Marika Pentikäinen

Co-Development of Work and Information Systems An Analysis of the Construction of the Activity-Driven ISD Methodology in 2001-2013

Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences

MARIKA PENTIKÄINEN

Co-Development of Work and Information Systems An Analysis of the Construction of the Activity-Driven ISD Methodology in 2001-2013

Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences No 153

Academic Dissertation To be presented by permission of the Faculty of Science and Forestry for public examination in Auditorium TTA in the Tietoteknia Building at the University of Eastern Finland, Kuopio, on September 27th, 2014 at 12 o’clock noon. School of Computing

Grano Oy Kuopio 2014 Editors: Profs. Pertti Pasanen, Pekka Kilpeläinen, Kai Peiponen, and Matti Vornanen Distribution: University of Eastern Finland Library / Sales of publications P.O.Box 107, FI-80101 Joensuu, Finland tel. +358-50-3058396 http//www.uef.fi/kirjasto ISBN: 978-952-61-1558-0 (print) ISSNL: 1798-5668 ISSN: 1798-5668 ISBN: 978-952-61-1559-7 (PDF) ISSN: 1798-5676

III

Author’s address

University of Eastern Finland School of Computing PO Box 1627 70211 KUOPIO FINLAND email: [email protected]

Supervisors

Docent Mikko Korpela, DTech. University of Eastern Finland School of Computing PO Box 1627 70211 KUOPIO FINLAND email: [email protected] University Lecturer Anne Eerola, PhD University of Eastern Finland School of Computing PO Box 1627 70211 KUOPIO FINLAND email:[email protected] Advisor Anja Mursu, Dr. Econ. Salivirta & Partners Oy Kauppakatu 28 40100 JYVÄSKYLÄ FINLAND email: [email protected]

Reviewers

Professor Steven Alter, PhD University of San Francisco School of Management 2130 Fulton Street CA 94117 San Francisco USA email: [email protected] Professor emeritus Jaakko Virkkunen, PhD University of Helsinki Institute of Behavioural Sciences Siltavuorenpenger 1-5 PO Box 9, 00140 University of Helsinki FINLAND email: [email protected]

IV

Opponent

Sampsa Hyysalo, PhD, Associate Professor, Co-Design Department of Design Aalto ARTS Hämeentie 135 C PO box 31000, 00076 Aalto University FINLAND email: [email protected]

V

ABSTRACT

Information systems should support the work activities which they are embedded in. The development of information systems should be addressed in work improvements, too. Actually, the development of work and information systems should happen as co-development, intertwined with each other. If we only improve the work activities, it is hard to find information systems or software that fit these activities. If we develop information systems only, work activities have to adapt to the constraints of the information systems. This is the motivation for this thesis. For understanding the work we need interaction and participation of the actors of the work. For modeling information systems requirements we need interaction and participation of the information systems developers. In order to support this we need guidelines for the co-development of information systems and work. We need a tool that is understandable to both the information systems developers and the actors of the work. During the years 2001-2013 the University of Kuopio (now University of Eastern Finland) had several information systems research projects in the context of healthcare. In the projects the research group participated in information systems development cases and strove to develop guidelines and a tool for the activitydriven development of information systems. In 2007 the outcome, the Activity-Driven Information System Development (ADISD) model and methodology was published. After that the methodology has been applied and adapted in various ways. This study tells the story of the development and use of the methodology as interpreted by the author. The process of the ADISD methodology development forms the case level of this study. On top of it there has been the meta level of a process of reflection on the case. The latter is presented as the analysis part of the thesis. To achieve understanding a lot of pictures were drawn. The outcomes of the analysis can be seen as general proposals to methodology developers. This thesis consists of two parts: the summary part and six research papers.

VI

Universal Decimal Classification: 004.414.2, 004.415, 331.103.32 INSPEC Thesaurus: information systems, systems analysis, modeling, health care, task analysis Yleinen suomalainen asiasanasto: systeemityö; tietojärjestelmät; suunnittelu; kehittäminen; käyttäjät; tarpeet; mallintaminen; toiminnan teoria; toimintatutkimus; terveydenhuolto

VII

Acknowledgements The starting point of this research was an idea of making the world (with computers) a better and easier place for human beings. This thesis is a story about attempts done towards the original and high idea. Significant step forward this research was taken when I was hired to my first project in 2002 by Anne Eerola, project leaded by Mikko Korpela. This thesis is the result of research carried out at the School of Computing, the University of Eastern Finland within three projects of health IS. I want to thank my supervisors Dr. Anne Eerola, Dr. Anja Mursu and Dr. Mikko Korpela for their kind support, intelligent advice, and all the attention they have paid to my work during the research. I also want to thank my reviewers Professor Steven Alter and Emeritus Professor Jaakko Virkkunen for their valuable and useful comments, and Associate Professor Sampsa Hyysalo for the promise to be the opponent in the public defense. The research was shaped up by the projects: PlugIT, ZipIT, and MyWellbeing funded by the Finnish Agency of Technology and Innovation (TEKES), the Finnish Work Environment Fund, and several Finnish companies and healthcare organizations. In addition, this research was supported also by Kuopion yliopistosäätiö, Ulla Tuomisen säätiö, and Itä-Suomen korkean teknologian säätiö with scholarships. Supported also by the EU, European Social Fund project “YLÄVÄT” through the Centre of Economic Development, Transport and the Environment in Northern Savonia. I want to thank all of the above mentioned organizations for financial support. I also want to thank all the great people in research networks in the University of Eastern Finland, the Savonia University of Applied Sciences, the University of Tampere, and the Aalto University in Finland. Special thanks go to my research groups: Anja, Anneli, Ansku, Heidi, Irmeli, Jari, Juha M, Juha R, Juho, Kristiina, Liisa, Marilla, Maritta, Merja, Mika, Pauliina, Pertti,

VIII

Päivi, Ritva, Susanna, Tomi, and Tuula. I would also like to thank the companies and health care organizations who participated in. I want to thank all my colleagues at the School of Computing; Special thanks go to Seppo, Matti, Markku, Paavo, Jukka, Merja L, Merja P, Leila, Piia, Paula L., Erkki, and Pekka K., who have been very supportive during the years. The warmest thanks I owe to Tanja Toroi and Maija Marttila-Kontio for their supporting and encouraging discussions with me. They have been as role models for me. I want to thank my co-authors Anne, Anja, Heidi, Irmeli, Juha, Kaija and Mikko, without whom it would have been impossible to do this work. Heidi Häkkinen opened her overall intelligence and special knowledge about health care to me. Juha Mykkänen gave me endless and intelligent advice about all the technological and methodological issues, or whatever topics came to my mind to be discussed. My dearest thanks I want to owe to Irmeli Luukkonen, my spiritual maid of honor, in research and in real life. She has been the most capable one to give me practical advice. I am grateful to my other sisters in research, too: Susanna Martikainen, Tuija Tiihonen, and Vilma Vainikainen have given wholehearted support and always been excited about my research ideas. With Daisy-sisters, I feel like a royal one. My personal thanks go to my friends and relatives Annukka, Anu, Sinikka, Sisko, Suvi and others for encouragement and specially for bringing joy and faith in to my life. My deepest thanks belong to my family. My mom Sirkka and dad Seppo have believed in me and have done everything to help me in my attempts. Finally, I thank our children Viltsu and Jimi for their love and patience. My very special thanks go to my beloved husband Ismo and his wise and loving heart. Thank you for being present in my life. Kuopio, 5 September 2014

Marika Pentikäinen

IX

X

XI

LIST OF ORIGINAL PUBLICATIONS

This thesis is based on data presented in the following articles, referred to by the Roman numerals I-VI. The publications have been included at the end of the printed version of this thesis with the permission of their copyright holders. I.

Luukkonen I, Toivanen M, Mursu A, Saranto K, Korpela M. Researching an Activity-Driven Approach to Information Systems Development. In Cruz-Cunha MM, Miranda IM, and Gonçalves P. (eds.) Handbook of Research on ICTs and Management Systems for Improving Efficiency in Healthcare and Social Care. vol 1, p. 431-450. Hersey (PA): IGI Global, 2013.

II.

Toivanen M, Eerola A, Korpela M. From information systems requirements to software components - home care case. In: Laukkanen S. and Sarpola S. eds. Electronic Proceedings of the 26th Information Systems Research Seminar in Scandinavia (Proceedings of the 26th IRIS), Haikko Manor, Finland, August 912, 2003, p. [14]. 2003.

III.

Toivanen M, Häkkinen H, Eerola A, Korpela M, Mursu A. Gathering, structuring and describing information needs in home care: a method for requirements exploration in a "gray area". In: Fieschi M, Coiera E, Li Y-C J, eds. MEDINFO 2004. Proceedings of the 11th World Congress on Medical Informatics: Part 2, San Francisco, September 7-11, 2004, p. 1398-1402. Amsterdam: IOS Press, 2004.

IV.

Toivanen M, Luukkonen I, Mykkänen J. Three-level analysis for shared understanding of information systems development. In: Breu R, ed. Proceedings of the IASTED International Conference on Software Engineering - SE 2009, Innsbruck, Austria, February 17-19, 2009, p. 43-48. Anaheim: ACTA Press, 2009.

V.

Toivanen M, Mykkänen J, Korpela M. Activity-Driven Information Analysis - Designing Personal Ubiquitous Health

XII

and Wellbeing Systems. In: Proceedings of International Workshop on Ubiquitous Healthcare and Supporting Services (UBI-HEALTH 2010), 31st May-2th June 2010, Shanghai, China. Science + technology (Aalto University publications series) 6/2011.

VI.

Toivanen M, Korpela M. How to Co-Develop Work Activities and Information Systems - Lessons Learnt from Developing the ADISD Approach (submitted)

XIII

AUTHOR’S CONTRIBUTION TO THE JOINT PAPERS

I made a significant contribution to all the papers. I was the corresponding author for Papers II-VI. (NB: I appear in all these papers by my maiden name Toivanen.) Paper I. The author contributed mainly to the construction of the Activity-Driven ISD methodology and pilot cases in the PlugIT and the ZipIT projects. The constructive analysis was conducted and the paper written jointly with Irmeli Luukkonen. Anja Mursu mainly contributed to the section concerning business applications of the AD approach, Kaija Saranto and Irmeli Luukkonen contributed to the section concerning early teaching experiences, and Mikko Korpela commented on the paper generally. The paper is part of the PhD theses of Irmeli Luukkonen and Marika Toivanen. Paper II. The story and the idea of the paper were contributed by the author. Anne Eerola contributed as a supervisor and contributed with the selection of the theoretical background. Mikko Korpela contributed mainly with Activity Theory. The paper was presented by the author at the IRIS26 conference. Paper III. The method explained with this paper was the author’s idea. The author conducted the real interviews and analysis explained within the paper. Heidi Häkkinen mainly contributed to the healthcare specific terminology, the section concerning healthcare practices in Finland, and the English. Heidi Häkkinen analyzed the method with the author. Anne Eerola commented on the paper as a supervisor and gave insights into the outlines. Mikko Korpela commented on the paper and layout. Mikko Korpela and Anja Mursu contributed the Activity Theory and ActAD framework as a background to the method. The paper was presented by the author at the MEDINFO 2004 conference. Paper IV. The methodology presented in this paper is the result of the ZipIT project, which was planned and conducted jointly by the research group. The methodology was based on the ideas of the author and analysis with the research group. The paper was jointly written and analyzed with Irmeli Luukkonen. Irmeli Luukkonen contributed with the English terminology. Juha

XIV

Mykkänen commented on the paper from the software engineering point of view. The paper was presented by the author at the IASTED SE’09 conference. Paper V. The materials for this paper were based on the MyWellbeing project. The materials were analyzed jointly with the research group. The author had the main responsibility and contribution of the development of the Activity-Driven information analysis. Juha Mykkänen contributed to the paper as an expert on architectures. Mikko Korpela commented on the paper as a supervisor. The paper was presented by the author at the UbiHealth’10 conference. Paper VI. This paper was based on a questionnaire study which was planned, carried out, analyzed, and documented by the author. Mikko Korpela contributed as a supervisor and expert on the background of information systems methods, and by copyediting the final version.

XV

LIST OF ABBREVIATIONS

ActAD AD ADISD AT BPR CASE CSCW DWR IS ISD RE SE SQ UML

Activity Analysis and Development (framework) Activity-Driven Activity-Driven Information System Development Activity Theory Business Process Re-engineering Computer-Aided System/Software Engineering Computer Supported Cooperative Work Developmental Work Research Information System Information System Development Requirements Engineering Software Engineering Sub-question Unified Modeling Language

XVI

XVII

CONTENTS

1 Introduction .................................................................................... 1 1.1 Motivation and Justification ....................................................... 1 1.2 Objectives ...................................................................................... 3 1.3 Definitions of Core Concepts and Conventions ...................... 4 1.3.1 Socio-technical ............................................................................. 4 1.3.2 Work, information system, software ............................................ 5 1.3.3 Activity, work activity, Activity Theory, ActAD ....................... 6 1.3.4 Development; work, IS, or software development; and codevelopment ............................................................................................. 7 1.3.5 Approach, methodology, method, technique, tool, model, framework ................................................................................................ 9 1.3.6 Activity-Driven approach, ADISD model and methodology .... 10 1.3.7 Action Research ......................................................................... 12 1.3.8 I, meta level, the (ADISD) group, case level, we ....................... 12 1.4 Research Questions and Expected Outcomes ........................ 13 1.5 Structure of the Thesis............................................................... 15 2 Research Methodology ............................................................... 17 2.1 Research Approach.................................................................... 17 2.2 Action Research ......................................................................... 18 2.3 Research Process and Design ................................................... 21 3 Theoretical Background ............................................................. 27 3.1 Development of Models and Methodologies ........................ 27 3.1.1 Struggling with the Terms ........................................................ 28 3.1.2 Justification for Methods and Methodologies ............................ 31 3.1.3 Characteristics of a Proper Methodology................................... 32 3.2 Activity Theory .......................................................................... 34 3.2.1 Classics ...................................................................................... 34 3.2.2 Engeström’s model of an activity system .................................. 36 3.2.3 ActAD framework and integrated levels of analysis ................. 37 3.3 Work and Work Development................................................. 41 3.3.1 Various approaches on work ...................................................... 41 3.3.2 The Change Laboratory ............................................................. 42 3.4 Software Engineering ................................................................ 44

XVIII

3.4.1 Requirements Engineering ........................................................ 46 3.4.2 Achieving Usable Software Modules......................................... 48 3.5 IS and ISD ................................................................................... 49 3.5.1 ISD Methodologies .................................................................... 50 3.5.2 The Work System Method ......................................................... 51 3.6 Summary ..................................................................................... 54 4 The Case: Developing the ADISD Methodology .................. 55 4.1 Healthcare and Health Information Systems as a Context .. 55 4.2 Projects as Sites of the Development of the Methodology... 56 4.3 PlugIT project ............................................................................. 58 4.3.1 Pilot Cases ................................................................................. 59 4.3.2 Outcomes ................................................................................... 62 4.3.3 Applications............................................................................... 64 4.4 ZipIT project ............................................................................... 67 4.4.1 Rationale and Realization .......................................................... 67 4.4.2 Pilot Cases ................................................................................. 69 4.4.3 Outcomes ................................................................................... 71 4.4.4 Applications............................................................................... 75 4.5 MyWellbeing Project ................................................................. 75 4.5.1 Rationale and Realization .......................................................... 76 4.5.2 Outcomes and Applications ...................................................... 76 4.6 Other Projects and Applications around Developing the Methodology ....................................................................................... 78 5 Summary of Papers ..................................................................... 81 5.1 Relations of the Research Papers and Projects ...................... 81 5.2 Relationships and Roles of the Research Papers ................... 83 5.3 Researching an Activity-Driven Approach to Information Systems Development ........................................................................ 84 5.4 From Information Systems Requirements to Software Components - Home Care Case ....................................................... 85 5.5 Gathering, Structuring, and Describing Information Needs in Home Care: A Method for Requirements Exploration in a "Gray Area".......................................................................................... 87 5.6 Three-level Analysis for Shared Understanding of Information Systems Development ................................................. 87

XIX

5.7 Activity-Driven Information Analysis – Designing Personal Ubiquitous Health and Wellbeing Systems .................................... 88 5.8 How to co-develop Work Activities and Information Systems - Lessons Learnt from Developing the ADISD Approach 89 5.9 About the Papers ....................................................................... 90 6 Analyzing the Development of the Methodology as an Activity ................................................................................................ 93 6.1 The Activity Network of the Development of the ADISD methodology ....................................................................................... 94 6.2 How can the Development Process be Analyzed as an Activity? ............................................................................................... 95 6.3 Analyzing the Phases of the development of ADISD........... 97 6.3.1 Phase 0: Producing the ActAD framework ............................... 98 6.3.2 Phase 1: Towards Work-centered ISD ..................................... 100 6.3.3 Phase 2: Development of the ADISD Methodology ................ 102 6.3.4 Phase 3: Applications of the ADISD Methodology ................. 104 6.3.5 Phase 4: Future Development of the Methodology .................. 105 6.4 Proposals for the Development of the Models and Methodologies ................................................................................... 108 7 Discussion and Conclusion ..................................................... 111 7.1 Summary of the Findings ....................................................... 112 7.2 Contributions of the Thesis .................................................... 117 7.2.1 Research Methodology ............................................................. 117 7.2.2 Theoretical Contribution at Meta Level .................................. 118 7.2.3 Practical Outcomes at Case Level............................................ 119 7.3 Strengths and Limitations of the Study ................................ 120 7.4 Future Work ............................................................................. 121 References ......................................................................................... 123 APPENDIX

Original publications

XX

LIST OF FIGURES

Figure 1. Action research is communication between research and practice to acquire knowledge. ...................................................................... 20 Figure 2. Initial plan of the research phases ................................................ 21 Figure 3. Two-level research process ............................................................ 22 Figure 4. Simplified figure of the continuous process of this cumulative study .................................................................................................................. 23 Figure 5. Relations of core concepts .............................................................. 30 Figure 6. The fundamentals of Vygotsky’s studies lie in the X-factor (modified from Vygotsky 1978, p. 40). ......................................................... 35 Figure 7. A triadic representation of actions, according to Engeström (1999, p. 30) ....................................................................................................... 36 Figure 8. A complex model of an activity system with an example, according to Engeström (1999, p. 31). ISCRAT is the international congress on activity theory, later renamed to ISCAR. ............................... 37 Figure 9. From triangular to oval presentation of work activity (Korpela & Mursu, 2003). Compare with Figures 8 and 10 for details..................... 38 Figure 10. The structure and relations of work activity as a systemic entity (Korpela et al., 2002)............................................................................. 40 Figure 11. Two times four integrative levels of analysis (Korpela et al., 2001)................................................................................................................... 41 Figure 12. The stages in software development confidence (Rook, 1986) 45 Figure 13. The shared understanding as the basis for the requirements (Luukkonen, 2012, p. 15; modified from Toivanen et al., 2007) ................ 49 Figure 14. The Work System framework (Alter, 2013, p. 78) .................... 52 Figure 15. The Work System Life Cycle Model for supporting IT innovations (bullets under headlines of the boxes) (Alter, 2004a, p. 6) ... 53 Figure 16. Research projects around the process. ....................................... 56 Figure 17. The relationships of the research projects to the process. (combination of Figures 3 and 16) ................................................................. 57 Figure 18. Organization of integration work in PlugIT project (Mykkänen et al., 2004, adapted from CORBA). .............................................................. 58 Figure 19. Picture from a workshop of the research group. ...................... 62 Figure 20. The early version of the model used in the home care pilot case of the PlugIT project ............................................................................... 63

XXI

Figure 21. Project’s example of the method’s steps of focusing (Toivanen, 2004) Activity network -> Activity -> Actions of process. (Essential here is the shapes of the diagrams and the details are not relevant) .................... 64 Figure 22. The project’s example of the method’s steps of focusing (Toivanen, 2004a) Actions of process -> Use cases. (Essential here is the shapes of the diagrams and the details are not relevant) .......................... 65 Figure 23. The project’s example of the method’s steps for focusing (Toivanen, 2004a) Use cases & sequences. (Essential here is the shapes of the diagrams and the details are not relevant) ............................................ 65 Figure 24. The modeling chain supporting traceability (Mykkänen et al., 2006)................................................................................................................... 66 Figure 25. Starting point and ideology of the ZipIT project: work and IS development in parallel (zipped up together). ........................................... 68 Figure 26. Elements around methodology development activity in the ZipIT project. Zippers demonstrate the need for balancing (Paper VI)... 69 Figure 27. One of the versions of the model under construction in the year 2005. Essential here is the shape of the model and the details are not relevant. ............................................................................................................ 72 Figure 28. A simple version of the ADISD model....................................... 73 Figure 29. The ADISD model in a table form (Paper IV) ........................... 74 Figure 30. The present state of a network of activities and information landscape in the pilot case: referral and referral feedback (adapted from Luukkonen et al., 2007). .................................................................................. 75 Figure 31. The Activity-Driven ISD model adapted to person-centered information analysis (adapted from Paper V). ............................................ 77 Figure 32. Information analysis of elements around the Coper (adapted from Tuomainen et al., 2010b). ...................................................................... 78 Figure 33. Research papers and their relations to the projects .................. 82 Figure 34. Research papers and their roles (capital letters) in this study 84 Figure 35. Essential phases of the development process with projects ... 94 Figure 36. Analyzing the development of the ADISD methodology as an activity ............................................................................................................... 96 Figure 37. Activity network of the essential phases of development ..... 98 Figure 38. The activity network of Phase 0 ................................................. 99 Figure 39. The activity network around the PlugIT project ................... 101 Figure 40. The activity network around the ZipIT project ..................... 103 Figure 41. Phase 3 as an activity ................................................................. 105

XXII

Figure 42. Analyzing the development of ADISD with the ADISD model. .......................................................................................................................... 107 Figure 43. Two-level research process (Figure 3) with the contributions of the essential chapters. ............................................................................... 117

XXIII

XXIV

1 Introduction “You can’t tell where you are going, unless you know where you have been” - Unknown Once upon a time there was a software implementation in an organization. The software was great and included all the new technology. The software was everything that had been set in the contract by the managers. The software was integrated with the legacy system of the organization as promised and everything worked, except the people. The end users were not working any more. This story is not real, even though it could be. Stories can be used to explain and understand things such as processes (Alexander & Maiden, 2004). Those stories can be scenarios, user stories, or personas. They can be made-up narratives or real stories. The story is always an instance. If we observe, analyze, and theorize about the instances we should be able to see the general-level process above the instances. This study is a real story about a process of striving to solve the problem of the above story, in which people were not working. This thesis is a story of developing a methodology for the co-development of information systems and work. This story has been written in order to produce new understanding about the methodology development process by making this whole process of ours visible.

1.1

MOTIVATION AND JUSTIFICATION

If we improve only the work activities, it is hard to find software to fit in with these activities. If we develop only software, work activities have to adapt to the changes caused by the software (Minkkinen & Eerola, 2007).

Dissertations in Forestry and Natural Sciences No 153

1

Pentikäinen M.: Co-Development of Work and Information Systems

There was already knowledge about the causes of information systems (IS) failures in the 1970s. Three reasons for such failures are named technical issues, organizational issues, and issues related to project management. Understanding the user needs and the use of information was also seen as an issue that was lacking in the information systems design process (Lucas, 1975). This thesis brings up issues concerning better information systems that are already known. But it also provides new insights into means and tools, and above all, it reports what a group of researcher-developers in Finland has done to improve Information Systems Development (ISD). The group’s solution to the problem of unfit IS was a new way to develop IS, a new methodology. This study is about the process of developing that methodology. Who could be interested in this study? While doing the research I was thinking of IS developers, IS users, the clients of IS users, management, researchers, and students interested in IS development. IS developers could apply the methodology that was developed or parts of it in order to gain an understanding of activities around software systems. IS users taking part in IS development might get help from the methodology. The ideal case might be that developers and users could apply the methodology together. Management, both of the developers and of the users, could use the methodology to see the big picture around development. The clients of IS users might get benefits in the form of better services from the users. For researchers the whole methodology development process is made transparent, which makes it possible to continue to develop the methodology or learn about the development of the methodology. Students get knowledge about IS development. Development is always a process. It is not strict but many activities influence the outcome of the process. Requirements for the outcome artifact might exist. Even milestones might be guided, for example, by a model or a method. But unfinished descriptions created during the process are not usually documented for final versions of products. Only the refined outcomes are documented instead. This leads to the 2

Dissertations in Forestry and Natural Sciences No 153

Introduction

disappearance of traceability and makes it hard to learn lessons from the processes. The process itself has a value, that of producing good products. We usually take it for granted that the quality of the process leads to the quality of the outcome. But studying the processes has its own value, too. Real-life processes hardly ever proceed in the same way as they do in textbooks and guides. We still need guidelines and generalizations. Novices need stricter and more detailed guidance than experts. Experts have knowledge and experience of how to apply guidelines. Within this study I analyze two processes: first, the information systems development process, and second and more important, the ISD methodology development process. These analyses are performed in order to produce better information systems development methods and methodologies, which lead to better information systems, which lead to meaningful work and improvements in the quality of the services provided by the work activities. Other socio-technical methods strive towards this same goal, but there is a lack of the practical use of such methods (Baxter & Sommerville, 2011). The methodology developed by research group has been applied in real practical pilot cases.

1.2

OBJECTIVES

The co-development of work and IS was the starting point of this study. To support this goal, a methodology was constructed by a research group in the University of Kuopio (since 2010 the University of Eastern Finland). This methodology was based on Activity Theory and named the Activity-Driven ISD (ADISD) methodology. The construction of the methodology mostly happened in the years 2001-2013 in several research and development projects. On top of this starting point, I gradually came up with a higher level goal, too. Through documenting and analyzing the actual process, and not only its refined outcome, we can learn about this process. The goal is to analyze the process. The lesson to be Dissertations in Forestry and Natural Sciences No 153

3

Pentikäinen M.: Co-Development of Work and Information Systems

learned here is how a methodology was developed for supporting the co-development of work and information systems. This thesis is not to pinpoint the weaknesses or strengths of the ADISD methodology. This thesis is not to evaluate the methods or frameworks used in constructing the methodology. But the objective of this thesis is to analyze the construction of the Activity-Driven ISD methodology in 2001-2013. That construction could be seen as an attempt to support the co-development of work and information systems. This historical analysis is done in order to envisage how this approach could be further developed in the future. The thesis also makes some proposals in support of a comprehensive development plan that requires more collaboration among researchers from different fields. The historical analysis of the case in this thesis can be used as one of the tools by those interested in further development of sociotechnical methodologies.

1.3

DEFINITIONS OF CORE CONCEPTS AND CONVENTIONS

This thesis is multidisciplinary by nature. The same terms are used in different meanings in different disciplines and schools of thought. To orientate readers from different research traditions, core concepts used in this thesis are briefly defined here upfront. References are made to the individual papers and sections of this summary where the terms and concepts are discussed more thoroughly. I also try to point out what the terms do not mean in this thesis. 1.3.1

Socio-technical The term socio-technical is used in this thesis as a generic term, not to refer to the Socio-Technical Systems theory or any other specific school of thought. It simply refers to real-life phenomena that have both social and technical aspects, and to theories and methods dealing with such phenomena. This is the common way 4

Dissertations in Forestry and Natural Sciences No 153

Introduction

of using the term in Information Systems research currently – “Socio-technical systems design methods are an approach to design that consider human, social, and organizational factors, as well as technical factors in the design of organizational systems” (Baxter & Sommerville, 2011). Particularly, I do not think that there are social and technical subsystems that are separate or parallel, like the old SocioTechnical Systems school of thought maintained. Rather, social and technical aspects are intertwined inseparable aspects of inherently socio-technical entities like activities and information systems. 1.3.2

Work, information system, software Work is activities, actions and tasks done by people for a purpose, motivated by the goals of a domain organization (e.g. business strategy, taking care of health, or being a good hairdresser). To do their work employees necessarily need an information system, which is a collection of all the elements which “offer the right information at the right time in the right place so that available information meets the needs of” the employees (Paper II). For example a hairdresser needs his or her notebook of appointments. He or she probably also has some information about each customer (name, phone number, hair treatments, etc). This is an information system even though there is no automated system. Information systems can have automated and manual elements. To develop a hairdresser’s information systems we can train him or her a new way of using the notebook, buy a better notebook, or introduce software that covers part of the information system. In this case some web-based appointment system could be the one. Information system as a real-life phenomenon is defined in Paper I: “The processes of managing (creating, using, storing, exchanging, etc.) information in an organizational setting (in work activities) for a purpose. An information system consists of people (actors), information (contents) and technology (means), linked together by a process directed towards a purpose. Information systems are mediated by various information Dissertations in Forestry and Natural Sciences No 153

5

Pentikäinen M.: Co-Development of Work and Information Systems

technologies. An information system is not always a ‘system’ (a systemic entity) but can consist of bits and pieces of processes and technology subsumed in a systemic work activity. Information system is a socio-technical entity that has both human and technological aspects.” Unfortunately the term information system has to be used for historical reasons although the management of information in a given organizational setting or activity does not always form a proper system. The relations between the terms work (system), information system and software are described in Figure 13, Sections 3.3 and 3.5, as well as Paper I. It is especially important to realize that an information system and a software system (product, package) are very different things. 1.3.3

Activity, work activity, Activity Theory, ActAD The activities concerned with in this thesis are mostly work activities, e.g., taking care of a patient or developing an information system. Human activities more generally are dealt with in Paper V. The term activity is not used here in the same meaning as in the Activity diagrams of the software analysis language UML (Unified Modeling Language). Work activity is defined in Paper I, definitions: “The systemic entity of purposeful, cooperative human action, where several actors work in an organized way upon a shared object of work to transform it into an intended outcome, by using different kinds of means of work and means of cooperation and coordination. The intended outcome forms the purpose (motive) of the activity. Information entities, information tools, and information systems are used within work activities alongside with other means of work and means of cooperation and coordination.” In this thesis we are interested in activities which are going to be developed and especially such activities which are going to be developed with information systems development. “Activity Theory: The theory of socio-cultural, mediated human activity, evolved from late 1800s, currently a multi-faceted tradition with several sub-traditions” (Paper I, Definitions). I do 6

Dissertations in Forestry and Natural Sciences No 153

Introduction

not pretend to know thoroughly or be an orthodox follower of various theorists of human activity like Hegel, Marx, Vygotsky, Scribner, Leont’ev, Engeström or Bedny for instance. In this thesis the main point is to see work as networks of systemic, mediated, dynamic entities of collective action called activities. Activity Analysis and Development (ActAD) is a framework, based on Activity Theory and Leont’ev and Engeström in particular, with which one can analyze an activity as a systemic entity. The framework was introduced by Korpela (1994) and further developed by Korpela et al. (2000; 2002; 2004). ActAD appears in this thesis in two roles: firstly, as an analytical framework used in the ADISD methodology by the ADISD group, and secondly, as the analytical framework that I use in analyzing the methodology development activity. The application of ActAD in the PlugIT project is explained in Paper I. Papers II & III and Section 3.2.3 introduce the framework more. 1.3.4

Development; work, IS, or software development; and codevelopment Development is seen in this thesis as an activity of changing things (e.g. work practice, information flow, or software). It can be desirable development (improvement) or maldevelopment from a specific stakeholder viewpoint, even both at the same time from different viewpoints, but my interest is in development that aims at improvement. Development can take place within a project in some certain period of time. The change period can be short or long. Some development can be continuous. Some development can be more planned, some even unconscious. I am interested in conscious changes. Some development happens in a way we planned but some is unplanned. We try to manage planned changes and understand the unplanned changes. To manage development we have to understand the current (its history) and goal situations. In order to support this we make things visible by documenting and modeling. My definition of development in this thesis does not refer to the endogenous development (evolution) of an activity for Dissertations in Forestry and Natural Sciences No 153

7

Pentikäinen M.: Co-Development of Work and Information Systems

example. It does not refer to socio-economic or human development, either, as in Development Research or educational theory for instance. Any purposefully made improvement in work practices is understood as work development (Section 3.3) in the context of this thesis. Development can be small changes, e.g. moving a work station or the procurement of new tools, or a longer period of developing and introducing new practices (e.g., quality assurance, business process modeling, or learning organization). Work development can start because of an observed or emerging disharmony in work practice, which is caused by, e.g., new goals, new rules, or new software becoming or being implemented. My definition thus does not mean extensive and heavy work development projects only. In this thesis the focus is on such “size” of work development that is possible in conjunction to an information systems development effort. Information systems development (ISD) (Section 3.5) is defined at the end of Paper I as “the process (activity) whereby a work activity or a larger organizational setting is facilitated by introducing a new socio-technical information system or modifying or expanding an existing one. […] Depending on the viewpoint, it can be seen as a software engineering process of a software producer, an application acquisition process of a software user, or a work development process” (spelling error in the original corrected). Software development (Section 3.4, Paper II) is a process or a project managed (usually) by software developers (hopefully, in interaction with end users) in order to specify, implement, test, integrate, or deploy and maintain software products. The discipline studying software development, among other things, is called Software Engineering. Software development can start from requirements engineering in order to produce a new software product, or to produce just a new part into an existing software product, or to integrate already existing parts (e.g. software components). Software Engineering always deals with automated data processing.

8

Dissertations in Forestry and Natural Sciences No 153

Introduction

By co-development I mean such improving the current situation that is done together, jointly, in a shared, cooperative, intertwined way, or at least taking each other into account as much as possible. In earlier phases and papers (e.g., Paper V) terms like ‘parallel development’ were used for the same purpose. By the term co-development I also mean development that takes into account both sides: work and IS. Development which leads to changes in information systems should always take into account what are the changes needed in work activities, too. On the other hand, when work practices change, information systems should be developed, too. Information systems are always embedded in some activities, but by developing the information systems only we cannot address all the possible disharmonies in activities. (Figures 13 and 25) 1.3.5

Approach, methodology, method, technique, tool, model, framework Being a study in the scientific field of Information Systems, this thesis uses this cluster of terms in the way they are most commonly used in IS research (section 3.1.1-3.1.2). At the core is the term method which means a planned way of doing something. A technique is a lower level way of performing an action in a prescribed way. Methodology is a coherent set of methods guided by a shared set of principles, a “philosophy”. An approach is a higher level term referring to a shared principle loosely grouping different methodologies together. For instance, Participatory Design is an approach within which Cooperative Design was a documented methodology which utilized methods like future workshops and techniques like wall charts. The “philosophy” binding together an approach or a methodology is often a model or a theoretical framework. The two latter terms are here used almost interchangeably for a theoretically justified set of concepts and relations between them which tries to model a real-life phenomenon. In a simplified way, “model/framework + methods = methodology”.

Dissertations in Forestry and Natural Sciences No 153

9

Pentikäinen M.: Co-Development of Work and Information Systems

These terms are used differently in different disciplines and schools of thought. Particularly method is often used in the way I defined methodology (e.g., Work System Method, Change Laboratory Method). Research methodology here means the coherent set of research methods selected to address the research questions of a study (sections 2.1, 2.3).

1.3.6

Activity-Driven approach, ADISD model and methodology Activity-driven (AD) approach refers to various research, consultancy, teaching etc. efforts that share the principle of “activity-drivenness” (Paper VI). The essential characteristics of AD are defined in Paper I: “We use the following statements to define an Activity-Driven approach to ISD. • The theoretical basis of the approach draws from Activity Theory, the Activity Analysis and Development framework, a socio-technical view of information systems development, and user participation. • The approach is developed for the early phases of ISD (planning, data gathering, analysis, describing, and validation) with an emphasis on intertwined work and information systems development. • Information systems development starts by studying work activities as systemic entities. • Technology, including computer-based technology, is seen as a tool to facilitate work, embedded in the work system. • Both collective and individual aspects of work are taken into account. • Work systems are studied in their organizational context.

10

Dissertations in Forestry and Natural Sciences No 153

Introduction

•

Participative and cooperative methods are used to incorporate representatives of different professional viewpoints within the work activities under study. • Understandable tools and visualizations are used to enable domain workers and other stakeholders to articulate their views of subject matter. The following three characteristics of an AD approach have been established within previous research projects: 1. utilization of the Activity Analysis and Development framework as a tool for analyzing human work activity as a systemic entity; 2. utilization of the Activity-Driven Information Systems Development methodology for guiding the gathering and analysis of the information, and for understanding the as-is situation and capturing the user needs for a to-be situation; 3. utilization of participatory and cooperative methods and tools with specific features driven by Activity Theory and the AD frameworks.” Within the AD approach, activity-driven ISD means “ISD that is driven by the requirements of the (work) activities” (Paper I, Definitions). ADISD methodology consists of the 3x3 model (Figures 28 and 29, explained in Paper IV) and a coherent but expandable and adjustable set of methods and tools (discussed in Paper I). The methods and techniques are mostly not unique to ADISD. The term ADISD model and methodology is sometimes used in this thesis to emphasize the existence of the model that binds the methods together. In earlier papers, ADISD model was used to refer to the entire methodology. This was mainly based on the Finnish word toimintamalli (literally ‘model of action’, ‘operating model’, ‘course of action’) used to describe the outcome of the ZipIT project (Toivanen et al., 2007). ActAD is used in ADISD as a theoretical framework with which one can analyze an activity as a systemic entity as well as networks of activities. The AD approach, ADISD methodology and ActAD framework are neither the main research objects nor among the main research outcomes of this study. They are Dissertations in Forestry and Natural Sciences No 153

11

Pentikäinen M.: Co-Development of Work and Information Systems

elements in the history of a methodology development that is the research object of this thesis. As usual in a case study, comments on specific elements of the case are produced as side products of the thesis. 1.3.7

Action Research Action research is used in this thesis as a generic term for approaches in which “the researcher enters a real-world situation and aims both to improve it and to acquire knowledge” (Checkland & Holwell, 1998)”, in the way it is commonly used in IS research (section 2.2; Baskerville, 1999). What is essential in my view is that researchers and practitioners collaborate, there are moments of both action and reflection, and the outcomes of reflection are fed back to the action, to the practitioners. We can speak about experiments and interventions. I do not commit myself to any branded orthodoxy within action research (for examples, see Baskerville & Myers, 2004), for instance the social engineering undercurrent of Lewin’s (1946).

1.3.8

I, meta level, the (ADISD) group, case level, we The first person singular subject I (my, etc.) is used in this thesis to refer to what the researcher has done for this doctoral thesis at the meta level of analysis on the methodology development experience. I am solely responsible for those actions and thoughts, and this thesis should be assessed on the basis of those parts. The term the ADISD group or simply the group refers to what the researcher-developers at the case level have done within the ADISD methodology development. I was indeed a member of the group for most of the time, at times in a leading position, so personally I do share the responsibility of those actions and thoughts as well. However, in terms of this thesis, those parts describe the research material. The first person plural subject we (our, etc.) is used to refer to you, my reader, and me and the scientific community – what we could or should think or do.

12

Dissertations in Forestry and Natural Sciences No 153

Introduction

1.4

RESEARCH QUESTIONS AND EXPECTED OUTCOMES

Taking end users into account during the information system development process is not enough. The final system should take into account the information needs of users during their activities. Information systems are there for supporting work activities (such as teaching or patient care) or work actions (such as writing notes or writing a recipe). Developing the information system that supports work means that the work has to be under analysis, too. Moreover, the intertwined development of IS and work should be the goal. This would make it possible to see where information system development is the solution and where changes in work practices are needed. The research problem and starting point for this thesis is that there was no proper methodology for supporting the co-development of work and information systems. There are models and methods for work development (Engeström et al., 1996; Virkkunen & Newnham, 2013; Alter, 2013). There are plenty of methods for IS development and software engineering (Avison & Fitzgerald, 2006; Sommerville, 2011; Alter, 2013). Some of the methods and models are in use but not always in the way that was initially intended. The issue of theory versus practice also arises. Some of the methodologies might be too theoretical to use in hectic practical development projects. If the development of a methodology is too far from real practices, can the methodology that is developed be practical to use? I think that methodologies should at least be tested in practice. Why should we design and develop another new methodology again? It was the “ideology”, the notion that there should be a methodology which enables IS and work to be co-developed, that pushed the development of the ADISD methodology. The methodology should enable both information and action to be analyzed and designed in tandem as much as possible. The ADISD group had the opportunity to test this idea in practice. Developing a good methodology for information system development is crystallized by Lucas (1975, p. 112) in the third of 12 design considerations for successful systems: “Let the user Dissertations in Forestry and Natural Sciences No 153

13

Pentikäinen M.: Co-Development of Work and Information Systems

design the systems if possible.” The user has to be in focus. “If possible” here means that users cannot design software by themselves but there has to be communication and shared understanding between users and developers as much as possible. Thinking about the desired methodology, the ADISD group had to find out the current state of the art in development. If there could be a methodology for the group’s purposes, what should be the requirements for that methodology? What existing methods could be included? Because the group did develop a new methodology (by utilizing theories and reported experiences), it is valuable to report the experiences of this development process. When the group strived to solve the problem by developing a methodology, it learned to construct a methodology by reflecting and reporting on the analysis of the development at the same time. To address the research problem, I formulated the main Research Question as follows: RQ: How can methodologies for the co-development of work and information systems be developed? This is the main question and the contribution of this study. The question is addressed by a single-case research design; that is, I analyzed the case (ADISD development) from the viewpoint of the main question, to produce lessons that contribute to answering the question but of course cannot provide a “complete” answer. To address the main question I formulated four SubQuestions, which are presented in Table 1. Sub-questions are formulated in order to get material and understanding needed to answer the main question.

14

Dissertations in Forestry and Natural Sciences No 153

Introduction

Table 1: Sub-questions, expected outcomes #

Question

Expected Outcome

Focused on

SQ1

What is the current state of

Theoretical basis and

the art in both ISD and Work

background,

Development

motivation

Ch 3

(methodologies)? SQ2

How can the co-development

Requirements for the

of IS and work be supported?

proper methodology

Ch 3, 4

(importance of communication, holistic view) SQ3

How could Activity Theory be

Experiences of

Ch 3, 4,

utilized to provide a coherent

applying the

5

methodology for the co-

methodology that was

development of IS and work?

constructed. Does it fulfill the requirements?

SQ4

How did the development of

Analysis of the

the ADISD methodology take

development process.

place?

Lessons learned and

Ch 6

activity network of experts and professionals.

1.5

STRUCTURE OF THE THESIS

This thesis consists of six research papers and the summary part, which has seven chapters. Chapter 1 provides the motivation and research questions. The methodology part in Chapter 2 introduces how this research is carried out. Chapter 3 gives an insight into the theoretical background of the essential concepts of the thesis, such as Activity Theory and Information Systems. Chapter 4 shows how the methodology was developed. The Dissertations in Forestry and Natural Sciences No 153

15

Pentikäinen M.: Co-Development of Work and Information Systems

projects and their outcomes are reported here. Chapter 5 introduces and summarizes the research papers and their relationships and roles in this research. The contributions of this research lie especially in documenting and analyzing the development of the artifact (methodology) and the process around that. Instead of listing the strengths and weaknesses of the outcome, analysis of the process is performed in order to gain an understanding of the phenomenon in general, too. This analysis is provided in Chapter 6. The research process and findings are discussed and summarized in Chapter 7.

16

Dissertations in Forestry and Natural Sciences No 153

2 Research Methodology Shaw (1990) presents the cycle of software methodology development as a circle: new problem, ad hoc solutions, folklore, codification, methodologies and theories, and improved practices. Like Shaw, I think good methodologies are the result of the interaction between science and engineering or practice. This kind of interaction is typical of action research. Moreover, I am interested in this process of action research. We can gain an understanding of this process by means of the concepts of Design Science. This chapter is organized as follows: 2.1 Research approach introduces the discipline of Information Systems; 2.2. Action research presents this method, and 2.3 Research process and design shows the methods with which this research was designed and performed.

2.1

RESEARCH APPROACH

“Information Systems is a new discipline and many of us come from very different backgrounds—from physics, chemistry, mathematics, psychology, management, sociology, philosophy, and computer science.” (Gregor, 2006) In order to clarify the researcher’s background and aims the researcher commonly opens up his or her philosophical assumptions. Information Systems research has been influenced by “soft” social sciences (but also by “harder” technology-oriented software engineering). Since information systems are created and used by humans it is natural to use social science and theories to study the phenomena associated with the development of Information Systems. As Gregor (2006) asks, what are the boundaries of the discipline called IS? What discipline is useful for this study?

Dissertations in Forestry and Natural Sciences No 153

17

Pentikäinen M.: Co-Development of Work and Information Systems

In research into Information Systems (IS) there is no single theoretical perspective. There are several philosophical assumptions (Orlikowski & Baroudi, 1991). The focus of this study is not on philosophies, but Orlikowski’s and Baroudi’s (1991) classification into positivist, interpretivist, and critical gives me the opportunity to categorize myself as an interpretivist. Within this study this means that I understand that we cannot reach the object (“as given”) as it is without our own interpretation of it. Assumptions and outcomes are colored by the mind of the researcher. The empirical world cannot be reached and modeled as it exists. This is the case in most qualitative research. Qualitative research is anyhow guided by philosophical assumptions (Myers and Avison, 2002). This thesis is based on interpretive assumptions, which means space to understand, and to make observations and categorizations during the process. While dealing with the nature of Information System Theory, Gregor (2006) states that: “A characteristic that distinguishes IS from other fields is that it concerns the use of artifacts in humanmachine systems.” This definition gives enough space to say that this is Information System research. Moreover, having Design Science as quite a new side of the IS field is encouraging. This means that not only is the “use of artifacts” a matter of interest but artifacts themselves and designing them are too. “Design is both a process (set of activities) and a product (artifact) a verb and a noun (Walls et al., 1992). It describes the world as acted upon (processes) and the world as sensed (artifacts)” (Hevner et al., 2004). Since I study the process of the development of an artifact through an interpretive lens, I would say this study falls into the Design Science category.

2.2

ACTION RESEARCH

“In this process the researcher enters a real-world situation and aims both to improve it and to acquire knowledge” (Checkland & Holwell, 1998). I as a researcher have been involved in the process which I study. I have planned, acted, observed, and reflected on 18

Dissertations in Forestry and Natural Sciences No 153

Research Methodology

the process together with the research group. This research has the features of action research. It focuses more on the process than findings. The findings demonstrate an understanding of the phenomenon under research. The research problem, “There was no a proper methodology for supporting the co-development of work and information systems”, is such it could be approached by a literature review, conceptualizing, and empirical testing. Quantitative analysis was not possible; there were no numerical data to use as statistics. Instead I have performed qualitative analysis during action research. There has also been a literature review, conceptualizing, and empirical testing, but also the interpretation of all these. Action research has its roots in the social sciences but has been applied to the development of information systems for 30 years (Baskerville & Wood-Harper, 1996; Baskerville & Myers, 2004). Action researchers had noticed that human activity is systemic by nature and each researcher involved has an influence on this system. This notion linked action research to system theory. It has also been said that Checkland’s use of action research to develop his soft system methodology (1981) is a landmark in the use of action research within IS (Baskerville, 1999). Like Goldkuhl (2008), I think that action research is collaboration between research and local practice. Collaboration provides the researchers with empirical data and also contributes to the development of local practice. The research community then contributes to the scientific body of knowledge (Figure 1).

Dissertations in Forestry and Natural Sciences No 153

19

Pentikäinen M.: Co-Development of Work and Information Systems

Planning

Reflection

Action Empirical data

Local practice contribution

Observations

Construction Scientific body of Knowledge

Practice

Research

Figure 1. Action research is communication between research and practice to acquire knowledge. As action research this study is multifaceted. It is qualitative, interpretive, a dialogue between theories and practices, and not repeatable. This study has been performed under certain circumstances and in certain contexts. It cannot be assumed that the action can be repeated. We cannot be sure about contexts or their similarities. We cannot be sure about future contexts, either. Using action research only to document failures, without reflection, does not lead to changes, but interpretation and construction are valuable as means to deepen understanding and the lessons learned. Action research does not lead directly to generalization, even though, through analysis, some ideas about generalization can be presented. Struggling with collaborative research and the individual work of writing a thesis is typical of action research (Herr & Anderson, 2005). This has been very much the case in this study. I, as the 20

Dissertations in Forestry and Natural Sciences No 153

Research Methodology

writer of this thesis, have made my own individual interpretations but at the same time I have gained many insights from the research groups and discussions with them.

2.3

RESEARCH PROCESS AND DESIGN

The research of this study has been multiphased. When I started this study the goal was quite clear. But, as usual, when knowledge grows more things to take into account come with the process. My first simple plan was to develop and document a method. This method ought to have some good points from existing methods and would deal with work activity when developing information systems. The planned process simply had three phases: 1. constructing the method from the literature; 2. testing the method empirically with action research, and 3. documenting the method (Figure 2). Literature

Action research

Constructing a method

Empirical testing of the method

Phase 1

Phase 2

Document the result: the tested method

Phase 3

Figure 2. Initial plan of the research phases Quite soon the method to be developed expanded and the group started to call this artifact a model. Empirical testing was also continuing and the group gained more perspectives on the goal. When starting with the new case, the tools and skills were more developed than in the previous case, the group had experience, and the model (input) to test had new parts included. This cumulative experience process influenced the model in such a way that it has had many variations. I will observe the

Dissertations in Forestry and Natural Sciences No 153

21

Pentikäinen M.: Co-Development of Work and Information Systems

variations of the model and methodology but concentrate more on analyzing the process of producing those. Because I have been involved in three different projects during this study I see that there have been three different phases of testing but also, at the same time, there has been reflection involved. The real development projects and collaboration with the professionals make this empirical testing action research. If we look at the whole picture under study within this thesis we can see that developing the methodology has happened through action research processes. Those processes have had typical phases of action research (action, observation, reflection, and construction). The artifact constructed here is the methodology for our purposes (as for the co-development of work and IS). Developing the methodology is the case I reflect on (Figure 3). Reflection: Analyzing the development of the methodology as an activity Methodology development in general

The Case: Developing the methodology

Preliminary research

Action research

Action research

Action research

Future

Figure 3. Two-level research process As action research usually does, this research focuses on practical cases, which means that the role of the literature is to be in a dialogue with the cases. Literature is needed for understanding theories, for planning the use of theories, and for reflecting on the use of theories. To make it simple, I would say that action research here has been a circle of action, observation, reflection, and construction. The action research processes (Figure 3) of this study took place within several cases in different organizations within three projects. The group called them pilot cases. The pilot cases occurred in the context of healthcare and social services in 22

Dissertations in Forestry and Natural Sciences No 153

Research Methodology

Finland (for more see Chapter 4). While working with the pilot cases the group got valuable information about real problems, environments, and practices. In its research the group actually created and tested its theory with the cases. After all, we could see two-level action research here: first, action research performed by a research group in collaboration with practitioners from the healthcare sector and the software industry, and second, action research conducted by me within the collaboration of the research group which I was in. Figure 4 shows some activities that were done, the outputs of the process, and the cumulative nature of this research.

Analysis Report of Report R case of the Report the case of the case

Previous research

Action Action Action

R ReflecReflectio Reflectio n ntion

The Thesis

O ObservaObservat Observat ions ions tions

c Construc construct construct ionion -tion

The Model

Planning

Figure 4. Simplified figure of the continuous process of this cumulative study The main phases of this research come from the action research methodology (Figure 4). This process has been continuous and cumulative. The research group has conducted action with real professionals, it has observed situations there, reflected on its experiences and observations, and have constructed versions of the methodology. The group has written case-specific reports. It has done all this with discussions with people involved in the Dissertations in Forestry and Natural Sciences No 153

23

Pentikäinen M.: Co-Development of Work and Information Systems

projects. The research group has done drawings and literature reviews, written scientific articles and got scientific feedback, and had more discussions and so on... The research group has done all the above. In addition, I myself with my own reflections and interpretations have done that. The output of this individual process is this thesis. Table 2 shows the methods and materials allocated to the separate research sub-questions introduced in Table 1. All the methods (literature, constructive analysis, action research, reflection) can be seen here as a part of the action research but as separate, too. Existing theories, previous research, existing methods, projects including cases, and the process itself are considered as the materials of this study. Table 2: Research methods and materials allocated to the sub-questions #

Questions

Method

Materials

SQ1

What is the current state of the art in both ISD and Work Development (methodologies)? How can the codevelopment of IS and work be supported? How could Activity Theory be utilized to provide a coherent methodology for the codevelopment of IS and work? How did the development of the ADISD methodology take place?

literature

existing theories and research

constructive analysis

literature, existing methods projects, cases, (also literature)

Ch 3,4

the process of development (Ch 4)

Ch 6

SQ2

SQ3

SQ4

action research

reflection: activity analysis

Focused on Ch 3

Ch (3), 4, 5

The methodology of this research comes from the discipline of Information Systems. The research has been done through action research processes. The materials are introduced in the following 24

Dissertations in Forestry and Natural Sciences No 153

Research Methodology

chapters: 3 Theoretical Background and 4 The Case: Developing the ADISD methodology, 5 Summary of Papers, and the original Papers I-VI.

Dissertations in Forestry and Natural Sciences No 153

25

Pentikäinen M.: Co-Development of Work and Information Systems

26

Dissertations in Forestry and Natural Sciences No 153

3 Theoretical Background This study is an analysis of the construction of the ADISD methodology (Activity-Driven Information System Development). The methodology was created to support the co-development of work and IS. This chapter introduces the theoretical basis and the background to the development of the methodology and strives to find relevant examples of work development and ISD methods. Here there are only fresh additions to the theoretical insights already gathered in Papers I-VI. In this chapter I seek from the literature IS and work development methods and motivation for the development of the methodology. This chapter is organized as follows: 3.1 Development of models and methodologies introduces general points needed in development; 3.2. Activity Theory presents the basics of this central theory for this study; 3.3 Work and work development looks into interesting work-oriented methodologies; 3.4. Software Engineering points out some points that are essential for the development of the methodology; and 3.5 IS and ISD explores methodologies that have a socio-technical emphasis and are evolutionary by nature.

3.1

DEVELOPMENT OF MODELS AND METHODOLOGIES

Concepts such as method, methodology, model, framework, technique, language, and tool are used in confusing ways in the fields of Information Systems (IS) and Software Engineering (SE). Usually, IS methods are socio-technical, whereas SE methods concentrate on designing and building the software-to-be. Methods which take into account both the social and the technical factors are called socio-technical (Baxter & Sommerville, 2011). “Method engineering” is a term proposed for the research field of

Dissertations in Forestry and Natural Sciences No 153

27

Pentikäinen M.: Co-Development of Work and Information Systems

the construction of development methods for information systems (Brinkkemper, 1996). The differences between the fields of IS and SE come from their roots. SE has a strong basis in, and history of, programming and IS has a broader scope through having organizational issues and changes included. Paper II inspected the methods used in both the IS and SE fields. This thesis does not attempt to define the line that demarcates the two disciplines but as a secondary benefit differences were identified during the study. By way of simplification, I would like to say that SE starts from thinking of people when they are end users, but IS takes people into account as a part of the organizational system, too. 3.1.1

Struggling with the Terms When striving to construct a methodology for the co-development of work and IS, the group looked into both sides of the process. It tried to take the required inputs to SE into account and to handle the required changes in work activities at the same time. The first of these raises a question about what the discipline we are working with is. Is it Information Systems Science, Software Engineering, or even some organizational theory? Struggling with this question for a while leads us to the conclusion that this study is necessarily multidisciplinary, and it is not only the disciplines that have already been mentioned that formed the input for this study. Overall and after all, the discipline is not the main point here; the goal is. The group has taken a pinch of each discipline that was needed for the recipe. This multidisciplinarity also gave us some new problems. Many terms are common but might have different meanings in different disciplines. In the end, like Baxter and Sommerville (2011), the group came to the conclusion that a socio-technical approach could be the answer. Socio-technical systems engineering lies between the systems engineering process and the change process. There might have happened some rapprochement between disciplines too.

28

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

What is a method or methodology? Brinkkempper (1996) stated that a method is something needed in engineering. Research methods are commonly known and divided into qualitative and quantitative methods (Myers & Avison, 2002). But development or engineering methods are usually divided into formal and informal methods (Pfleeger & Hatton, 1997). In conclusion, with this study, the method is an aid when engineering an artifact. The method is there for guiding the process. But the method is only a part of the methodology. An information systems development methodology is defined by Hirschheim et al. (1995, p. 22) as “an organized collection of concepts, methods, beliefs, values and normative principles supported by material resources”. In the light of the above definition, I would say that the group constructed the ADISD model and the methodology. A methodology is a systematic composition of methods, techniques, and tools. All these concepts – theory, approach, methodology, model, method, framework, technique, and tool – overlap with each other and it is hard to demarcate them clearly, but for the purposes of this study I made one interpretation (Figure 5).

Dissertations in Forestry and Natural Sciences No 153

29

Pentikäinen M.: Co-Development of Work and Information Systems

THEORY METHODOLOGY MODEL

AP

PR

OA

Method

CH

FRAMEWORK Method technique technique q technique tool

Method

Figure 5. Relations of core concepts An approach is a way of thinking and approaching the problem area with some theoretical background and “an ideology”. For example, the Activity-Driven approach is defined in Paper I. A model, on the other hand is something used in the methodology. The model provides a framework for the development process, and is usually illustrated by a figure (aka a model). A method gives the steps of the progress, and does not necessarily have its own beliefs, concepts, and values. A framework is related to a model. It gives frames but not necessarily the steps. The framework could often be seen as a checklist or used as a tool. But a framework can be used as a notation, too. The size of the framework, as compared to the other concepts here, might be the most arguable. Especially when the discussion is on the level of a discipline, the framework could be philosophical (Baskerville, 1991). In software engineering the

30

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

term is used to mean useful, reusable abstractions of best practice solutions (Johnson, 1992). A technique can be, for example, a notation (e.g., modeling language) or way of grasping the field. Scenarios, stories (Alexander & Maiden, 2004), or the wall-chart technique (SaarenSeppälä, 1997) could be seen as techniques as well. A tool can be any kind of manual or automated tool used to apply the method, e.g., CASE tools such as Rational Rose, or another drawing tool, such as Microsoft Visio, or a dictaphone. These definitions do not try to be complete but they are used in this way in this study (Figure 5). 3.1.2

Justification for Methods and Methodologies When thinking about what kind of methods we need, we should go backward, thinking about why we need methods and what we do with methods. We draw a picture when we cannot explain our purposes only with text or speech. With a picture we can see many things and connections at a glance. A shared language with drawings makes reading figures quicker. That is why we even have standardized modeling languages, e.g., UML. Standardization is performed in order to help software developers’ work. Most of the models and methods are for that purpose, too. Little by little, customers as end users have been taking into account, e.g., Use Case Diagrams. We can list many useful reasons to use methods: in brief, for gathering, structuring, and describing the ideas, requirements, analysis, and solutions, but also for communication and documentation. With a method one can: - analyze the problem area for and by oneself - analyze the problem between developers - analyze the requirements with the customer - adjust the client’s requirements to the software requirements

Dissertations in Forestry and Natural Sciences No 153

31

Pentikäinen M.: Co-Development of Work and Information Systems

-

-

document the software in order to share information in any phase of the process (requirement specification, design, implementation, testing, introduction, and maintenance) design the solution of a program describe the software requirement detailed to the programmer produce the code of the program in an automated way prove the solution before realization make the process visible make the process traceable simplify real-world abstractions pick up what is essential

To achieve all these good aspects of methods, the applier has to know what he or she is doing. When one has knowledge of several different methods (know-how), it is possible to apply them in useful ways in different situations (know-when). After our justification of methods, models, and methodologies, we should be aware that most of these precepts are not applied as intended (Baskerville, 1991; Truex et al., 2000; White Baker 2011). 3.1.3

Characteristics of a Proper Methodology A large number of reasons and needs for methods are presented above. Primarily, the group’s need was for guidelines for IS development in such way that work improvements would be supported, too. At first the group was searching for a method or methods. As explained in Chapter 2.3, this search turned into the construction of a model and methodology. In this explanatory construction of the methodology there are many inputs that deal with the important features of the methodology. Since information systems should support work, their developers should take work activities into account. Only workers can talk about their work, so communication (interaction) is one big characteristic of the methodology-to-be. Communication between developers and end users has been mentioned as a successful factor since 1965 (Churchman & 32

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

Schainblatt, 1965). The group searched for a method that would be understandable for developers and end users and also all the other necessary stakeholders. The group even hoped that the method would work as a common language and foster shared understanding, as was found important by Coughlan & Macredie (2002). Not all the stakeholders who might have something to do or say about the system-to-be always participate in the development process, but a representative group of each should take part. These groups are real participants. New challenges will arise when we do not know the end users or their skills, e.g., internet-based systems (Chakraborty et al., 2010). What was a good methodology in the 1990s is not necessarily still one today. As studied by Misic & Graf (2004), even the change during 1994-2001 was huge: “The environment in which systems analysts perform their job is characterized by constant change. This change and uncertainty places pressure on systems analysts to regularly and consistently update their skills as their tasks and activities also change. The last decade has seen technologies and tools like the World Wide Web and object-oriented analysis rapidly change the very nature of systems development. The explosive growth of Internet-based business (ebusiness) has provided a variety of opportunities and challenges. In addition, the process of systems development has faced pressure to be quicker, and the very process itself handled, at least in part, by individuals who are not systems analysts.” What the group wanted from the methodology was applicability at different times and in various kinds of projects. The methodology should be scalable, comprehensive, and multidimensional, but simple enough and documented and easy to use. The methodology-to-be should have: - an emphasis on similar development of IS and work - helpful guidelines to get the big picture, from which you can zoom in and zoom out (aka traceable) - tools to share understanding and support communication - an emphasis on modeling Dissertations in Forestry and Natural Sciences No 153

33

Pentikäinen M.: Co-Development of Work and Information Systems

-

scalability for projects of different sizes features of a framework, checklist, and process path a theoretical background to ensure its systematic and methodical nature practically useful frames a path to software requirements engineering applicability and adaptability over the years

Since there was no single methodology in existence that met the requirements, the group took a look at some promising disciplines in order to increase its understanding and gather the pieces needed to construct its own methodology.

3.2

ACTIVITY THEORY

Nowadays Activity Theory is a multidisciplinary research approach, but it has its roots in the Soviet Russian culturalhistorical psychology of the 1920s and 1930s. Scientists such as Vygotsky, Leont’ev, and Luria were the progenitors of AT. There were impacts from classical German philosophy (from Kant to Hegel) and the writings of Marx and Engels. (Engeström, 1999) 3.2.1

Classics Vygotsky (1978) argued against the stimulus-response psychology which was developed in the 1880s. He suggested that the interaction of a human being with the world is always mediated; the interaction with material world by material tools and mental interaction by immaterial signs. These mediating artifacts are the “X-factor” between a stimulus and a response, or rather, between a subject and the world (Figure 6).

34

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

X=mediation

Stimulus

Response

Figure 6. The fundamentals of Vygotsky’s studies lie in the X-factor (modified from Vygotsky 1978, p. 40). The principle of mediation is a cornerstone of AT since Vygotsky. Another widely applied concept introduced by him is the zone of proximal development. While studying the psychology of child development, he suggested that at any moment in a child’s life there is a zone of how far he or she can develop from the current state, and the limit of the zone can only be achieved if the child is supported by adults. The idea of a zone within which development is possible has been applied by analogy more widely in later AT. Vygotsky also focused on analyzing the process, not objects (1978, p. 61). Objects would be easier to analyze because they are stable, but it is more valuable to achieve the development (of psychological processes). Moreover, he strived for explanations, not only descriptions, of phenomena. Leont’ev (1977) introduced in the 1970s the concept of structured human activity. Vygotsky had already introduced the concepts of tools, instrumental operations, goals, and motives in relation to an individual. Leont’ev expanded Vygotsky’s proposition to activities by collective subjects. He introduced the levels of Activity, Action, and Operation as the structure of the activity. These are connected with each other so that activity consists of actions and actions consist of operations. Leont’ev emphasized that individual human actions can only be understood through the collective activity which they are part of. The three-level model was summarized by Engeström (1990) as in Table 3.

Dissertations in Forestry and Natural Sciences No 153

35

Pentikäinen M.: Co-Development of Work and Information Systems

Table 3: Leont’ev’s three-level model (Engeström, 1990, p. 197) Unit Activity Action Operation 3.2.2

Directing Factor Object/motive Goal Conditions

Subject Collective Individual or group Non-conscious

Engeström’s model of an activity system Engeström further developed Leont’ev’s theory of the systemic structure of human activity in his PhD thesis (1987) and presented it as a graphic model. The basic part of the model presents the elements of an action as a triad (Figure 7).

Mediating artifacts

Subject

Object

Outcome

Figure 7. A triadic representation of actions, according to Engeström (1999, p. 30) An action is performed by a subject (individual or group) on an object with mediating artifacts to achieve the outcome. Moreover, around the action there is the collective activity. Engeström expanded the model from action to activity by adding the community of the activity and the “social infrastructure” that mediates its relations with the object (division of labor) and the subject (rules). Figure 8 shows Engeström’s model of an activity with the example of international activity-theoretical collaborations.

36

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

Mediating artifacts:

x Works of Vygotsky, Luria, Leont’ev, etc. x Congress and newsletter

Subject:

Object:

Outcome:

Group of scholars organizing ’ISCRAT’

Central issues of Activity Theory

New intellectual tools and patterns of collaborations

Rules:

Division of Labor:

Conventions of scientific collaboration, statuses of

’ISCRAT’

Community: Colleagues inspired by Activity Theory worldwide

Compartmentalization based on disciplines, nationalities, languages, ’schools’

Figure 8. A complex model of an activity system with an example, according to Engeström (1999, p. 31). ISCRAT is the international congress on activity theory, later renamed to ISCAR. Other significant additions to Activity Theory by Engeström (1987, p. 89) include the concept of activity network and the proposition of contradictions within and between the elements of the model as the driving forces of change. These are discussed in the next section. 3.2.3

ActAD framework and integrated levels of analysis Kuutti (1991b) suggested that when information systems are being developed, the object of analysis should be a work activity rather than an information system. Kuutti (1994) and Korpela (1994) applied Engeström’s (1987) model of an activity system in information system development. Korpela’s motivation was grasping “the role of technology in work” in the context of a developing country (Korpela, 1994). Korpela et al. (2000) presented a modified framework to analyze work activities, including information systems, and all the elements of work. The modified model of the structure and dynamics of an activity (Figure 10) has the same elements as Engeström’s original model of an activity system and some new ones (comparison in Figure Dissertations in Forestry and Natural Sciences No 153

37

Pentikäinen M.: Co-Development of Work and Information Systems

9). This framework is called ActAD, Activity Analysis and Development (Korpela et al., 2000). The developers summarized the reasons for the modifications that they had “simplified the model by ignoring non-mediated relations, highlighted the dissimilarity of the elements by depicting them by different symbols, highlighted the difference between individual and collective elements by presenting various subjects and their instruments explicitly, elaborated on the ‘social infrastructure’ or ‘means of coordination’, and underlined the systemic relation between the elements by ‘mode of operation’”. They argued that the modified notation “can especially help in reminding of the multi-actor nature of an activity, which has often been reduced into single-actor models in DWR projects” (Korpela et al., 2000, p. 196). They also added mediated relations between activities.

Mode of operation, historical phases Collective actor: group or team

Relations with other activities, mediated by means of networking

Mediating artifacts

Means of coordination and communication: division of work, rules, etc. Actors, subjects

Subject

Object

Outcome

Means of work, instruments, facilities Work process: Object

Outcome transforms into

Rules

Division of Labor Community

Elements of a work activity

Figure 9. From triangular to oval presentation of work activity (Korpela & Mursu, 2003). Compare with Figures 8 and 10 for details. The starting point and a driver of this study were the ActAD framework and its Activity Theory roots. In this study Activity Theory has many roles. It is used in many ways. Two main roles are: 1. as a basis for the development of the ADISD methodology (Chapter 4)

38

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

2. as a framework for analyzing the development of ADISD methodology (Chapter 6) The ActAD framework is explained in Paper II. Its background as the bedrock for the ADISD methodology is shown in Paper I. When a work activity is being analyzed with ActAD, all the elements of an activity should be taken into account (Figure 10, Paper II). The object of the activity and its transformation to the outcome provide the motive for the activity. This transformation can be seen as the collective work process. The work process consists of the actions by the actors in which they use the means of work they have in their hand or in their mind. The actors’ actions are coordinated by means of coordination and communication, e.g., rules and division of work, but also schedules, meetings, multiuser software systems, etc. The individual actors taken together, i.e. the collective actor, might be a highly organized team or unaware of each other. At a given point of time, all the elements must more or less fit each other within the current mode of the activity, which characterizes the state of the system as a whole. Contradictions – tensions, discrepancies, misfits – within and between elements or between the activity as a whole and its environment cause changes in activities. Any activity develops through historical phases, each of which has a different mode, a different temporary fit between the elements. Which phases are the subject of interest, depends on the objectives of the analysis (Korpela et al., 2000). Furthermore, one activity is not separate from the rest of the world. There are always supporting activities and supported activities for any one activity. The relations between activities are also mediated, by means of networking. Material and mental artifacts thus appear in three different roles: as means of work in individual actions, as means of communication and coordination between actors within an activity, and as means of networking between activities (Korpela et al., 2000). After all, we should look at networks of activities.

Dissertations in Forestry and Natural Sciences No 153

39

Pentikäinen M.: Co-Development of Work and Information Systems

Mode of operation, historical phases Collective actor: group or team

Relations with other activities, mediated by means of networking

Means of coordination and communication: division of work, rules, etc. Actors, subjects Means of work, instruments, facilities Work process: Object

Outcome transforms into

Elements of a work activity

Figure 10. The structure and relations of work activity as a systemic entity (Korpela et al., 2002) Korpela emphasized in his doctoral thesis (1994) that activities must be studied in their organizational and societal contexts. This was later elaborated into a framework of four integrative levels of analysis (Figure 11; Korpela et al., 2001, 2008). The levels of analysis identified here are societal, organizational, group, and individual. Activity Theory operates on the group and individual levels, but other theories are needed on the “higher” levels.

40

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

Figure 11. Two times four integrative levels of analysis (Korpela et al., 2001)

3.3

WORK AND WORK DEVELOPMENT

Work can be viewed from several very different perspectives. In this section I briefly introduce some research approaches on work, in addition to Activity Theory which was discussed before and which is the perspective applied in this thesis. After that I introduce Change Laboratory as an important example of work development methodologies. 3.3.1

Various approaches on work Work can be studied, for example, from the perspectives of business (as a cost), political economy (as labor), operations management (as production processes), etc. In this thesis the viewpoint is on work as practice. Dissertations in Forestry and Natural Sciences No 153

41

Pentikäinen M.: Co-Development of Work and Information Systems

Paper VI briefly touches some approaches on work practice. The main scientific discipline studying practical work is ergonomics, which is called Human Factors in North America. It is a broad field, within which only organizational ergonomics (macroergonomics) deals with multiple actors, organizational structures, collective technologies, processes and practices. The emphasis in research is often on work processes (Damodaran et al., 1982; Holden et al., 2011). While organizational ergonomics aims at optimizing work processes and conditions to fit generic human characteristics, various quality-related approaches see work processes as business processes to be described for management purposes (Scheer & Nüttgens, 2000; van der Aalst et al., 2003). Business process reengineering (BPR) is a common and well-known technique that is used when developing the work of organizations. Even the original ideas of BPR study the inputs and valuable outputs of processes, and processes for the automation of those processes are often described. This has been criticized, e.g., by Virkkunen and Kuutti (1999) and further analyzed by Luukkonen and Mykkänen (2012). Work concerns individuals but also groups and teams. Activity Theory gives us an opportunity to analyze work on different levels (Engeström, 2000). 3.3.2

The Change Laboratory Engeström’s PhD thesis (1987) not only introduced the model of an activity but outlined a methodological cycle of “expansive developmental research”; that is, research that would develop an activity from the current form to a new one. Together with a group of researchers, he elaborated on the methodology which was named Developmental Work Research (DWR). A decade later a new generation of the methodology was launched under the name Change Laboratory (Engeström et al., 1996). The Change Laboratory is a work development methodology (its developers call it a method) which has its basis in Activity Theory and Engeström’s model of an activity. The use and 42

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

development of the Change Laboratory has been continuous since the 1980s. A new book that guides this methodology was published by Virkkunen and Newnham in 2013. This development of guidelines was made possible by getting the development needs from surrounding business and, on the other hand, having a growing group of researchers. The research group grew up as a community and the Center for Activity Theory and Developmental Work Research was established by Engeström and his colleagues in 1994 at the University of Helsinki, Finland. Nowadays this research center is named CRADLE (the Centre for Research on Activity, Development, and Learning, www.helsinki.fi/cradle), because of its having merged with another unit. The main analysis and development in the Change Laboratory takes place in five to twelve sessions and a follow-up session with practitioners and researchers. Collaborative analysis in sessions is supported by a 3*3 set of Surfaces and an Expansive learning cycle (see, e.g., Engeström, 2001; Virkkunen & Ahonen, 2011). Expansive learning actions are (according to Engeström & Sannino, 2010): 1. questioning, criticizing, or rejecting some aspects of accepted practice and existing wisdom; 2. analyzing the situation. Analysis involves mental, discursive, or practical transformation of the situation in order to find out causes or explanatory mechanisms; 3. modeling the new explanatory relationship in some publicly observable and transmittable medium. This means constructing an explicit, simplified model of the new idea that explains and offers a solution to the problematic situation; 4. examining the model, running, operating, and experimenting on it in order to fully grasp its dynamics, potentials, and limitations; 5. implementing the model by means of practical applications, enrichments, and conceptual extensions; 6. reflecting and evaluating the process, and 7. consolidating the new practices. Dissertations in Forestry and Natural Sciences No 153

43

Pentikäinen M.: Co-Development of Work and Information Systems

The surfaces are described in Table 4. Table 4: 3*3 set of surfaces used in analysis and design in a session of the Change Laboratory. The numbers are a possible sequence of steps through the process. (Virkkunen & Newnham, 2013)

Future Present Past

Model/Vision 7 6 5

Ideas/Tools 8 2 4

Mirror 9 1 3

The mirror is a collection of facts about work and organizations, e.g., statistics, customer feedback, or videotaped work situations. Ideas and tools start from identifying problem areas and the need for development and sharing concerns. The model and vision phases concern modeling the essential activities, changes, and visions. Engeström’s triangular model is the main tool used for the analysis and modeling. Sessions in the Change Laboratory method are held in order to gain a shared understanding of current problems and solutions. This method even concerns psychological issues such as sociocognitive processes and the dialogues in the session are analyzed by the researchers (Virkkunen & Newnham, 2013).

3.4

SOFTWARE ENGINEERING

Software Engineering is considered here briefly. Instead of working on the differences between the disciplines of IS and SE, it is worthwhile to say something about the endeavors of Software Engineering. The group’s attempt, when developing its own methodology, was somehow to cover the steps from work development to software engineering. First of all, the discipline is young and the extremely fast development of technology means that methods are developed and applied in practice in the best way that practitioners have the time and will for. 44

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

Something that is characteristic of SE is the process of producing the software product. This process could have different kinds of lifecycles, depending on the size and ideology of the software company and product. Figure 12 shows one of the most widely used lifecycle models, the “v-model”, first proposed by Rook (1986) and later used specially to emphasize the testing and evaluation of software (Burnstein et al., 1996, Toroi, 2009).

Project initiation

Operation and maintenance

Requirement specification

Product phaseout

Accepted software

Specificati on

Acceptance test

Structural design

Integrated software

Design

Integration and test

Detailed design Module Module Module designs designs designs

d Debugged Debugged Debugged modules modules modules

Code and unit test

Figure 12. The stages in software development confidence (Rook, 1986) Typically, software is analyzed, designed, implemented, tested, and introduced to the end users as an iterative process. The biggest part of SE work is maintaining existing applications. Many of the SE models take the nature of the project into account. Factors such as risks, evaluation, version, resources, cost estimation, and project planning come with SE, but are not considered here.

Dissertations in Forestry and Natural Sciences No 153

45

Pentikäinen M.: Co-Development of Work and Information Systems

3.4.1

Requirements Engineering This thesis does not contribute to the discipline of SE but it is worthwhile to create bridges and to try to see similarities in our purposes. Requirements Engineering (RE) focuses on the first steps of SE but covers the whole process, at least in terms of the need to manage the requirements. As Brooks in 1987 wrote, “the hardest single part of building a software system is deciding precisely what to build (Brooks, 1995).” According to Pohl (1994), there are three dimensions in RE: the specification dimension, the representation dimension, and the agreement dimension. These could also be called elicitation, modeling, and communication. Those are the attempts of the ADISD methodology, too. “There was a time when the epigram “requirements say what the system will do and not how it will do it” summarized all of requirements engineering. That time is long past. Research in requirements engineering has now produced a body of knowledge including terminology, methods, languages, tools, and issues acknowledged to be critical” (Zave and Jackson, 1997). RE has become more guided and methods have been developed for better specifications. “Modeling appears to be a core process in requirements engineering. The existing system has to be modelled in some way or another; the alternative hypothetical systems have to be modelled as well” (van Lamsweerde, 2000). Not only the guidelines but also the necessity of modeling before implementation arose as a central activity. The ADISD methodology should also support the modeling of the systems. “Whether viewed at the systems level or the software level, RE is a multi-disciplinary, human-centred process. The tools and techniques used in RE draw upon a variety of disciplines, and the requirements engineer may be expected to master skills from a number of different disciplines.”(Nuseibeh & Easterbrook, 2000). This is a fact that the group had in mind and tried to create a methodology for communications and sharing understanding. “RE by nature is about its core activities, such as eliciting, specifying and representing requirements.” (Sparrow et al., 2006). 46

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

Requirements engineering is said to be a most important phase of the success of the product. At the same time RE is seen as being large and difficult to achieve (Cheng & Atlee, 2007). I think that importance is identified but should be recognized in practice, too. “...Requirements engineering is about defining precisely the problem that the software is to solve.” (Cheng & Atlee, 2007). Traditionally, RE focuses on software, not environments or work activities. The core activities of Requirements Engineering are eliciting, modeling, requirements analysis, validation and verification, and requirements management as an umbrella for all these (Cheng & Atlee, 2007). For more about RE see, e.g., Wiegers (2003); Pfleeger & Atlee (2010). There have been many attempts to bridge requirements and programming (e.g., Harrison et al., 2000, Zundorf, 2001; Pastor et al., 2001). Even computerized tools have been used for this endeavor (the CASE tool, Computer-Aided System/Software Engineering), for example, Rational Rose (see more at IBM’s website: www.ibm.com). There have been attempts to apply Activity Theory to software engineering, especially to the specification of requirements. Turner et al. (1999) used AT to elicit requirements from the analysis of work situations. Martins & Daltrini (1999) used precepts from AT in requirements elicitation. Georg & Troup (2013) introduced a Requirements Language based on AT. Even UML-AT has been presented (Fuentes et al., 2006; FuentesFernández et al., 2010). To simplify, this means Activity Theory concepts are described with the Unified Modeling Language, a common language and notation used in software engineering. Andreev et al. (2012) also use Activity Theory and UML Use Case notations in their ActCPG framework (Activity Clinical Practice Guidelines, a combination of ActAD, Martins & Daltrini, and AT). (For more about UML see, e.g., Fowler, 2004). The ADISD methodology works as a bridge from work analysis and understanding the domain to the requirements engineering and specification. There are many similarities in

Dissertations in Forestry and Natural Sciences No 153

47

Pentikäinen M.: Co-Development of Work and Information Systems

working with RE or ISD (the need for a method, multilayering, multi-disciplinarity, human-centeredness etc.). 3.4.2

Achieving Usable Software Modules There are quality features, such as performance, maintainability, security, scalability, flexibility, and manageability, which are strived to achieve during the software engineering process. Features such as usability and maintainability are very valuable ones. There are many more quality attributes but here I focus only on these two, which are taken into account during the development process of the ADISD methodology. Usability is more than a nice user interface. Designing usability starts from understanding the context, work activities, and user needs (Nielsen, 1993; Martikainen et al., 2010). Nowadays, service-orientedness is the answer to the maintainability requirement (Arsanjani, 2004). Software-Oriented Architecture, SOA, means a special way to identify and specify the artifacts of service-oriented modeling. To simplify, architecture means defined pieces of software with defined interfaces and connections. The development of Software Engineering methods has come a long way from procedural paradigms to object-oriented, component-based, and serviceoriented, but improving maintainability still has the same principles as those recommended by Meyer (1988). These are the principles of modularity, i.e., high cohesion and low coupling of pieces of software. Modularity in its component-based form was one of the starting points for the development of the ADISD methodology to bridge the gap between RE and SE (Paper II). Component-based software engineering has raised interface definitions, the interoperability of applications, and compliance with standards as important factors (Szyperski, 2002, Toroi, 2009).

48

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

3.5

IS AND ISD

There are various definitions of Information Systems, which are broadly compared by Alter (2008a). This pair of words is often understood in everyday use as meaning a computerized tool used by people. In Information Systems Science it is defined rather as a system to process, store and give information in order to support a work system. In my definition an IS is a socio-technical entity and includes people, processes, and artifacts (Paper I). Moreover, I would say that a work system includes an information system, which includes a software system (Figure 13). This leads to the principle that developing the process of any of those should take the others into account.

Work activity system Information system Software

Information system development

Work development

Needs Analysis => Shared understanding and requirements

Work Information development system development RE RE

RE Software development

Software development

Figure 13. The shared understanding as the basis for the requirements (Luukkonen, 2012, p. 15; modified from Toivanen et al., 2007) If we think of information systems development as a sociotechnical process it includes the improvement of work activity by its nature. When talking about software we can easily think only of the implementation phase of the process, but this phase necessarily has an impact on the information systems and that has an impact on the work activity system. The following sections briefly introduce some ISD methodologies that are of interest to this thesis. Particularly the Work System Method, or WSM (by Alter, 1996-2013) is an

Dissertations in Forestry and Natural Sciences No 153

49

Pentikäinen M.: Co-Development of Work and Information Systems

important example of an ISD methodology that takes work as its starting point, like ADISD. 3.5.1

ISD Methodologies There are methodologies which emphasize the process or data modeling. Structured analysis (DeMarco, 1978) focused on to structure processes. At the same time there was Chen’s (1976) Entity Relationship (ER) model for data modeling. In the 1990s came Object-oriented methods (e.g., Booch, 1986; Shlaer & Mellor, 1988; Jacobson, 1992), which were an attempt to associate process and data design. The above-mentioned methods are focused on software but in the early history of socio-technical system design there are a couple of interesting methods: Soft System Methodology, or SSM (Checkland, 1981; 1999) and ETHICS (Mumford, 2000). Sociotechnical design has its roots in the Tavistock Institute over 50 years ago. At the same time there was a feeling that systematic thinking is the answer to analysis. The action research approach started in that same institute (Chapter 2). Roots of socio-technical approach lie in psychology and that makes these methods different (Mumford, 2006). SSM has good ideas for understanding and communication; Checkland (1999) calls them a rich picture. The ADISD group wanted to use rich pictures as well, even some cartoon-like ones. ETHICS takes work into account in design and with this method there are reasonable stages with simple steps that enable us to make sense of the problem and understand its scope. Since Engeström’s (1987) activity model and Developmental Work Research methodology were published, Activity Theory has also been applied in IT-related subfields, e.g., in HumanComputer Interaction (HCI) by Nardi (1996), Kaptelinin & Nardi (2006), Mwanza (2001), and Bertelsen & Bødker (2003) as well as in Computer-Supported Cooperative Work (CSCW) by Bardram et al. (1998, 2009, 2012), Kuutti (1991a), Collins et al. (2002), and Halverson (2002). Information systems development methodologies in the sense used in this thesis are those suggested 50

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

by Mwanza and Bardram above as well as the ones suggested by Hasan (1998) and Quek (2006). Bardram’s and Hasan’s groups are still active but otherwise the attempts at applying AT in ISD have remained once-off suggestions. 3.5.2

The Work System Method The Work System Method has evolved over the years since it was published in 1995 by Alter as work-centered analysis (Alter, 2013). Despite the name of the “Work System Method” (WSM), it could be seen as an ISD methodology. The Work System method is for the development of work-supported IS. According to Alter (2004a), WSM is very useful for ISD. The systemic nature of work can be identified and developed with the method. According to Alter (2013), the work system includes people and machines performing processes and activities, but also socio-technical work systems and even totally automated systems, i.e., computerized systems could be a work system. The basics of WSM lie in its approach to business but the scope is considerably richer than traditional business process methodologies (compared by Luukkonen et al., 2010). The business-oriented aspect of this method appears through attributing a high level of relevance to the customer and thinking about the values of products. Figure 14 shows the Work System framework, which provides a basic understanding of the work system. With the Work System framework we get a static view of the work system. It is like a checklist for taking a snapshot of the work system.

Dissertations in Forestry and Natural Sciences No 153

51

CUSTOMERS

ES

VI

GI

RO

TE

NM

RA

EN

ST

EN

T

Pentikäinen M.: Co-Development of Work and Information Systems

PRODUCTS/ SERVICES

PROCESSES and ACTIVITIES

PARTICIPANTS

INFORMATION

TECHNOLOGIES

INFRASTRUCTURE

Figure 14. The Work System framework (Alter, 2013, p. 78) Another essential framework of WSM is the Work System Life Cycle Model (WSLCM), which in turn is dynamic by nature (Alter, 2008b). WSLCM shows the change process of the work system. This iterative process has phases: initiation, development, implementation, and operation & maintenance. The special feature of this model is that WSLCM takes into account the fact that changes can be unplanned but still manageable. Alter (2013) calls them unanticipated opportunities and adaptations. Since the Work System, according to the definition, could be manual or automated, there are close similarities to the ADISD methodology. Examples of Work Systems that have been analyzed are differently-sized and different levels of computerization, e.g., renewing insurance policies, receiving materials at a large warehouse, controlling marketing expenses, operating an engineering call center, administrating grant budgets, etc. (Alter, 2013). These are business cases analyzed as work systems by students. The evolution of WSM has happened as a result of over 700 students using the various versions of the method and publishing revised versions (Alter, 2004a; 2006; 52

Dissertations in Forestry and Natural Sciences No 153

Theoretical Background

2008a; 2013). There have been very comprehensive, and sometimes even overly complicated, versions and simplified and applicable versions. In Figure 15 there is a version of WSLCM for IT innovations (2006).

Figure 15. The Work System Life Cycle Model for supporting IT innovations (bullets under headlines of the boxes) (Alter, 2004a, p. 6) A brief summary of using WSM is to remember to take into account the nine elements of the Work System framework (Figure 14) as a snapshot and be aware of the changing (and unplanned changing) nature of systems (Figure 15). Moreover, there are various helpful tables with further defined questions for the Dissertations in Forestry and Natural Sciences No 153

53

Pentikäinen M.: Co-Development of Work and Information Systems

analysis of work systems; for example, one table (from the 2006 version) is to identify the systems and opportunities, to analyze the system and identify possibilities, and recommend and justify changes (Alter, 2013).

3.6

SUMMARY

The general situation is that work is more and more informationoriented and information systems and software are becoming more complicated. The disciplines are young and undergo change and development all the time. This means that methodologies have to be adaptable or they are not used. I explored the methodologies that cross over the disciplines to find features needed for the ADISD methodology, which will support the co-development of work and IS. The context of the PlugIT and ZipIT projects was healthcare, and from the view of methodologies it was natural to start with Activity Theory. While applying it, the research group adapted it and documented this use, and eventually realized that it had created the ADISD methodology (Chapter 4). There are attempts to conceptualize the development of methods and methodologies (e.g., Brinkkemper, 1996; Song, 1995) but not necessarily to analyze why the methods are as they are or what they should be like. In this thesis an analysis of the development of the ADISD methodology is carried out in Chapter 6.

54

Dissertations in Forestry and Natural Sciences No 153

4 The Case: Developing the ADISD Methodology This chapter introduces the process of developing the methodology. The outcome of the process, the ADISD model and the methodology, the so-called artifact of the case, is presented, too. Developing the methodology happened in the context of healthcare in several projects. This chapter is organized as follows: Section 4.1 Healthcare and health information systems as a context introduces briefly the specific context in which the methodology development took place; 4.2 Projects as sites of the development of the methodology provides an overview of the projects during the development process; 4.3 PlugIT; 4.4 ZipIT; and 4.5 MyWellbeing present the most significant projects in more detail, and 4.6 Other projects and applications around developing the methodology provides complementary information around the main projects. Versions of the methodology are presented among the stories of the projects as outcomes and applications.

4.1

HEALTHCARE AND HEALTH INFORMATION SYSTEMS AS A CONTEXT

In Finland it is characteristic that healthcare and social services are provided by the local authority, the municipality. The legal responsibility to provide these services leads to non-profit organizations, the activities of which are run from a different perspective than those of organizations in the private sector. Healthcare, as a context, is challenging in many ways. The information needs of health professionals when taking care of patients are various. When, for example, healthcare is compared with some commercial field where information is mainly numerical and strategies are run by finances, healthcare is Dissertations in Forestry and Natural Sciences No 153

55

Pentikäinen M.: Co-Development of Work and Information Systems

different and run by caring. The information involved in healthcare, even that concerning only one patient, is various: e.g., numerical data, narrative descriptions, lists, separate recipes, and X-ray imaging. Decision making in healthcare is a complicated process performed by professionals. They use the information that they can get. The most important informant is, of course, the patient him- or herself. Issues such as legislation, the security of information, or the quality of information systems can have an effect on the patient’s life. Developing the ADISD methodology in the context of healthcare was challenging.

4.2

PROJECTS AS SITES OF THE DEVELOPMENT OF THE METHODOLOGY

As introduced in Chapter 2.3, the development of the methodology happened in several action research projects. These projects form the process of developing the methodology. After each action research cycle the methodology was released as a new version or at least some upgrades were included. That version of the methodology was both the input and output of an action research cycle. The development of the ADISD model and methodology took place in several research projects (Figure 16): Pre-Indehela, Indehela-methods, PlugIT, ZipIT, Indehela-context, ChinaFinland eHealth Partnership, MyWellbeing, and SOLEA. All the projects are introduced briefly in Paper I. This study focuses on three of these projects: PlugIT, ZipIT, and MyWellbeing, because I worked with these projects.

Pre

Indehelamethods 19982001

PlugIT 2001-2004

ZipIT 2004-2007

IndehelaContext 2004-2007

China-Finland eHealth Partnership 2007-2008

MyWellbeing 2008-2010

SOLEA 2008-2011

Figure 16. Research projects around the process. 56

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

These three projects that are in focus were funded by the Finnish Funding Agency for Technology and Innovation, Tekes. In addition, software companies and healthcare organizations, healthcare districts, the Finnish Work Environment Fund, and the Academy of Finland were involved. The projects form the materials of developing the ADISD model and methodology (Figure 17).

Reflection: Analyzing the development of the methodology as an activity

Preliminary research

Action research

Action research

Action research

The Case: Developing the methodology

Pre

Indehelamethods 19982001

PlugIT 2001-2004

ZipIT 2004-2007

IndehelaContext 2004-2007

China-Finland eHealth Partnership 2007-2008

MyWellbeing 2008-2010

SOLEA 2008-2011

Figure 17. The relationships of the research projects to the process. (combination of Figures 3 and 16) The preliminary research took place in pre-Indehela and Indehela-Methods projects (1998-2001). The most important output of these projects was the ActAD framework (Korpela et al., 2000). The projects contributed to the AD approach by identifying the requirements for a socio-technical approach and the methods to be used in ISD (introduced in Paper I). The theoretical basis for the approach was established and the first action research experiments were done. The focus was on work activity and activity networks (Paper I). The contexts of the cases were healthcare and software companies in Finland and Nigeria. Dissertations in Forestry and Natural Sciences No 153

57

Pentikäinen M.: Co-Development of Work and Information Systems

4.3

PLUGIT PROJECT

The PlugIT project (http://www.oppi.uef.fi/uku/plugit/english/) in the years 2001-2004 focused on integration and integration methods in the healthcare context (Mykkänen et al., 2004). There were 13 pilot cases, three of which concentrated on activitydriven methods. The project focused on an integration specification process (Figure 18). This process was developed and piloted during the project. -health know-how -pilot systems -existing technologies -functional and quality requirements

Health service provider (e.g. hospital)

Work process improvement

Application vendors

-existing integration solutions -pilot systems -existing technologies and tools -existing production and quality process

product implementation

Acceptance Deployment Adaptation

Implementation (pilot)

Integration need Project plan

Piloting plan

-list of initial integration needs -specification guidelines -content definitions for specifications -example solutions -templates -tools

Description of current status

local requirements specification work iteration, new versions

-documentation -evaluation implementation -integration process work development -generalization -method validation

4. IMPLEMENTATION DESCRIPTION

Moderator (PlugIT project)

Solution specification

Context studies, technology evaluations

Open, reusable integration specifications Requirements analysis 1.INTEGRATION REQUIREMENTS specification 2. PLATFORMwork INDEPENDENT INTERFACE SPECIFICATION 3. TECHNOLOGYSPECIFIC INTERFACE SPECIFICATION

Figure 18. Organization of integration work in PlugIT project (Mykkänen et al., 2004, adapted from CORBA).

58

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

4.3.1

Pilot Cases The methodological results of the whole PlugIT project fell into three categories: (1) the open integration specification process (in Figure 18); (2) Activity-driven methods for requirements analysis; (3) Applications and integration methods (for more see Mykkänen et al., 2004). In this study the focus of interest is on (2) Activitydriven methods for requirements analysis. Activity-driven methods were piloted in three cases of the PlugIT project: 1. information needs exploration in an activity network of home care; 2. an information systems needs study in an organizational unit of maternity care; 3. an outline of an activity-based method for the software requirement specification of data administration software. For each pilot case a research group was constituted. The research groups had 3-6 persons from the multidisciplinary and multiprofessional research group. The PlugIT workers came from four different units: 1. University of Kuopio, Health Policy and Management Department, SHIFTEC; 2. Savonia Polytechnic, Savonia Business, Information Processing; 3. University of Kuopio, IT Services, Health Information Systems S R&D Unit, and 4. University of Kuopio, Department of Computer Science, Software Engineering. This meant that the teams had viewpoints such as those of healthcare professionals, IT professionals, methods and tools developers, and software professionals. There were also six software companies participating in the project, nine software companies following it, and eight healthcare organizations participating in Finland. The team leaders of each pilot case met each other at the “tiimitiimi” (the team of the teams) meetings. Those meetings concentrated on the integration process and methods (Figure 18). During the meetings the group discussed the methods, pilot cases, plans, and the context itself, because not all of the researchers in group were familiar with healthcare. Moreover, the group discussed and defined the terms to use in Finnish and in English in connection with the context and with the development of the method (PlugIT, 2004). Dissertations in Forestry and Natural Sciences No 153

59

Pentikäinen M.: Co-Development of Work and Information Systems

As stated above, the interesting outcomes of the PlugIT project for this study are the activity-driven methods that were piloted, developed, and described. The ActAD framework was used in three differently sized pilot cases. The PlugIT project’s activitydriven cases are introduced in Paper I. The methods used with the cases are described in Finnish (2004b) by Toivanen et al. They are also briefly presented in the following articles: home care (Paper III); maternity care (Häkkinen & Korpela, 2007), and data administration software (Riekkinen, 2004). The pilot cases were real IS development projects in which members of the research group took part as action researchers. Table 5 shows the starting points, main methods and tools, and outcomes of each case. Table 5. Summary of the activity-driven pilot cases within the PlugIT project (adapted from Paper I) Objective

Starting point

Information needs in home care

Previously poorly known “gray area” of home care services Information management problem in service providers’ network Need for new software for user identification

Maternity clinic information flows Requirement specification of data administrati on software

Main Methods and Tools 12 thematic interviews with Activity-driven themes and questions, workshops with wall technique, enriched ActAD Focus group interviews and brainstorming workshops with Activity-driven themes, wall technique, activity network with information flows and means Interviews with Activitydriven themes, work process models, enriched ActAD, UML use cases and data modeling, UI interaction design

Outcomes Describing the activities and information needs in a holistic way List of the most problematic spots from one organization’s point of view, integration needs software requirements

The driver of these three cases was to use the ActAD framework in different kinds of situations and to formulate activity-driven methods to be used in IS development. The group selected the ActAD framework under study in order to move the 60

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

development of information systems towards work-centered IS development. This kind of IS development means better facilitation of the workers discharging their responsibilities. If the starting point of the development is work activity rather than new technological opportunities, this leads to IS which supports the work instead of IS which forces the work to be done in a certain way. The methods used in the cases were activity-driven, the materials were the real object of development (pilot cases), and the resources were the research groups and participants, mostly from healthcare organizations. Each case had specific goals, but in addition the research group had a shared goal of method development. The group got case-specific results but also shared the experiments involving the development of the activity-driven methods. The research groups had meetings around the cases and they reported the case-specific findings. In addition the group had “Activity club” meetings which were focused on discussions of Activity Theory and the ActAD framework (Figure 19). Each element of the framework was discussed and thought was given to what they might mean in different cases. This led to a deeper understanding of this theory and ensured that the group used the theory with the same principles in different cases.

Dissertations in Forestry and Natural Sciences No 153

61

Pentikäinen M.: Co-Development of Work and Information Systems

Figure 19. Picture from a workshop of the research group. 4.3.2

Outcomes The outcomes of the activity-driven cases of the PlugIT project were: x case-specific results; x experiments with using activity-driven methods (Table 5); x three method descriptions (Toivanen et al., 2004b): exploratory method, rapid assessment, and componentbased; x the ActAD-based method, which provides a holistic overview of the domain and a checklist for studying various areas or systems, and x the main activity-driven method that resulted from the home care case was a traceable requirements chain from activities to software components (see Figures 21-24) The goal was to test and document activity-driven methods to be used in information system development projects. Even the cases were separated and differently sized and had their own 62

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

focus of development. The methods described had three main phases or levels (Toivanen et al., 2004b). For example, the version of the model used in the home care case (Figure 20) had the following phases (Toivanen, 2004): x structuring and describing the problem domain (feasibility study, interviews, analysis, cross-checks, and IT needs spots); x aiming at and focusing on the targets of development (work processes and information needs in more detail), and x focusing on the software specification (architecture, use cases etc.).

Figure 20. The early version of the model used in the home care pilot case of the PlugIT project

Dissertations in Forestry and Natural Sciences No 153

63

Pentikäinen M.: Co-Development of Work and Information Systems

4.3.3

Applications The traceable requirements chain from activities to software components was tested with the home care case (Paper II, Toivanen, 2004, 2004a, 2004b). The chain followed the steps of focusing. First, the activity network was sketched, and then essential activities and their elements were described. The work processes with the means of managing information were figured out (Figure 21). From the work processes we can see the actions of the actors. Actions which are done or to-be-done with the software can be seen as UML use cases (Figures 22-23) and use cases can be described as sequence diagrams of software requirements’ specifications.

Oma

As iakas

hoitaja

BASIC SERVICES

Ambu lanssi

Käynti tiimituvalla "T yökansio"

Herättelee ja kys elee vointia

Team

Various actors: PLANNING VISITS

Puhuu sekavia

Kyselee vaivoja ja mittaa kuumeen

Privacy Protection Tarkistaa illan merkinnät

VIEST IVIHKO

FAX Kysyy neuvoja

Plan

Ksh

Hätä -

Soittaa ambulanssin

Home helper DOMESTIC AID

keskus

Customer?

Doctor

Physiotherapist

Home Helper Nurse

Am bulanssi saapuu

Relative/ Friend On-call Cleaner helper

Kertoo tilanteen, antaa Hopasun

Kirjaa käynnin

Service Providing

Siirretään ambulanssiin

HOPASU

Kuljettaa asiakkaan

Plan

RN, MD: HOME HEALTH CARE

HOPASU

Kirjaa tilanteen VIEST I VIHKO

päivystykseen

Ilm oittaa omais ille Omai nen

Ilm oittaa kpo:lle

Kirjaa poiss aolon

Kpo

Ateri apalv.

POISSA OLOT

Figure 21. Project’s example of the method’s steps of focusing (Toivanen, 2004) Activity network -> Activity -> Actions of process. (Essential here is the shapes of the diagrams and the details are not relevant)

64

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

Asiakas

Oma -hoit.

HOPASU

Tarkista perustiedot

Tilaus Tarkistaa perustiedot Kertoo tiedot Kirjaa tiedot

Hoitojaksojen tiedot

Asiakas

Tarkista sairaus- ja lääkitystiedot

Asiakasrekisteri

Terveyskertomusjärjestelmä

Muita järjestelmiä:

Ksh

Tarkistaa sairaus- ja lääkitystiedot

Kartoita tarpeet ja toiveet

Erikoissairaanhoito

Pegasos

Näyttää epikriisin ja reseptit

Kirjaa tiedot

Tekijä

Kartoittaa tarpeet & toiveet

Tallenna HOPASU

Käynnistä palvelut Avuntarve aamutoimissa

Aluetietojärjestelmä

Kpo

Sosiaalitoimi

Tulosta HOPASU

Kirjaa tarpeet & toiveet Lisää käynnin listalleen

HOPASU HOPASU HOPASU

Allekirjoittaa HOPASUn

Jakele HOPASU

Yksityinen palveluntuottaja

HOPASUn jakelu eri tahoille

Figure 22. The project’s example of the method’s steps of focusing (Toivanen, 2004a) Actions of process -> Use cases. (Essential here is the shapes of the diagrams and the details are not relevant)

Figure 23. The project’s example of the method’s steps for focusing (Toivanen, 2004a) Use cases & sequences. (Essential here is the shapes of the diagrams and the details are not relevant)

Dissertations in Forestry and Natural Sciences No 153

65

Pentikäinen M.: Co-Development of Work and Information Systems

The traceability from domain modeling to software design with starting from activities is studied also by Mykkänen et al. (2006) (Figure 24)

Figure 24. The modeling chain supporting traceability (Mykkänen et al., 2006) In the context of home care the group used ActAD as an explorative requirements analysis method (Paper III). Another context was a maternity clinic, which is a multifaceted system, too. With the maternity clinic ActAD was used as a rapid participatory assessment method for integration needs (Häkkinen & Korpela, 2007). This method was based on only two workshops, each two hours long, called the critique workshop and the bridged future workshop. The contribution of the PlugIT project to the Activity-driven approach and the method was describing the IS analysis process as a process with three different levels of detail and using the ActAD framework (Toivanen, 2004). Moreover, showing the traceability of the path from activities to software requirements was promising (Figures 21-24). Using the ActAD framework in problem domains of different sizes gave us an idea of the scalability of the framework.

66

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

4.4

ZIPIT PROJECT

The ZipIT project (http://www.oppi.uef.fi/uku/zipit/english/) (in the years 2004-2007) focused on narrowing the distance between work improvement and the development of information systems. The project was involved the following: three research groups from the University of Kuopio (Department of Computer Science, Health Information Systems R&D unit, Health Policy and Management Department), one group from Savonia University of Applied Science, six software companies, and four healthcare organizations. 4.4.1

Rationale and Realization The goal of the ZipIT project was to continue the study by testing and developing the model outlined in the previous PlugIT project. The Activity-driven approach was seen as being promising but needed more experimentation and validation. The basics for the model and methodology under construction were: three-levelness, the parallel development of work and IS, and understandability (Figure 25). Understandability means that the model should work as a tool for communication between professionals in software companies and healthcare organizations. The group strove to develop a model which could provide a common language for all the stakeholders taking part in a process.

Dissertations in Forestry and Natural Sciences No 153

67

Pentikäinen M.: Co-Development of Work and Information Systems

Health care organization

services and processes

1. IT needs

business

- activity networks -information system

Software Company The tool of communication

2. Information management

Information system development needs - processes, information architecture

3. Requirement specification -Use case descriptions

introductions

software

Activity networks development

--Workflows, task descriptions

changes in the activities

Health care organization

Figure 25. Starting point and ideology of the ZipIT project: work and IS development in parallel (zipped up together). The ZipIT project had two main processes in parallel: the development of the model and methodology as well as the pilot cases (Figure 26). The process of the development of the model and methodology was based on the cumulative knowledge of theories and experiences. Descriptions of the methods were balanced with background theories, e.g., activity-driven, participatory, user-centered, and software requirements. Processing the knowledge of theories and experiences led to better descriptions of the methods. Through processing those with the new experience gained from the pilot cases the group filtered the methodology as a guide for activity-driven information system development (upper part of Figure 26). Methodological discussions took place between the research groups. In terms of action research the group took part in real development projects in the pilot cases (lower part of Figure 26). 68

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

For each case 2-4 researchers took part and the rest of the group worked as consultants when the cases were discussed. Each of the cases gave us an understanding about organizations and the way they carry out their work duties. The target of development was based on the goals of the case. Each pilot case was affected by the methods that served as a plan for research group. The target of development was the balance between work activity and the development of information systems. In addition to pilot case-specific results, the group produced study reports.

Figure 26. Elements around methodology development activity in the ZipIT project. Zippers demonstrate the need for balancing (Paper VI). 4.4.2

Pilot Cases During the project there were 9 practical pilot cases of the development of information systems in the context of healthcare. These pilot cases are described in Papers I and VI (Table 6). The pilot cases affected the methodology that the group constructed and the Activity-driven approach affected the cases and how they were carried out. Dissertations in Forestry and Natural Sciences No 153

69

Pentikäinen M.: Co-Development of Work and Information Systems

Table 6. Summary of the activity-driven pilot cases within the ZipIT project (modified from Paper I) Objective:

Starting point, rationale

Main Methods and Tools

Outcomes

Crossorganizational care chain

Inadequate tool for communication between organizations Old hardware and software

Cooperative planning, focus group interview, storytelling, interactive writing Enriched ActAD, fishbone diagram, activity story Workshop, interactive modeling, patient journey walkthrough Activity story, AD diagrams, workflow diagram Usability study, workshop User interface sketches

Process integration between two healthcare organizations Fit the intended technical solutions to the work process

Further development of digital radiology Medicinal care system

Inadequate software

Introduction of electronic patient records

Software already acquired

Mobile phone application for home care

New market area, new application, problems in current practice

Developing the home care activity with new technology Purchase and requirement process of operation application in hospital Analysis of the use of a decision support system

Problems in current practice

Analysis of the current practice of fundus imaging

70

Decision to acquire new software

Software already acquired, how or why it was used not known Local differences in practices

Cooperative planning, workshop, interview Wall technique, semi-structured question list and tables, AD diagrams, flow diagram Cooperative planning, workshop, observation Textual reports, datasheets and UI specifications, wall technique, UML activity diagram Cooperative planning, workshop, interview, observation Textual reports, wall technique, AD diagrams Cooperative planning, workshop, observation, visits to peer organizations which had already acquired similar software Enriched ActAD Interviews Usability testing

Interviews Modeling AD diagrams

New component for existing electronic patient records (with better usability) Guidelines for introduction of electronic patient records Prototyping new application

Reporting how new technology could help User requirements for the software acquisition process Reporting for better usability

Recommendations on better usefulness of software and common course of action

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

The focus of the ZipIT project was on intertwined IS and work development. Organizational and cross-organizational settings were tested. The emphasis was put on software companies’ clients’ views of their requirements (Minkkinen & Eerola, 2007). In addition to the case-specific results (described in Table 6), the Activity-Driven Information System Development model was constructed (Paper IV). Moreover, the theoretical underpinnings of the ADISD model were clarified (Mursu et al., 2007). 4.4.3

Outcomes The activity-driven methods used during the PlugIT project had three steps. These steps were seen as phases. During the ZipIT project the group identified three levels, too. The group wanted to keep the levels and the phases separated. It tried to describe the levels, phases, methods and documents in a single description of the methodology to be as a model. One of the versions is shown in Figure 27. The group thought that certain tools are useful on certain levels and in certain phases.

Dissertations in Forestry and Natural Sciences No 153

71

Pentikäinen M.: Co-Development of Work and Information Systems

Three Phases and three Levels of the model Methods/tools

Phase 1 Present Present state

Level 1 Activity network

notation

Workshops

Level 3

Goal state

Tarkastelun Modeling of work kohdistaminen ja and information syventäminen flows

UI design

Development of work supportive information system

Development spots Goal state work flows information architecture

Phase 3 To-do

Software requirements

method x

Tools and actions

Network of activities essential information entities

Present state

Focusing

User interface descriptions

Phase 2: Goal Development spots

information architecture

Check out

Observations

Network of activities essential information entities

work flows

workflow diagram

Level 2 Work flow and information management

Interviews

Focusing

Pictures, diagrams ActAD-

Check out

Analyzing and modeling of the domain

Observations

Improving work activities

Plans of work improvement

UM L

Figure 27. One of the versions of the model under construction in the year 2005. Essential here is the shape of the model and the details are not relevant. After this comprehensive but complex version of the model had been created the group realized that all of the levels and phases overlap with each other. The phases form the time dimension of the development. The levels represent the examinations and descriptions zooming in and out. When one moves from Level 1 to Level 3 the descriptions become more detailed. The other way around, zooming out, means more general descriptions. A simple version is shown in Figure 28.

72

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

Figure 28. A simple version of the ADISD model Eventually, the group started to describe the ADISD model as a table with three rows as the levels and three columns as the main phases (Figure 29). This model is explained in Paper IV. The phases go forward quite simply. Only stopping and thinking is needed between Phases 2 and 3. There should be validations and verification of whether we have done the “right things” so far and whether we have done “things right”. Moreover, we need to make decisions as to whether to continue or not.

Dissertations in Forestry and Natural Sciences No 153

73

Pentikäinen M.: Co-Development of Work and Information Systems

Phases Levels & Describing models 1. Network of activities &

Information landscape

ORGANIZATION UNIT

PAPER

DATA STORE ACTIVITY

2. Work activity (processes) & Information system (data flows)

ACTIVITY

Overview; what services or products do we produce and for whom, who are our stakeholders, what are the essential activities?

Considering - context (buildings, infrastructure, legislation etc.) - changes in the network of activities and - changes in the information landscape.

What are the most important information entities in our activities, where are these information entities located, how do we communicate in the network, and which tools are used? Zoom in on our essential activities: who is involved in the work process, who makes decisions in different stages, how is the work coordinated, what means (mental and physical) are needed in the process, what information is needed and where from, what is written down, and what information tools are used?

WORK PROCESS

3. Actions & Information tools valinta

Text: OK

USE CASES

3. Making the plans for change

DRAFT OF USER INTERFACE

Zoom in on our essential work processes, what actions need to be developed, and what detailed information sets and data items are needed in central actions. What information tools (forms etc.) are needed? We can outline use cases from the actions if software is used.

In information system development (e.g. the introduction of new or tailored software or an integration project), we have to outline how and where change affects the network of activities and information landscape. Zoom out from processes and information flows to see the range of changes.

In ISD (e.g. the introduction of new or tailored software or an integration project), we have to outline how and where we utilize IS in work processes, what our work processes will be after the change, and what impact change has on the information systems and data flows. Zoom out from the actions (and use cases) and information tools to see the range of changes. In information system development, we have to outline how we utilize the IS that will be developed in actions, how it would be used, and what effects the changes have on information tools. Users' needs, wishes, and requirements must be considered. We can specify use cases from the actions if software is or is-to-be used.

Validation, verification, decisions

ORGANIZATION

1. Analysis for 2. Design for shared understanding of shared understanding of the goal state the present state

Planning - re-organized services and activities - purchase of software and hardware and - systems integration. Planning - changes in the information system - changes in work processes e.g. introduction of new software and new work practices step by step, unit by unit

Planning - changes in information tools (e.g. software) and their use, - changes in actions and duties.

Figure 29. The ADISD model in a table form (Paper IV) The roots of the model and the methodology are in Activity Theory (e.g., Vygotsky’s zone of proximal development) and the ActAD framework (Mursu et al., 2007). There are also ideas of the integrative levels (Korpela et al., 2001) shown in section 3.2.3. The societal and organizational levels are merged into Level 1 in the ADISD model, since the model and methodology was used mainly locally and only in Finland.

74

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

4.4.4

Applications With the pilot cases the group modeled situations with the ActAD framework, not only to use in discussions with the professionals, but also to understand the contexts. One of the modeled situations is shown in Figure 30. The object of this pilot case was the cross-organizational care chain.

Figure 30. The present state of a network of activities and information landscape in the pilot case: referral and referral feedback (adapted from Luukkonen et al., 2007). Case specific reports published in Finnish present many drawings such Figure 30 (ZipIT, 2007). Paper I shows where the model is used. More detailed examples are in the thesis of Luukkonen (2012).

4.5

MYWELLBEING PROJECT

MyWellbeing (http://omahyvinvointi.utu.fi/) was a national research and development project on a citizen-centric wellness management concept, named “Coper”. The project had three Dissertations in Forestry and Natural Sciences No 153

75

Pentikäinen M.: Co-Development of Work and Information Systems

units from the University of Kuopio: the Health Policy and Management Department, Health Information Systems R&D Unit, and Department of Computer Science, but units from the Universities of Tampere, Turku, and Åbo were also involved. Turku played the role of the coordinator. There were also 10 organizations involved, such as the postal service, software companies, and healthcare organizations. 4.5.1

Rationale and Realization The aim of the MyWellbeing project was to present the concept of the “Coper”, which is to be a personal health and wellbeing system. This “Coper” concept was studied from the design point of view (Paper V and Tuomainen et al., 2010b). The project studied various Personal Health Records and their opportunities to help individuals to cope with their health information (Tuomainen et al., 2010a). Moreover, Activity-driven information analysis from individuals’ point of view was the main goal. The target group was families having a baby (families that had a baby or that are going to have a baby) (Tuomainen et al., 2010b). Families having a baby were interviewed about, e.g., their information management habits and their needs for social media support in their life situations. Palmén et al. (2012) describe those studies and results in the greater detail.

4.5.2

Outcomes and Applications We cannot say that the ADISD model and the methodology gained new features but rather the group saw its applicability. In Figure 31 the ADISD model is applied to person-centered information management and to analyzing that on different levels and in different phases. With this case I focused on developing concepts and reference models to person-centric system design.

76

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

Figure 31. The Activity-Driven ISD model adapted to person-centered information analysis (adapted from Paper V). The MyWellbeing project analyzed information associated with individuals’ life activities. The role of the Coper was analyzed and concepts for its design were developed and documented (Tuomainen et al., 2010b). In Figure 32 is an example of an analysis of information entities around life activities and the Coper.

Dissertations in Forestry and Natural Sciences No 153

77

Pentikäinen M.: Co-Development of Work and Information Systems

Individual

n

has

Life situation (activity of life)

n

n n

n

n

include

use

need

associate

n

n

Coper

n

n

n

include

n

Information entity n

n n

Data item

n

compose of

include

Figure 32. Information analysis of elements around the Coper (adapted from Tuomainen et al., 2010b). The MyWellbeing project and, especially, information analysis are described in Paper V.

4.6

OTHER PROJECTS AND APPLICATIONS AROUND DEVELOPING THE METHODOLOGY

During the ZipIT project the ADISD model and methodology was introduced in the form it has nowadays (Paper IV). Since then several projects have utilized the methodology and created some new applications. All these are introduced in Paper I. The China-Finland eHealth Partnership project in the years 2007-2008 applied the methodology to the preliminary domain analysis of the maternity path (Luukkonen et al., 2011). The INDEHELA Context project in 2004-2011 used an activitydriven approach to the contextual analysis of IS use to be utilized in IS development and broadened it towards Landscape Methodology (Korpela et al., 2008). Within the SOLEA project (2008-2011) an Activity-driven approach was used, for example, to analyze the process modeling, which is a typical way to model organizations’ work (Luukkonen et al., 2012). 78

Dissertations in Forestry and Natural Sciences No 153

The Case: Developing the ADISD Methodology

Teaching the Activity-Driven approach has been done since Korpela’s thesis (1994). Newer teaching experiences are documented by Luukkonen et al. (2011a). The application of the Activity-Driven approach in business started after the ZipIT project, since researchers involved in the project moved into business (Papers I and VI). This chapter introduced the materials and methods used in developing the ADISD model and the methodology.

Dissertations in Forestry and Natural Sciences No 153

79

Pentikäinen M.: Co-Development of Work and Information Systems

80

Dissertations in Forestry and Natural Sciences No 153

5 Summary of Papers This chapter introduces the research papers of this study, the aim of which is to tell the story of the development of a methodology. This can be seen as a process, but not a simple and strict one. The papers represent different times, projects, and phases of the process of the development of the methodology. We should consider the papers as windows on the process. All the papers have their own focus, even though the same themes recur. This chapter is organized as follows: Section 5.1 Relations of the research papers and projects links the projects introduced in Chapter 4 with the research papers; 5.2 Relationships and roles of the research papers explains the division of roles between papers during the research; Sections 5.3 – 5.8 summarize each research paper in turn; and 5.9 The contributions of the Papers summaries the findings from the papers in light of the research sub-questions.

5.1

RELATIONS OF THE RESEARCH PAPERS AND PROJECTS

The projects around the process impacted on this research and research papers (Figure 33). The development of the ADISD methodology took place within the projects and pilot cases of healthcare. In each project there was a different research group, different healthcare organizations and software organizations, and different cases on which to apply and test the methodology.

Dissertations in Forestry and Natural Sciences No 153

81

Pentikäinen M.: Co-Development of Work and Information Systems

VI How to Co-develop Work Activities and Information Systems: Lessons Learnt from Developing the ADISD approach

II From information systems requirements to software components - home care case

IndehelaPre-Indehela methods 1998-2001

III Gathering, structuring and describing information needs in home care: a method for requirements exploration in a "gray area"

PlugIT 2001-2004

ZipIT 2004-2007

IV Three-level analysis for shared understanding of information systems development

IndehelaContext 2004-2007

China-Finland eHealth Partnership 2007-2008

V Activity-Driven Information Analysis - Designing Personal Ubiquitous Health and Wellbeing Systems

MyWellbeing 2008-2010

SOLEA 2008-2011

I Researching Activity-Driven Approach for Information Systems Development

Figure 33. Research papers and their relations to the projects Paper I gathers all the projects and cases involved in the development process. The other papers focus on projects in which I was involved. Paper II was written in the early phase of the process of the development of the methodology. The goal of the PlugIT project was to develop integration methods. Ideas of integration and component-based systems played a role in the first attempts to develop a method for gathering the requirements for systems. Furthermore, the cases in the PlugIT project also influenced the process. Papers II and III are heavily influenced by pilot case of the home care. Paper II rather relates to the goals and Paper III the results of the home care case study. One of the lessons here was that imagining a case study is different from implementing it. Anyhow, having a goal helps realization. The ZipIT project and eight cases that were studied during the project were very important influences on the contents of the ADISD methodology. Paper IV was written after the ZipIT project and it introduces the model and methodology in the form it still has to this day. 82

Dissertations in Forestry and Natural Sciences No 153

Summary of Papers

After the ZipIT project I had the opportunity to be involved in the MyWellbeing project. This project and its different starting point gave us the chance to stretch and apply the ADISD methodology. Paper V describes this applicability of the methodology. After that, I wanted to take a look backwards. Paper VI is a reflection on the steps I was involved in. The ZipIT project is the main presence because the questionnaire was carried out mainly for the research group of the ZipIT project.

5.2

RELATIONSHIPS AND ROLES OF THE RESEARCH PAPERS

The research papers show the honest history of the process. Paper I introduces this long run, with its different milestones. The goal was clear at the beginning, as Paper II shows. The goal was a method for a traceable path from work activity to software components. However, the whole path was not covered, as the first steps needed more attention. The benefits of Activity Theory and the ActAD framework (Korpela et al., 2004) were tested within interviews and workshops, as Paper III shows. After the first few experiences of the application of Activity Theory, developing the model and methodology seemed quite clear and easy to guide. When the group went deeper into the development of the model and methodology, it saw the complexity that was involved. The draft of the model was extended before the central cases started. After several cases the group published a guide for the planned changes (in Finnish) and the ADISD model was seen as a tool for shared understanding during the ISD process (Paper IV). Paper V is about the application of the methodology to the design of an individual health information system. The ADISD methodology was used to create a model for analyzing information from the Activity-Driven point of view. Paper VI is a summary, as is Paper I, but Paper VI also involves reflection. Some of the papers present action research, some others reflection and all have some theory and lessons learned. All in all, I see the Dissertations in Forestry and Natural Sciences No 153

83

Pentikäinen M.: Co-Development of Work and Information Systems

papers as having their own roles and relationships to each other (Figure 34). VI REFLECTION: How to Co-develop Work Activities and Information Systems: Lessons Learnt from Developing the ADISD approach

II GOALS: From information systems requirements to software components - home care case

III ACTION: Gathering, structuring and describing information needs in home care

IV CONSTRUCTION: Three-level analysis for shared understanding of information systems development

V APPLICATION: Activity-Driven Information Analysis

I PROCESS: Researching Activity-Driven Approach for Information Systems Development

Figure 34. Research papers and their roles (capital letters) in this study Paper I presents the process. Paper II introduces the goals. Paper III is a report of the action. Paper IV shows the final construction of the methodology. An application of the methodology is introduced in Paper V. Reflection is performed and written down for Paper VI.

5.3

RESEARCHING AN ACTIVITY-DRIVEN APPROACH TO INFORMATION SYSTEMS DEVELOPMENT

First of all, Paper I introduces the Activity-Driven approach and how it is used. The contributions of the paper are a historical view and a collection of experiences of using the Activity-Driven approach. Paper I is a walk-through from the early steps of the ActAD framework to the Activity-Driven approach, and the methodology and applications on the way. 84

Dissertations in Forestry and Natural Sciences No 153

Summary of Papers

Paper I covers a timeline of over ten years from 1998 to 2011. The paper gives a big picture of the development of the ADISD methodology. This offers an insight into the projects, cases, and experiences that influenced the methodology. A definition of the Activity-Driven approach for ISD is given and its benefits are analyzed, especially in the context of healthcare. Moreover, some future directions are outlined in Paper I. The methods used for this paper were a literature review and the research group’s own experiences. The contribution of the paper is to gather projects and cases around the process of developing and using the ADISD methodology.

5.4

FROM INFORMATION SYSTEMS REQUIREMENTS TO SOFTWARE COMPONENTS - HOME CARE CASE

Paper II presents the reasons for constructing the model or method for requirements specification. This paper focuses on striving towards and defining the method for component-based requirements (integration needs) with the Activity-Driven approach. The starting points and initial assumptions were that requirement specification works as a tool for communication during the development process, and component-based systems should be the goal. Communication between the developers and end users is seen as being the key to the success of ISD (Coughlan, & Macredie, 2002). The principles of the components (e.g., reusable software pieces, documentation, and interface definition) are seen as factors that lead towards a maintainable, understandable, and manageable software process. The paper presents the existing methods selected for upgrading. At the center is the Activity Analysis and Development framework (ActAD) because of the possibilities it offers for modeling work activities and basis in Activity Theory (Korpela et al., 2000; Korpela et al., 2002). From the beginning of the development of the methodology the group followed the principle that because information systems should support the work, the method used for the development of information Dissertations in Forestry and Natural Sciences No 153

85

Pentikäinen M.: Co-Development of Work and Information Systems

systems should take account of work activities. ActAD was selected as the main method. This paper explains the ActAD framework in depth. Other selected methods are “software-oriented” use case design (floated by Jacobson, 1992) and work-oriented eventdriven use cases (Robertson & Robertson, 1999). “Softwareoriented” use cases can be seen as one of the most common methods in requirements engineering. The use cases are generally described by means of the Unified Modeling Language (UML), and their roles involve the interaction between software systems and the end user. Event-driven use cases are derived from understanding the work and the goals of the product. The above methods are compared to ActAD by finding similarities and additions. The ambitious outcome for the development of the method is to identify and describe the method as a path from activity and work analysis to use cases and up to software components. The paper justifies the importance of component-based software engineering (e.g., Herzum & Sims, 2000). “Componentbased” is a term used to mean quite similar things than idea already exists and was called “modularity” (Meyer, 1988). Component-based means principles for pieces of software which have high cohesion and low coupling, are reusable, have welldefined interfaces, and could have different levels of granularity. Identifying software components helps to manage large and complicated systems and to see the need for integration. Moreover, the paper has an example. Home care was taken as a context for testing the first steps of the method path. This can be seen as a feasibility study for the method and for home care. The method was given a preliminary test, and the home care context was briefly explored. Paper II introduces the needs for and benefits of a method covering the path from work activities to software components. However, the description of the path itself as a whole traceability chain is not clear. The attempt to cover this chain was an intention but was not in focus, because the group focused on the first steps of the path. However, Toivanen et al. (2004b) show the ideology of a traceability chain from activities to software components and 86

Dissertations in Forestry and Natural Sciences No 153

Summary of Papers

the home care case is documented as an example of a chain in Toivanen et al. (2004a). The chain consists of elements such as an activity network, activity descriptions, workflows, actions, use cases, and use case descriptions (Mykkänen et al., 2006).

5.5

GATHERING, STRUCTURING, AND DESCRIBING INFORMATION NEEDS IN HOME CARE: A METHOD FOR REQUIREMENTS EXPLORATION IN A "GRAY AREA"

Paper III explains the action research conducted in the home care context. Home care in Finland is a multi-professional and multiorganizational activity. While different professionals help and provide care for their customers in their homes, questions arise such as who they are, what information they need, and where they can get the information they need. The basics of the requirements exploration method involved applying the ActAD framework (Korpela et al., 2000). The paper presents the influences and usefulness of the framework used within the home care activity exploration process. The ActAD framework was used, for example, to select interviewees, to produce interview themes, and to draw pictures to check the results in a workshop, aka “gathering, structuring, and describing”. This paper has traces on the method levels or steps that should be taken towards defining the software requirements and architecture from domain modeling. Home care-specific results are from different levels of viewpoints, e.g., activity networks and organizations, service provision activity in detail, and the need for information by computer-based means.

5.6

THREE-LEVEL ANALYSIS FOR SHARED UNDERSTANDING OF INFORMATION SYSTEMS DEVELOPMENT

Paper IV presents the Activity-Driven ISD methodology in the form it took after the experiences of the ZipIT project and eight Dissertations in Forestry and Natural Sciences No 153

87

Pentikäinen M.: Co-Development of Work and Information Systems

pilot cases within the project (explained in depth in the tables in Paper I). At the same time as the research group had several case studies of action, it had the reflection and construction of the methodology in mind. A detailed discussion about the theoretical basis of the methodology took place and was published by Mursu et al. (2007). The starting points for this methodology are the same as those set down in the so-called “goal” Paper II. Striving for a collection of methods turned into the construction of an applicable methodology. Theoretical approaches on socio-technical phenomena, such as Activity Theory, seen as a key to a shared understanding. Moreover, dividing the model into different levels and phases made it useful. The materials for the construction of the methodology were eight pilot cases and reflection on the cases and the various drafts of the methodology that now existed. The paper introduces the constitution of the research group and acts as a reminder of the healthcare context. The paper acts as a summary of a guide printed in 2007 in Finnish as a result of the ZipIT project (Toivanen et al., 2007). The levels of the Activity-Driven ISD methodology are in focus within this paper. Levels are named as following: 1 Network of activities & Information landscape; 2 Work activity & Information system; and 3 Actions & Information tools. Additionally, the paper is an attempt to see the methodology in a more general way and suggest its applicability to new contexts. As summarized, the methodology, as presented in this paper, aims to narrow the gap between work and IS development and to create a shared understanding of the development as a whole.

5.7

ACTIVITY-DRIVEN INFORMATION ANALYSIS – DESIGNING PERSONAL UBIQUITOUS HEALTH AND WELLBEING SYSTEMS

Paper V concentrates on analyzing information. This paper explains how to apply the Activity-Driven approach to information analysis. The Activity-Driven ISD methodology takes 88

Dissertations in Forestry and Natural Sciences No 153

Summary of Papers

both activities and information into account. Now the focus was on information. The paper is based on the MyWellbeing project, the goal of which was a personal “Coper”. During the project the basis for the Coper design was made and documented (Tuomainen et al., 2010b). What kind of concept would be useful for the design of personal ubiquitous health and wellbeing systems was the one of the questions answered by the multiprofessional research group (Pärjäin, 2010). The project had two main focuses: families having a baby and retiring consumers. My interest was in the first of these. For this case the group identified health- and wellbeing-related information management activities and needs in families (Palmen et al., 2010). Both for identifying family activities and creating the information analysis, the group applied Activity Theory and especially the ADISD methodology. Paper V presents the personcentered information analysis, the reference model for the personcentered information analysis, and the framework for information architecture design derived from the ADISD methodology. These applications with new challenges, of activities centered on the individual (rather than a focus on work activities), and of information analysis that would lead toward a decision about architecture decision (rather than a focus on activities), held out promise for us of the applicability and flexibility of the ADISD methodology.

5.8

HOW TO CO-DEVELOP WORK ACTIVITIES AND INFORMATION SYSTEMS - LESSONS LEARNT FROM DEVELOPING THE ADISD APPROACH

The role of Paper VI is to rise above the methodology development process and draw a picture of the meta level. Reflection in the paper is based on a questionnaire study conducted by the author among the members of the research group of the ZipIT project. The first theme was to explore what Dissertations in Forestry and Natural Sciences No 153

89

Pentikäinen M.: Co-Development of Work and Information Systems

the researchers think about the project and the results in hindsight. The second theme was to find out if the developers of the methodology have used it after the project and if so, how. Finally, the developers’ views on the need for and future of the ADISD methodology were gathered. This reflection Paper shows through the findings of the questionnaire study that there is a need for the guidelines for the co-development of work and IS. The respondents regarded that the ADISD methodology is quite good but needs some further development to be more useful.

5.9

ABOUT THE PAPERS

At the same time as the papers show the process of the study, they also have relationships with each other. Paper I and Paper VI show the process as a whole from different points of view. Papers II-V are focused on particular phases of the process. Table 7 summarizes the papers contributions and relations to the sub-questions presented on Table 1 (page 15). In addition to the contributions of the papers listed in Table 7, their total contributions are analyzed together in Chapter 6

90

Dissertations in Forestry and Natural Sciences No 153

Summary of Papers

Table 7: Contributions of the papers to the sub-questions RQ# SQ1.

SQ2.

Questions What is the current state of the art in both ISD and Work Development (methodologies)? How can the codevelopment of IS and work be supported?

P# I

II

I II

III SQ3.

SQ4.

How could Activity Theory be utilized to provide a coherent methodology for the codevelopment of IS and work?

I

How did the development of the ADISD methodology take place?

I

II III IV V

VI

Outcome & contribution Socio-technical approaches concerned and earlier application of Activity Theory (DWR, HCI, CSCW) Starts from method “map” of Iivari & Lyytinen Definition of Activity-Driven approach and characteristics of use of it. Thinking about methods for work development and software engineering (ActAD, Use Cases from UML, and EventDriven Use cases from Robertson & Robertson). Compares ActAD and EventDriven use cases. Looking for systematic model for interviews Presentation of ADISD methodology and projects where it was developed and applied. Applying ActAD for studying home care Using Activity Theory to gather, structure, and describe information needs. Description of the ADISD methodology Experience of applying ADISD for individual-centric information analysis Experiences so far and future directions for the AD approach Lessons learned

Dissertations in Forestry and Natural Sciences No 153

91

Pentikäinen M.: Co-Development of Work and Information Systems

92

Dissertations in Forestry and Natural Sciences No 153

6 Analyzing the Development of the Methodology as an Activity Why are ISD methods developed? What is the reason for the existence of development? The motivation for development activity could be found, e.g., from supported activities. There have been good attempts at developing tools and models, but especially methodologies that take care of work improvement are fewer in number. Some of the models are never adopted for practice because they are too theoretical. Some of the methods used in practice are not even documented and that is why they are not used broadly. The attempt in the case studied in this thesis was to develop a practically useful methodology which has a theoretical basis. In this chapter a metamodel of the development of an ISD methodology is presented by analyzing the development of the ADISD methodology as an activity within an activity network. As explained in Chapter 2, this reflection is performed in order to get ideas about the future of the methodology but also to observe general points that relate to methodology development activity. This chapter is organized as follows: Section 6.1 The activity network of the development of the ADISD methodology starts the analysis by looking at the case (Chapter 4) as a process; 6.2 How to analyze the development process as an activity presents the generic framework (from Section 3.2.3) with which I will analyze the case; 6.3 Analyzing the phases of the development of ADISD goes deeper Dissertations in Forestry and Natural Sciences No 153

93

Pentikäinen M.: Co-Development of Work and Information Systems

into analyzing each of the phases of the case (Chapter 4); and 6.4 Proposals for the development of models and methodologies takes a look at the future and makes proposals derived from the analysis.

6.1

THE ACTIVITY NETWORK OF THE DEVELOPMENT OF THE ADISD METHODOLOGY

The ADISD methodology was developed over a period of several years in different projects, as introduced in Chapter 4. The experiences and results are documented in the papers presented in Chapter 5. The development process was actually a network and many unplanned issues and unplanned changes affected the development of the methodology. I will start this analysis from the network level (Korpela et al., 2000). In Figure 35 the projects (Chapter 4, Paper I) are linked to the phases of the development process. The Case: Developing the methodology

Preliminary research

Action research

Phase 0

Phase 1

Phase 2

PlugIT 2001-2004

ZipIT 2004-2007

Pre

Indehelamethods 1998-2001

Action research

IndehelaContext 2004-2007

Action research

Future

Phase 3

Phase 4

MyWellbeing ChinaFinland 2008-2010 eHealth Partnership SOLEA 2007-2008 2008-2011 d h IndehelaEducation & Exchange 2009-2014

Teaching

Business applications ISD4D 2011-2015

Figure 35. Essential phases of the development process with projects I see this development process as having phases 0-3 and the future. I will analyze each of the phases separately in Section 6.3. 94

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

6.2

HOW CAN THE DEVELOPMENT PROCESS BE ANALYZED AS AN ACTIVITY?

When analyzing an activity, the starting point is figuring out the motive. What is the object of the activity? Transforming the object to the intended outcome is the collective motivation for an activity. More information on the motive is found from looking into the activities that were supported. Activities waiting for an outcome are the reason for the existence of the focus activity (Korpela et al., 2000). According to Activity Theory, there are subjects or actors who work upon the object. Individual actions are mediated by their mental (education, experience) and physical (available information, software) tools and means (Leontjev, 1977). When there are many actors it is meaningful to see their means of coordination and communication, e.g., professional secrecy, rules for meetings, or computerized communication tools. When analyzing one activity with the ActAD framework (introduced in Chapter 3.2.3 and Paper II) we identify the elements of the activity, such as object, outcome, process, participants, means of work, means of communication and coordination, and collective actor. We also recognize the supporting activities (which need to exist before the central activity) and the supported activities (which are the reason for existence of the central activity). Figure 36 presents the model for analyzing the activity of developing the ADISD methodology in a general way.

Dissertations in Forestry and Natural Sciences No 153

95

Pentikäinen M.: Co-Development of Work and Information Systems

Activity = Development of the ADISD methodology Collective actor: research group

Need for for Need developNeed for Need for developm ment developm developm ent - pilot ent ent cases

case

Participants

Object development cases worked with actitivitydriven methods

User-interface e design Socio-technical approach

Use ccase design (UML)

Component-based C development

Experiences, literature

Means of communication and coordination: projects’ settings and plans, e-mail, reports, timetables

Health H Software Healllth th Software Health Healltlt lthh IS IS research research interest providers providers interest informatics informaticcs

Modeling, testing

Previous research & projects

Knowledge

Meetings Meetings

Interviews, Workshops

Education & experience

plans

ActAD framework

Projects and managem ent

Outcome Guidelines of the ActivityDriven development (ADISD model)

Process: constructing the model

Knowledge

ActivityActivityActivityDriven Driven Driven IS IS ISD developmen d practices developmen t t

Means of workk ADISD model Teaching ADISD

ADISD model ADISD model

Researching (& developing) ADISD

ADISD model Version 0.x

Supporting activities

Supported activities

Figure 36. Analyzing the development of the ADISD methodology as an activity The historical development of the activity over time can be divided into phases by analyzing how the mode of the activity has changed. Each phase is also characterized by a slightly different motive (i.e., what was the outcome strived for during that phase) and object (i.e., the initial “raw material” that was to be transformed into the outcome). The outcome of each phase becomes the starting point of the next one. Figure 36 presents this as part of the generic model and Figure 37 names the specific outcomes-inputs mediating between the phases. According to the theory, any contradiction within the activity or with its outer world could be the driving force from one mode and phase to the next one. In practice, the driving force always appeared to be the

96

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

contradiction between the outcome of the previous phase and the new challenges which created the motive for the next one. The motive, since Phase 1, was definitely related to supporting the Activity-Driven analysis and development (first called worksupported). In the different phases of the development process (03) there was a change in the emphasis of the activity elements. During the years 2001-2013 there were many participants in this process, each with their own educational backgrounds and experiences. The participants formed multiprofessional groups over this time. There was interest from the IS research and software points of view’, a health informatics professional and even health providers. The participants’ backgrounds affected their selection of the means of work, and the means of work affected the outcome. There were also limitations and directions from project resources and plans. We should not underestimate the pilot cases. The methodology was created with the cases and for the cases. The context of healthcare had an influence on the outcome. Later the group also noticed that it was reasonable to teach, research, and further develop the methodology. Next I will perform a more detailed analysis of the development phases with the ActAD framework.

6.3

ANALYZING THE PHASES OF THE DEVELOPMENT OF ADISD

There have been projects around Activity Analysis and Development (ActAD) since Korpela’s dissertation (1994). I regard the early projects as Phase 0 of the methodology development studied in this research (Figure 37). Phase 1 was centered on the PlugIT project, the main input of which was the ActAD framework from previous research (Korpela et al., 2000; 2002, Mursu, 2002). The outcome of Phase 1 and input to Phase 2 was activity-driven three-level experiences (Korpela et al., 2004; Toivanen, 2004, Toivanen et al., 2004b, Papers II, III). Phase 2 was particularly significant for the construction of the methodology. The main result of the ZipIT project was a documented model and methodology (Mursu et al., 2007; Dissertations in Forestry and Natural Sciences No 153

97

Pentikäinen M.: Co-Development of Work and Information Systems

Toivanen et al., 2007; Paper IV). After ZipIT, applying the methodology was performed in Phase 3, but no major changes were made to the methodology after Phase 2 (Papers I & VI).

Phase 0

Phase 1

Phase 2

Develop ment projects

Phase 3

Future Phase

Applications

Teaching Feedback

Previous research

ActAD framework

PlugIT

Three-level method experiments

ZipIT

MyWellbeing

ADISD model

Research Business

Application

Future

Upgrades

Applications

Figure 37. Activity network of the essential phases of development The future phase remains to be seen but some ideas can be proposed in the light of the analysis and findings from the questionnaire shown in Paper VI.

6.3.1

Phase 0: Producing the ActAD framework The origins of the activity-driven approach lay in Korpela’s doctoral research (1994). His aim was to study the practice of ISD in Nigerian software companies. To that end he needed a theoretically sound lens for analyzing work practice, and identified Engeström’s activity model and DWR methodology (1987) as the most suitable one. During his doctoral research he was the only actor that was acting on the object of DWR to modify it to better suit his needs for an analytical framework for studying work practice (Figure 38). The outcome presented in his

98

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

thesis was not yet a practicable methodology but rather theoretical argumentation for, and elements of, a model. Activity = Development of ActAD Collective actor: a researcher or research group

Projects and management

plans

Means of communication and coordination: projects’ settings and plans, airmail, e-mail, reports, timetables

Meetings Meetings

Knowledge

Object NeedNeed for of Needfor for Need developme d developme development developme nt ntntin projects

case

Development projects in developing countries

Modeling, testing

Interviews, Workshops

Literature Previous research

Socio-technical approach

Health Health providers providers

IS IS researcher researcher DWR model

Education & experience

Knowledge Activity

Participants

A Activity-Driven Activity-Driven Analysis and IS IS d Development development development

Means of work

Outcome Actitivy Analysis and Development framework ActAD

ActAD

ActAD ActAD

Teaching ActAD

Research projects

Process: formulating ActAD ActAD version

Supporting activities

Supported activities

Figure 38. The activity network of Phase 0 Phase 0 was completed in the first of the INDEHELA (Informatics Development for Health in Africa) projects, the research project INDEHELA-Methods, which was funded by the Academy of Finland, Development Research. Two new doctoral students, Anja Mursu in Finland and H. Abimbola Soriyan in Nigeria, now participated in the methodology development activity. The collective actor of the activity thus became a closely working research group (Figure 38). It is notable that the field work of Phase 0 took place completely in Nigeria. Issues that are particularly relevant in developing countries were thus addressed in the development of the methodology together with local actors; for example, the role of communities (Korpela et al., 1996) was Dissertations in Forestry and Natural Sciences No 153

99

Pentikäinen M.: Co-Development of Work and Information Systems

elaborated on with local healthcare professionals and the sustainability of information systems with local IS professionals (Mursu, 2002). The process in Phase 0 consisted mainly of acts of theoretical analysis by the researchers, supported by interviews and workshops with local professionals. The means of coordination and collaboration were meetings while the actors were in the same place, and then airmail and (later) e-mail correspondence. The initial motive for Phase 0 was thus the need for a proper analytical model for research, and the mode moved from desktop research into experimenting with preliminary ideas together with practitioners.

6.3.2

Phase 1: Towards Work-centered ISD When the PlugIT project started in 2001 the ActAD framework was well documented. The PlugIT project’s main target was to produce methods for the integration of information systems. This was the reason why some means of work were selected. The background of the participants also had an effect. In the research group there was strong knowledge about software engineering but also about healthcare and health informatics (Figure 39). The group searched through the literature to find methods that emphasized the integration of information systems. That is why the group had component-based ideas as the centre of the attention. At the same time the group had knowledge about patient record systems which were not useful. The idea that information systems should support work played a remarkable role in the selection of Activity Theory and ActAD framework as the means of work. The development of information systems has to enable work processes to be developed. ActAD was selected because it was available, clear, and easy to adopt, and the framework worked as a modeling tool. It was able to describe activity networks in order to gain an understanding of the big picture. It worked as a checklist to identify and analyze 100

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

activity. It took into account people but also the tools and other means that supported people’s work. Moreover, ActAD had a theoretical background based on Activity Theory and that made it a coherent and trustworthy approach. Activity = Development of Work-centered ISD Collective actor: research group Projects and management

Participants

Object Need Need forfor Need for developmen development developmen t cases - pilot t

case

Apply the selected methods in A. Home care B. Maternity care C. Data administration

Use case design (UML)

Component-based development

Experiences, literature

Modeling, testing

Health Health Health IS IS research researchSoftware Software Health provid ders informaticcsproviders interest interestinformatics Interviews, Workshops

Previous research

Meetings Meetings

Means of communication and coordination: projects’ settings and plans, e-mail, reports, timetables

Knowledge

ActAD framework

Education & experience

plans

Knowledge

Means of work

AD methods Teaching ActAD

Outcome Documented ISD methods for integration, interoperability indentification & information needs analysis

Activity-Driven A WorkActivity-Driven IS centered ISD IS development d development

experiences

AD methods

Researching & developing Activitydriven methods

Process: apply and document Activity-driven method description

Supporting activities

Supported activities

Figure 39. The activity network around the PlugIT project The means of work here are focused on methods. ActAD and component-based approach were in focus that the group used to find a path from activity to software requirements specifications. That was why use case design was also considered. Something that is noteworthy is that the research group had various meetings. It had case-specific meetings, meetings to discuss Activity Theory, meetings about healthcare organization in Finland, meetings about the method to apply to the case, meetings to compare experiences, and meetings of project Dissertations in Forestry and Natural Sciences No 153

101

Pentikäinen M.: Co-Development of Work and Information Systems

managers of the pilot cases. In the last mentioned meetings there were representatives of all the PlugIT project cases (13 cases, 7 persons). The objective was to apply selected methods in order to test if they are useful and then to use them as documented guidelines. The final outcomes were three case methods and the need for further development.

6.3.3

Phase 2: Development of the ADISD Methodology While the objective of the PlugIT project was methods for integration, the objective of the ZipIT project was to develop and document an activity-driven model for ISD. The emphasis was more on activity-drivenness. Component-based development was not the focus any more. The group had already proved that it is possible to identify a traceable path from work activity to use cases. Now the group focused on documenting the experiences and guidelines for how work and IS could be co-developed. The consequences of new participants and pilot cases becoming involved also approach of user interface design became as the centre of the attention (Figure 40). The pilot cases from the healthcare context gave a challenging environment. I think that willingness to cooperate with healthcare professionals and the idea that the group should create guidelines for helping them had an impact on the final outcome of the model and methodology. At the same time the group had software companies involved and thought that the final methodology should work as a tool for communications and shared understanding for both sides (software producers and healthcare professionals).

102

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

Activity = Development of the ADISD model Projects and management Education & experience

Collective actor: research group

plans

Means of communication and coordination: projects’ settings and plans, e-mail, reports, timetables TEAMWORK TEAM MWORK

Knowledge

Participants

Object Need Need for for Need for Need for development developme developme developme - nt pilot cases ntnt

case

9 development cases worked with actitivitydriven methods

Use case design (UML)

User-interface design

Comments

Socio-technical approach

Scientific conferences & discussion

Modeling, testing

Experience, literature

Interviews, Workshops

Previous research

ActAD framework

Software Sof Softwarrre Health Health Health IS IS research research interest Health interestinformatics providers informaticcsproviders

Activity-

Knowledge

Activity-Driven A Activity-Driven Driven IS IS ISD d development development

Means of work

ADISD model

Outcome Guidelines of the ActivityDriven development (ADISD model)

Teaching ADISD A ADISD model ADISD model

Researching & developing ADISD

Process: constructing the model ADISD model Version 0.x

Supporting activities

Supported activities

Figure 40. The activity network around the ZipIT project The group now had more knowledge gained from the literature. It had in mind Information Systems Research, Health Informatics, Participatory Methods, healthcare services, Software Engineering, and a deeper understanding of Activity Theory. This led it to an understanding that it had much in common with others that are working on a socio-technical approach. Input also came from scientific forums when the group presented versions of the methodology and experiences at international conferences. If meetings were significant in the previous phase, they still were. Moreover, the teamwork of the research group showed its worth in reflections and the further development of the ADISD methodology, which developed into many variations (Chapter 4.4.3).

Dissertations in Forestry and Natural Sciences No 153

103

Pentikäinen M.: Co-Development of Work and Information Systems

The significant outcomes were a book of guidelines (Toivanen et al., 2007) and the methodology in its final from (Paper IV). 6.3.4

Phase 3: Applications of the ADISD Methodology After the ZipIT project the researchers used the ADISD methodology in their own way in their own work. One of the applications is presented in this thesis and it was the MyWellbeing project (2008-2010). The ADISD methodology was applied to personal activity analysis and as a starting point to analyze information on personal health and wellbeing (Chapter 4.5; Paper V). More applications in Phase 3 are shown in Paper I and analyzed in Paper VI. Other projects were the China-Finland eHealth Partnership (2007-2008), Indehela-Context & Education (2004-2011), SOLEA (2008-2011) and, additionally, ISD4D (20112015) (not mentioned yet in Paper I, Moreover, the ADISD http://www.uef.fi/indehela/isd4d). methodology was taught and used in business in this phase. As a starting point for Phase 3 we can see that the ADISD methodology exists and is documented but not widely used. Even after Phase 3 the methodology is not widely used. However, there are experiences and evaluations which will be helpful in the further development of the methodology. This phase was so multilayered and multiform that the participants worked separately, but still some remained in communication with each other in order to apply or further the development of the ADISD model and methodology. I call them a group of researchers with the same interest. Anyhow, the documenting of the experiences and sharing the discussions were not so systematic any more. This phase was not one project and was no longer goaloriented, but each actor had their own goals. I still consider that all those outcomes could be used to support teaching, researching, further development, and, very importantly, Activity-Driven ISD practice (Figure 41).

104

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

Activity = Applying the ADISD methodology Projects and management

Targets, scope

Collective actor: groups of researchers with the same interests Means of communication and coordination: timetables, research papers

Experience

Software Software Health Health h developers developers developers informatics ers informatics develope IS IS researchers researchers

Applying ADISD Modeling, testing M Component-based C development p ntt Applying ADISD UML o ng, testing Modeli face design erseerUser-inter Socio-technical approach

Previous research & projects

Knowledge

tAD framework ActAD Modeling, testing Workshops e vie Interviews,

Education & experience

case

Development cases

Knowledge

Means of work

A ADISD model

Object Real cases, for Need need Need forfor Need for developm development developm developm ent ent ent

Participants

Outcome Understanding, share undertanding, modeling, gathering information, documenting of the cases

ActivityActivityActivityDriven Driven Driven ISISISD d developmen developmen tt

Teaching ADISD

ADISD model

ADISD model

Researching & developing ADISD

Process: applying the model Interpretations of ADISD model

Supporting activities

Supported activities

Figure 41. Phase 3 as an activity

6.3.5

Phase 4: Future Development of the Methodology If we look at possible inputs for the future phase of the development, we could recognize feedback and upgrades from appliers and students, various application experiences, and needs for further development. What would be the desired outcome of the future phase? I suppose it would be more and wider use of the methodology, easier and quicker use of the methodology, and well-documented guidelines. As long as we have appliers we have a future for the methodology and the development of the methodology. But we need somebody who has an interest in gathering, documenting, and spreading information. One possible future can be as part of a bundle of methodologies (Korpela et al., 2013). Dissertations in Forestry and Natural Sciences No 153

105

Pentikäinen M.: Co-Development of Work and Information Systems

The group’s future goal is presented in Figure 42. Action plans and the evaluation that is needed are analyzed, as is the current state. The analysis is performed on three levels, as the ADISD model advises.

106

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

Figure 42. Analyzing the development of ADISD with the ADISD model. Dissertations in Forestry and Natural Sciences No 153

107

Pentikäinen M.: Co-Development of Work and Information Systems

6.4

PROPOSALS FOR THE DEVELOPMENT OF THE MODELS AND METHODOLOGIES

“Systems thinking in the IS discipline often focuses on either the technical system or the social system and its context, but rarely focuses on both in combination. … Orlikowski and Iacono (2001) say that too little of IS research takes a system view that explicitly recognizes the overlap between the social and technical aspects of a system … Both research and practice might benefit from more systems thinking that spans the technical and the social.” (Alter, 2004b, bolding by Pentikäinen) The highest-level research question of this thesis is: MRQ: How can methodologies for the co-development of work and information systems be developed? There is no single answer but some points are analyzed here. It seems that often the development of the methodology is struggling with the requirements, between being comprehensive and simple enough. Why do some models or methods become useful? How are methods picked up in practice? If you have to climb up on the roof, you certainly look around for a ladder. If there is no ladder you can choose to build one, or use an ad hoc solution to fix the problem (e.g., pile up big pieces of furniture on top of one another). Two obvious reasons why a method is used or not are its availability or its suitability for the problem at hand. Moreover, the method should be easy and quick to use. Big pieces of furniture are used to reach the roof if the ladder is far away and there is a need to hurry. Essential lessons learned are listed as proposals for the development processes of models and methodologies: x theoretical background to having a systematic approach x practical knowledge because of its usefulness x multiprofessional participants so as to have enough insights x teamwork to share understanding and test the ideas x various cases and contexts for empirical testing

108

Dissertations in Forestry and Natural Sciences No 153

Analyzing the Development of the Methodology as an Activity

x x x x

x

documenting the observations and experiences for later use and reflection discussions, reflections, and comparing of the experiences to improve the methodology an open mind to see alternative ways don’t be afraid of unfamiliar disciplines (Activity Theory was proposed to study psychological phenomena at first) think about who the appliers are and how they put your methodology into use.

Ultimately, you should be prepared for the fact that your methodology could be misunderstood and misused or not used at all.

Dissertations in Forestry and Natural Sciences No 153

109

Pentikäinen M.: Co-Development of Work and Information Systems

110

Dissertations in Forestry and Natural Sciences No 153

7 Discussion and Conclusion “The qualitative researcher seeks not truth and morality, but rather understanding”- Eija Karsten On the ground level, the motivation for this study was to help end users work, with and without computers. This would happen with a new development methodology, the ADISD methodology. The methodology is for end users, managers, and software producers to get the big picture about the needs and context of the development. To create an extensive guide, we need to document phases as a step-by-step process. The guide can take the form of methods, models, or instructions. When striving for a perfect guide, there is a contradiction that needs to be solved, between being versatile and simple enough. This thesis is about developing the methodology for ISD. The attempt made here is for the methodology to work as a guiding tool. The development process of the methodology is shaped by its history and context. I, as the writer of this thesis, have moved from being a novice towards expertise. Within that process the group’s goal, the ISD methodology, has moved from a strict step-by-step method towards an applicable model and methodology that includes compromises. At the same time as diversity has increased, the group has tried to attain simplicity. This process is described in Chapter 4. On a higher level, the motivation was to analyze and document the phases of the development process of the ADISD methodology in order to see possibilities of further development of ADISD and methodology development in general. This analysis is performed in Chapter 6. This chapter aims to discuss and summarize the findings on all the levels. The chapter is organized as follows: Section 7.1 Summary of the findings looks back to the research question and Dissertations in Forestry and Natural Sciences No 153

111

Pentikäinen M.: Co-Development of Work and Information Systems

summarizes the main findings to the sub-questions; 7.2 Contributions of the thesis and 7.3 Strengths and limitations of the study assess the study as a whole; and 7.4 Future work opens the door to future research.

7.1

SUMMARY OF THE FINDINGS

The sub-questions introduced in Section 1.4 are rich and the expected outcomes even richer. How did the findings actually reflect the questions? And what do they bring to the Main Research Question: How can methodologies for the co-development of work and information systems be developed? Each of the subquestions should be seen through the main research question. SQ1: What is the current state of the art both in ISD and Work Development (methodologies)? The findings in Chapter 3 are that there are methods for the development of IS and work but, for co-development, only to a small extent. Work development and IS development are usually seen as dissimilar and separate processes. This acts as motivation for studying and developing a new methodology. The same observations were found in different fields: more communication is needed, end users should be taken into account more, and practical, easy, and quick methods are needed. Moreover, there are methods and methodologies, but they are rarely used. All in all, the findings confirm the initial assumption that there indeed is a gap to be zipped up. SQ2: How can the co-development of IS and work be supported? In Chapter 3 I explore the fields to find a theoretical basis and justification for the construction of a new methodology. In Chapter 4 the projects are introduced as examples of supporting the co-development of IS and work. During the projects and through the dialogue with the literature the requirements for the proper methodology were sharpened (Section 3.1.3). We need an IS development methodology (guiding tool) that supports the development of the work, and should be useful in communication between end users and IS producers. The ADISD 112

Dissertations in Forestry and Natural Sciences No 153

Discussion and Conclusion

group chose Activity Theory as the basis of the methodology to ensure that it fulfilled the requirements of being work-oriented, coherent, theoretical, comprehensive, and, above all, understandable. SQ3: How could Activity Theory be utilized to provide a coherent methodology for the co-development of IS and work? This question is covered in Chapters 3, 4, and 5, first theoretically, then practically. The Papers tell the rest of the story. The case of developing the ADISD methodology is one example and attempt to evidence that it was indeed possible to utilize Activity Theory to support the co-development of work and IS. It was possible to utilize AT for providing a coherent basis for the methodology. Throughout the ADISD development process the group applied Activity Theory. Moreover, the ActAD framework based on Activity Theory was used because of its graphical nature and its way to see many subjects in the same picture. The group needed the framework for utilizing the Activity Theory. The group used this framework as a tool for modeling but also as a checklist to cover Activity Theory’s perspectives. Moreover, while applying Activity Theory in the pilot cases the group saw its usefulness to bring together, into the same picture, both the work and the IS. The experiences showed that using Activity Theory is a very promising choice for creating a coherent methodology for the co-development of IS and work. SQ4: How did the development of the ADISD methodology take place? I analyze the development process of the methodology in Chapter 6. With these experiences, introduced in Chapter 4 and the Papers, I cannot proclaim a universal truth. But I identified things which might affect the process and thus the product of the process. Usually, methodologies do not start from scratch. That is the case in many other development projects nowadays. We might easily think that a new methodology is summarized from existing methodologies, but how? The development process is always a unique process which is under the influence of the participants and their backgrounds, education, and experiences, the means Dissertations in Forestry and Natural Sciences No 153

113

Pentikäinen M.: Co-Development of Work and Information Systems

and tools of work that are on hand, and the means of communication, tools, and skills of the participants. The process is under the influence of the supporting activities (inputs) and the context environment (in this case it was healthcare). The process is directed by its objective. Moreover, the purpose of the methodology affects the development (supported activities). Now let us get to the main RQ: How can methodologies for the codevelopment of work and information systems be developed? This thesis is a story about that. As found out in Chapter 6, it seems to be beneficial to have a multiprofessional, open-minded, and flexible team, an empirical test context, and a clear ambition. The ambition works as a driving force. Table tests only are not a convincing proof of the quality of an artifact. Multiprofessionality is very good when ideas are being discussed and reflected on. In Table 8 the essential findings to each of the sub-questions are summarized.

114

Dissertations in Forestry and Natural Sciences No 153

Discussion and Conclusion

Table 8: Sub-questions, main materials and methods, essential findings, and additional considerations (Ch=Chapter of this thesis, P#=Research Paper # of this thesis, see Section 5.9 for more details of Papers) #

Question (Ch 1)

Main materials & Research method (Ch 2)

SQ1

What is the current state of the art in both ISD and Work Development (methodologies)?

Existing theories and research Literature review

SQ2

SQ3

SQ4

How can the codevelopment of IS and work be supported? How could Activity Theory be utilized to provide a coherent methodology for the codevelopment of IS and work? How did the development of the ADISD methodology take place?

Essential findings Additional considerations

Our objectives are considered important in the fields of SE and IS. There are a couple of promising methods. Because work becomes more information and computer centric, we need more comprehensive methods to analyze and develop. Literature, With a combination of methods existing or a methodology that takes methods both sides into account. Constructive We can analyze, model and design them together. analysis Projects, Using AT as a basis of the methodology made it become cases (in dialogue Activity-Driven. Work with activities are always literature) considered first and IS is seen Action as a part of it. research The long history of using AT in different disciplines makes its use quite systematic but still easy to apply. Process of There were several development development projects and cases (Ch4) as objects of the multiReflection: professional research group. activity The research group’s attempt to analysis develop and apply the ADISD methodology was achieved in years 2001-2013

Dissertations in Forestry and Natural Sciences No 153

Focused on in

Ch 3 PI P II

Ch 3 Ch 4 P II

Ch 3 Ch 4 Ch 5 P I-V

Ch 6 PI P VI

115

Pentikäinen M.: Co-Development of Work and Information Systems

None of the sub-findings can answer the main research question by itself but in the light of the sub-findings and lessons learned I can summarize the main findings. The question “How can methodologies for the codevelopment of work and information systems be developed?” could have been approached from many perspectives and basics. I decided to split it into sub-questions for my own purposes to understand the phenomenon of methodology development and to support the co-development itself. One essential guiding factor of this research was the decision to use Activity Theory in the very beginning. Without this decision it had been very different kind of a study, even if the main question had been the same. I think there can be different ways to develop methodologies. Only one example is analyzed within this thesis. I still propose some general issues I regard as essential in developing a methodology: x The real need for a methodology plays the role of motivation (found from preliminary research (SQ1) and practice). x Considering the existing and promising methods gives insights and perspectives to your methodology-to-be (SQ2) x Testing and strengthening the idea of the methodology provides experience of its usefulness (SQ3). x Having several different cases and a multiprofessional team, which reflected on the cases and the emerging methodology, were major factors in this case where methodology was developed for the co-development of work and IS (SQ4) Beyond the sub-questions I see that when developing a methodology in a multidisciplinary field more research on each of the fields of science should be done. Moreover, defining the concepts explicitly is important.

116

Dissertations in Forestry and Natural Sciences No 153

Discussion and Conclusion

7.2

CONTRIBUTIONS OF THE THESIS

This thesis is multilayered and multiphased (Figure 43). At the case level, the practical outcomes are the methodology and the experiences of applying it. At the meta level, the thesis introduces the historical development of the ADISD methodology, but it also analyzes that. The research methodological contribution lies in this thesis using iterative action research.

Ch 6.4 Ch 6

Reflection: Analyzing the development of the methodology as an activity

Ch 4

Methodology development in general

The Case: Developing the methodology

Preliminary research

Action research

Action research

Action research

Future

Ch 6.3.5 Ch 3 Figure 43. Two-level research process (Figure 3) with the contributions of the essential chapters. The contribution of this thesis is to confirm and remind us of simple and already-known things, but also to share understanding about one case of constructing a methodology. We can see other level contributions as well: What was the historical process of developing the ADISD methodology and moreover, how one can analyze such a process. With this analysis the group can start a further development of ADISD. 7.2.1

Research Methodology Once more we see with this study that plans are there to guide research but we cannot fully control qualitative research even if we wished to. During this study the research groups and I were Dissertations in Forestry and Natural Sciences No 153

117

Pentikäinen M.: Co-Development of Work and Information Systems

just interpreters. We cannot state the objective truth. Notwithstanding, I think we still need the action research type of research. That is the way we get a practical flavor for the research and empirical arrangements. When real world development projects are cases of research it is hard to apply one strict research methodology. This research was not done with only one methodological approach. That made it challenging. However, I think that contributes to the richness of this study. Having features of different disciplines and crossing the lines between them might be confusing but it seems promising in the light of this study. The real application of an action research methodology here was on many levels and I think it can be adapted more as an iterative process in the future, as I see it (Section 2.3). 7.2.2

Theoretical Contribution at Meta Level This study shows that it is valuable to analyze not only the product but also the process, as Design Science proposes (Gregor, 2006). The main distinguishing feature in the product in this case is that the ADISD methodology is developed on the basis of Activity Theory and the ActAD framework. This research shows the potential that is present in those. The main theoretical contribution of this thesis is the increased understanding about the development of the ADISD methodology by documenting that process. Moreover, there are some general points about the development of methodologies. The lessons learned from the development of the methodology are that one methodology cannot cover everything. If we strive for it to be comprehensive we probably lose some of its usability and simplicity. If a methodology is intended for practical use it has to be easy to put into use. To use the ActAD framework for creating a shared understanding of developing activities seems promising, as can be seen in Chapter 6. Moreover, making action plans for progress toward the desired future on different levels is possible with the ADISD methodology (Section 6.3.5). 118

Dissertations in Forestry and Natural Sciences No 153

Discussion and Conclusion

7.2.3

Practical Outcomes at Case Level Findings on the case level can be seen as side products of this thesis. At the level of the case in question, the ADISD model and methodology and its applications were the practical outcomes. I would say that the methodology has many good features for practical use: the co-development of IS and work; the codevelopment of information and processes on different levels; scalable, coherent, works as a checklist, and takes important phases into account. What is still required are a quick guide and CASE tool support (Paper VI). The methodology supports the traceable requirements engineering process and usefulness of IS products. The methodology can be used just to gain a shared understanding of the situation in an organization. It can be used to identify the development needs of work or IS. Above all, it can be used for the co-development of work and IS. Analyzing the process gave me one more insight: the process is there to produce an artifact for the users of the artifact and that means the process is there for the end users, not only for the producers. The process should be transparent for the end users so that they can take a stand on the process and product. While it is important to produce a great and maintainable product, it is also valuable to reflect on the process. Through seeing the phases and elements affecting the process we can make the planned changes in time. This is why we should document the process. This works in practice but also in theoretical discussions. One more practically significant outcome of this case has been the knowledge spread by individuals of the research group after the ZipIT project. The ADISD methodology has been applied and shared but not as much as it could be. Based on the experiences with pilot cases (introduced in Section 4.3.1 and 4.4.2) ADISD seems to be most worthwhile in the first phases of ISD (needs analysis, requirement specifying). A domain area that is large or fuzzy (e.g. home care), too, seems particularly fitting to ADISD.

Dissertations in Forestry and Natural Sciences No 153

119

Pentikäinen M.: Co-Development of Work and Information Systems

7.3

STRENGTHS AND LIMITATIONS OF THE STUDY

It has been a long time and there are a lot of materials associated with this study. It is a strength but at the same time it turns into limitation as well. Many phases happened during the period covered by my inspection, 2001-2013, but many interesting phases happened before too. To have a more extensive view I could have reflected on more materials. Even if I did not analyze all the available information, some might require more test cases, particularly to show the practicality of the ADISD methodology. I view it as a strength that the methodology and the process of its construction are documented here. The limitation is that there is no comparison with other methodologies or systematic evidence of the ADISD methodology’s usefulness. Healthcare as a context was challenging and that was a strength of the process. The group applied the methodology in a context where information is very important, information can be in different formats, and many professionals might need the same information, etc. At the same time the fact that the group applied the methodology almost solely in the context of healthcare, not in different contexts, is a limitation. Another limitation of this study is also that it was an action research type of study. Action research is always tied up with the circumstances, and we cannot assume that a study of that kind can be repeated. But if we see gaining a better understanding as our goal, this kind of study works well. To answer the research problem about co-development of work and IS, the group constructed a methodology which supported this goal. The methodology gives guidance on how to start and how to do modeling, but it is not a self-sufficient methodology in projects. The ADISD methodology supports mainly the first phases of ISD projects, which means that other methods are needed as well, if a project continues to a software project or to the long period of a work development process. Still ADISD could be used to see confluences of these separate developments, and further co-development where it is possible.

120

Dissertations in Forestry and Natural Sciences No 153

Discussion and Conclusion

The theoretical roots of the methodology constructed in the case could have been better documented. The selection of certain theories should be more explicitly justified. The ADISD methodology should be compared to other similar methodologies. However, the ADISD methodology is not in the focus of this thesis, but the process of its development is.

7.4

FUTURE WORK

Like this study, I also see the future work as being on two levels. On the case level, the ADISD methodology should be applied and tested more; experiences should be gathered and reflected on. Probably, the ADISD model and methodology will be developed further and on the theoretical level, models and methods in general should be in focus – what we can do to develop those, how they can be more useful. Some comparing with other promising methods supporting our purposes should be done with the ADISD. Moreover, what are the useful methods to use with ADISD? How one can continue from ADISD to UML? Those are future questions to study more specifically. On the meta level, why are methods and methodologies not used? There are many methods in the ISD field, even sociotechnical ones. It is possible that they are not available, or not so practical after all, or that they are not marketed and “sold” enough. The importance of communication with and participation of end users have been written about since 1970 but still something is missing. If the practice is in such a hurry that developers have no time to take the end users into account, they have the wrong tools and models. How methodologies come into use in practice is one possible future research topic. Moreover, I see that historical analyses such as the one done about ADISD in this thesis should be done about the other promising methodologies, to get more understanding on how they can be developed more practical and become more widely used. As long as there is the belief that things can be done better… One hopes that one day there will be a software implementation Dissertations in Forestry and Natural Sciences No 153

121

Pentikäinen M.: Co-Development of Work and Information Systems

in an organization. The software will be great and include all the new technology. The software will be everything that was stated by the managers in the contract. The software will be integrated with the legacy system of the organization, as promised, and everything will work, including the people. The end users will work effectively and be happy about the changes made to their work activities during the development process of the information system (e.g., introducing the new software). This will all happen with the help of a socio-technical methodology (called ADISD).

122

Dissertations in Forestry and Natural Sciences No 153

References Alexander, I. & Maiden, N. (2004) Scenarios, Stories, Use Cases through the Systems Development Life-Cycle. John Wiley & Sons Ltd, West Sussex, England Alter, S. (2004a) IT Innovation Through a Work Systems Lens. In IT Innovation for Adaptability and Competitiveness (pp. 43-64). Springer US. Alter, S. (2004b) Desperately Seeking Systems Thinking in the Information Systems Discipline. In ICIS (pp. 757-770). Alter, S. (2006) The work system method: connecting people, processes, and IT for business results. Larkspur CA, Work System Press. Alter, S. (2008a) Defining information systems as work systems: implications for the IS field. European Journal of Information Systems, 17(5), 448-469. Alter, S. (2008b) Service system fundamentals: Work system, value chain, and life cycle. IBM Systems Journal, 47(1), 71-85. Alter, S. (2013) Work System Theory: Overview of Core Concepts, Extensions, and Challenges for the Future. JAIS 14( 2), 72-121, Andreev, P., Michalowski, W., Kuziemsky, C., & Hadjiyannakis, S. (2012) Application of activity theory to elicitation of user requirements for a computerized clinical practice guideline: the ActCPG conceptual framework. ECIS 2012 Proceedings. Paper 205. http://aisel.aisnet.org/ecis2012/205 Arsanjani, A. (2004) Service-oriented modeling and architecture. IBM developer works, 1-15. http://www.immagic.com/eLibrary/ARCHIVES/GENERAL/IBM/I04 1109A.pdf Avison, D. & Fitzgerald, G. (2006) Information systems development: methodologies, techniques and tools. 4th edition. McGraw Hill. Bardram, J. (1998) Designing for the dynamics of cooperative work activities. In Proceedings of the 1998 ACM conference on Computer supported cooperative work (pp. 89-98) ACM. Dissertations in Forestry and Natural Sciences No 153

123

Pentikäinen M.: Co-Development of Work and Information Systems

Bardram, J.E. (2009) Activity-based computing for medical work in hospitals. ACM Transactions on Computer-Human Interaction (TOCHI), 16(2), 10. Bardram, J., Gueddana, S., Houben, S., & Nielsen, S. (2012) Reticularspaces: Activity-based computing support for physically distributed and collaborative smart spaces. In Proceedings of the 2012 ACM annual conference on Human Factors in Computing Systems (pp. 2845-2854). ACM Baskerville, R. (1991) Practitioner autonomy and the bias of methods and tools. In Information Systems research: Contemporary Approaches & Emergent Traditions (eds.) Nissen, H-E., Klein, H.K. & Hircshheim, R. 673-698. North Holland, Amsterdam. Baskerville, R.L. (1999) Investigating information systems with action research. Communications of the AIS, 2(3es), 4. Baskerville, R.L. & Wood-Harper, A.T. (1996) A critical perspective on action research as a method for information systems research. Journal of Information Technology, 11(3), 235-246. Baskerville, R. & Myers, M.D. (2004) Special issue on action research in information systems: Making IS research relevant to practice — Foreword. MIS Quarterly, 28(3), 329-335. Baxter, G. & Sommerville, I. (2011) Socio-technical systems: From design methods to systems engineering. Interacting with Computers, 23(1), 417. Bertelsen, O.W. & Bødker, S. (2003) Activity theory. In Carrol, J.M. (eds.) HCI models, theories, and frameworks: Toward a multidisciplinary science, 291-324. Booch, G. (1986) Object-oriented development. IEEE Transactions on Software Engineering, (2), 211-221. Brinkkemper, S. (1996) Method engineering: engineering of information systems development methods and tools. Information and software technology, 38(4), 275-280. Brooks, F.P. Jr. (1995) The Mythical Man-Month. Essays on Software Engineering, The 20th Anniversary edition. Addison Wesley Longman, Inc.

124

Dissertations in Forestry and Natural Sciences No 153

References

Burnstein, I., Suwanassart, T., & Carlson, R. (1996) Developing a testing maturity model for software test process evaluation and improvement. In Test Conference, 1996. Proceedings., International (pp. 581-589). IEEE. Chakraborty, S., Sarker, S., & Sarker, S. (2010) An exploration into the process of requirements elicitation: a grounded approach. Journal of the Association for Information Systems, 11(4), 212-249. Checkland, P. (1981) Systems Thinking Systems Practice. Chichester Wiley. Checkland, P. (1999) Soft Systems Methodology: a 30-year retrospective. Wiley Checkland, P., & Holwell, S. (1998) Action research: its nature and validity. Systemic Practice and Action Research, 11(1), 9-21. Chen, P. (1976). The entity-relationship model—toward a unified view of data. ACM Transactions on Database Systems (TODS), 1(1), 9-36. Cheng, B. H., & Atlee, J. M. (2007) Research directions in requirements engineering. In Future of Software Engineering (pp. 285-303). IEEE Computer Society. Churchman, C. W., & Schainblatt, A. H. (1965) The researcher and the manager: A dialectic of implementation. Management Science, 11(4), B-69. Collins, P., Shukla, S., & Redmiles, D. (2002) Activity theory and system design: A view from the trenches. Computer Supported Cooperative Work (CSCW), 11(1-2), 55-80. Coughlan, J., & Macredie, R. D. (2002) Effective communication in requirements elicitation: a comparison of methodologies. Requirements Engineering, 7(2), 47-60. Damodaran, L., Simpson, A., & Wilson, P. (1982) Designing systems for people. NCC Publications. DeMarco, T. (1978) Structured analysis and system specification. Yourdon Press. Engeström, Y. & Sannino, A. (2010) Studies of expansive learning: Foundations, findings and future challenges. Educational Research Review, 5(1), 1-24. Engeström, Y. (1987) Learning by expanding. An activity-theoretical approach to developmental research. Orienta-Konsultit, Helsinki.

Dissertations in Forestry and Natural Sciences No 153

125

Pentikäinen M.: Co-Development of Work and Information Systems

Engeström, Y. (1990) Learning, working and imagining: Twelve studies in activity theory. Orienta-konsultit. Engeström, Y. (1999) Activity theory and individual and social transformation. Perspectives on activity theory, 19-38. In Engeström, Y., Miettinen, R., & Punamäki, R. L. (Eds.). (1999) Perspectives on activity theory. Cambridge University Press. Engeström, Y. (2000) Activity theory as a framework for analyzing and redesigning work. Ergonomics, 43(7), 960-974. Engeström, Y. (2001) Expansive learning at work: Toward an activity theoretical reconceptualization. Journal of education and work, 14(1), 133-156. Engeström, Y., Virkkunen, J., Helle, M., Pihlaja, J., & Poikela, R. (1996) The change laboratory as a tool for transforming work. Lifelong Learning in Europe, 1(2), 10-17. Fowler, M. (2004) UML distilled: a brief guide to the standard object modeling language. Addison-Wesley Professional. Fuentes, R., Gómez-Sanz, J.J., & Pavón, J. (2006) Integrating agentoriented methodologies with UML-AT. In Proceedings of the fifth international joint conference on Autonomous agents and multiagent systems (pp. 1303-1310). ACM. Fuentes-Fernández, R., Gómez-Sanz, J.J., Pavón, J. (2010) Understanding the human context in requirements elicitation, Requirements Eng (2010) 15:267–283 Georg, G. & Troup, L. (2013) Experiences Developing a Requirements Language Based on the Psychological Framework Activity Theory. Proceedings of the MODELS 2013 OCL Workshop Goldkuhl, G. (2008) Practical inquiry as action research and beyond. In European Conference on Information Systems, 2008. Galway: AIS. Gregor, S. (2006) The nature of theory in information systems. Mis Quarterly, 30(3), 611-642. Halverson, C. A. (2002) Activity theory and distributed cognition: Or what does CSCW need to DO with theories?. Computer Supported Cooperative Work (CSCW), 11(1-2), 243-267. Harrison, W., Barton, C., & Raghavachari, M. (2000) Mapping UML designs to Java In OOPSLA '00 Proceedings of the 15th ACM SIGPLAN conference on Object-oriented programming, systems, languages, and applications, pp. 178-187. 126

Dissertations in Forestry and Natural Sciences No 153

References

Hasan, H., Gould, E., & Hyland, P. (Eds.) (1998) Information systems and activity theory: tools in context. Wollongong,, Australia: University of Wollongong Press. Herr, K. & Anderon, G.L. (2005) The action research dissertation. A guide for students and faculty. Sage Publications Ltd, London. Herzum, P. & Sims, O. (2000) Business Component Factory, Wiley Computer Publishing, New York. Hevner, A.R., March, S.T., Park, J., & Ram, S. (2004) Design science in information systems research. MIS quarterly, 28(1), 75-105. Hirschheim, R., Klein, H.K., & Lyytinen, K. (1995) Information systems development and data modeling: conceptual and philosophical foundations (Vol. 9). Cambridge University Press. Holden, R.J., Brown, R.L., Alper, S.J., Scanlon, M.C., Patel, N.R., & Karsh, B.T. (2011) That’s nice, but what does IT do? Evaluating the impact of bar coded medication administration by measuring changes in the process of care. International journal of industrial ergonomics, 41(4), 370-379. Häkkinen, H., & Korpela, M. (2007) A participatory assessment of IS integration needs in maternity clinics using activity theory. International Journal of Medical Informatics, 76(11-12), 843-849. Jacobson, I. (1992) Object-oriented software engineering: a use case driven approach. Pearson Education India. Johnson, R.E. (1992) Documenting frameworks using patterns. In ACM Sigplan Notices (Vol. 27, No. 10, pp. 63-76). ACM. Kaptelinin, V. & Nardi B.A. (2006) Acting with technology: Activity theory and interaction design. Mit Press. Korpela, M. (1994) Nigerian practice in computer systems development. A multidisciplinary theoretical framework, applied to health informatics. Doctoral dissertation. 1994. Helsinki University of Technology. Department of Computer Science. Reports TKO-A31. Available at: http://www.uef.fi/documents/1020021/1942317/Korpela1994_DTech_NG-ISD.pdf. Korpela, M., & Mursu, A. (2003) Means for cooperative work and activity networks: An analytical framework. Workshop at ECSCW. 8th European Conference of Computer-Supported Cooperative Work, Helsinki, Finland, 14th September 2003.

Dissertations in Forestry and Natural Sciences No 153

127

Pentikäinen M.: Co-Development of Work and Information Systems

Korpela, M., de la Harpe, R., & Luukkonen, I. (2008) Depicting the landscape around information flows: Methodological propositions. In: SIG GlobDev Workshop Proceedings, Paris, France, 13 December 2008. Association for Information Systems. Korpela, M., Ikävalko, P., Luukkonen, I., Martikainen, S., Palmén, M., Tiihonen, T., Toivanen, M., & Vainikainen, V. (2013) How to CoDevelop Services, Work, and Information Systems in Healthcare: The Daisy Approach. In: Beuscart-Zéphir M-C, Jaspers M, Kuziemsky C, Nøhr C, Aarts J (eds). Context Sensitive Health Informatics: Human and Sociotechnical Approaches. Studies in Health Technology and Informatics 194:126-132. doi 10.3233/978-1-61499-2936-126. http://ebooks.iospress.nl/publication/34715 Korpela, M., Mursu, A., & Soriyan, H.A. (2001) Two times four integrative levels of analysis: a framework, in N.L. Russo, B. Fitzgerald and J.I. DeGross (eds.), Realigning research and practice in information systems development: the social and organizational perspective, (pp. 367-377), Boston, MA: Kluwer Academic. Korpela, M., Mursu, A., & Soriyan, H.A. (2002) Information systems development as an activity, Computer Supported Cooperative Work, vol. 11, 2002, pp. 111-128. Korpela, M., Mursu, A., Soriyan, A., Eerola, A., Häkkinen, H., Toivanen, M. (2004) Information systems research and development by activity analysis and development: dead horse or the next wave? In: Kaplan B, Truex III DP, Wastell D, Wood-Harper AT, DeGross JI, eds. Information Systems Research. Relevant Theory and Informed Practice. IFIP TC8/WG 8.2 20th Year Retrospective: Relevant Theory and Informed Practice-Looking Forward from a 20-Year Perspective on IS Research, Manchester, United Kingdom, July 15-17 2004, p. 453-471. Boston: Kluwer Academic, 2004. Korpela, M., Soriyan, H.A., & Olufokunbi, K.C. (2000) Activity analysis as a method for information systems development: General introduction and experiments from Nigeria and Finland. Scandinavian Journal of Information Systems 12(1), 191-210. Available at: http://aisel.aisnet.org/sjis/vol12/iss1/8

128

Dissertations in Forestry and Natural Sciences No 153

References

Korpela, M., Soriyan, H.A., Olufokunbi, K.C., Onayade, A.A., DaviesAdetugbo, A., & Adesanmi, D. (1996) Community participation in health informatics in Africa: An experiment in tripartite partnership in Ile-Ife, Nigeria. In: Blomberg, J., Kensing, F., and DykstraErickson, E.A., eds. PDC'96: Proceedings of the Participatory Design Conference, Cambridge, Massachusetts, USA, 13-15 November 1996, p. 25-34. Palo Alto: Computer Professionals for Social Responsibility, 1996. Kuutti K. (1994) Information systems, cooperative work and active subjects: the activity-theoretical perspective, University of Oulu, Department of Information Processing Science, Research Papers, Series A 23, University of Oulu, 1994. Kuutti, K. (1991a) The concept of activity as a basic unit of analysis for CSCW research. In Proceedings of the Second European Conference on Computer-Supported Cooperative Work ECSCW’91 (pp. 249-264). Springer Netherlands. Kuutti, K. (1991b) Activity theory and its applications to information systems research and development. H-E. Nissen, H. K. Klein & R. Hirscheim (eds.), Information systems research: contemporary approaches and emergent traditions. Amsterdam: Elsevier, 529-549. Leontjev, A.N., Hakkarainen, P., & Helkama, K. (1977) Toiminta, tietoisuus, persoonallisuus. Kansankulttuuri. (Finnish version of Leont'ev A.N. (1978), Activity, Consciousness and Personality, Engelwood Cliffs, NJ: Prentice Hall.) Lewin, K. (1946) Action research and minority problems. Journal of social issues, 2(4), 34-46. Lucas, H.C. Jr. (1975) Why Information System Fail. Columbia University Press. New York and London Luukkonen I. & Mykkänen J. (2012) Analyzing Process Modeling as Work Activity. In: Molka-Danielsen, J., Keller, C., and Wiberg, M. (Eds.) Selected Papers of the Information Systems Research Seminar in Scandinavia Nr. 3 (2012): IRIS 35 Designing the Interactive Society. p.9-24. Trondheim, Norway: Tapir Akademisk Forlag.

Dissertations in Forestry and Natural Sciences No 153

129

Pentikäinen M.: Co-Development of Work and Information Systems

Luukkonen I., Korpela M., & Mursu A. (2011) Studying the Applicability of Methods - Activity Driven Needs Analysis Applied to Maternity Pathway in Pudong, Shanghai. In: Proceedings of the International Workshop on Ubiquitous Healthcare and Supporting Services (UBIHEALTH’10), 31st May – 2nd June 2010, Shanghai, China, 2011. Luukkonen, I. (2012) Activity-Driven Needs Analysis and Modeling in Information Systems Development. Publications of the University of Eastern Finland. Dissertations in Forestry and Natural Sciences. 74 Luukkonen, I., Korpela, M., & Mykkänen, J. (2010) Modelling approaches in the early phases of information systems development. T. Alexander, M. Turpin, J.P. van Deventer (Eds.), Proceedings from –the 18th European Conference on Information Systems, IT to Empower. Luukkonen, I., Saranto, K., & Korpela, M. (2011a) Assessing the role of a site visit in adopting activity driven methods. In: Moen A., Andersen S.K., Aarts J., Hurlen P. (eds.), Proceedings of the XXII European Medical Informatics Conference (MIE 2011), User Centred Networked Health Care. IOS PRess BV. Studies in Health Technology and Informatics 169. pp. 422–426. Luukkonen, I., Toivanen, M., & Mursu, A. (2007) Toward planned changes: An activity-driven ISD model. In Proceedings of the 30th Information Systems Research Seminar in Scandinavia–IRIS30. Tampere, Finland (pp. 752-773). Martikainen S., Ikävalko P., Korpela M. (2010) Participatory interaction design in user requirements specification in healthcare. In: Safran C, Reti S, Marin HF, eds. Medinfo 2010 - Proceedings of the 13th World Congress on Medical Informatics, Cape Town, 12-15 Sep 2010. Studies in Health Technology and Informatics 160(Pt 1):304-308. Martins, L.E.G., & Daltrini, B.M. (1999) An Approach to Software Requirements Elicitation Using the Precepts from Activity Theory. In ASE 14th IEEE International Conference (pp. 15-23). Meyer, B. (1988) Object-oriented software construction. Prentice Hall, London.

130

Dissertations in Forestry and Natural Sciences No 153

References

Minkkinen, I., & Eerola, A. (2007) Improving requirements engineering from the client’s perspective in the health care domain. In Proceedings of the IASTED International Conference on Software Engineering as part of the 25th IASTED International Multi-Conference on Applied Informatics (Hasselbring W, Ed.), Innsbruck, Austria (pp. 93-98). Misic, M.M., & Graf, D.K. (2004) Systems analyst activities and skills in the new millennium. Journal of Systems and Software, 71(1), 31-36. Mumford, E. (2000) A socio-technical approach to systems design. Requirements Engineering, 5(2), 125-133. Mumford, E. (2006) The story of socio-technical design: reflections on its successes, failures and potential. Information Systems Journal, 16(4), 317-342. Mursu, A. (2002) Information systems development in developing countries: Risk management and sustainability analysis in Nigerian software companies. University of Jyväskylä. Mursu, A., Luukkonen, I., Toivanen, M., Korpela, M. (2007) Activity theory in information systems research and practice: theoretical underpinnings for an information systems development model. Inf Res 2007:12(3):[26 p.]. Available at: http://informationr.net/ir/123/paper311.html Mwanza, D. (2001) Where Theory meets Practice: A Case for an Activity Theory based Methodology to guide Computer System Design. In: Proceedings of INTERACT’ 2001: Eighth IFIP TC 13Conference on Human-Computer Interaction, 9-13 July 2001, Tokyo, Japan Myers, M. D., & Avison, D. (2002) An introduction to qualitative research in information systems. In: eds. Myers, M.D. and Avison D. Qualitative Research in Information Systems. Mykkänen, J., Minkkinen I., Pöyhölä, A. and Riekkinen, A. (2006) Improving traceability of functional requirements to information needs and applications in healthcare. 6th Nordic Conference on eHealth and Telemedicine August 31 - September 1, 2006 Helsinki, Finland.

Dissertations in Forestry and Natural Sciences No 153

131

Pentikäinen M.: Co-Development of Work and Information Systems

Mykkänen, J., Porrasmaa, J., Korpela, M., Häkkinen, H., Toivanen, M., Tuomainen, M., Häyrinen, K., Rannanheimo, J. (2004) Integration models in health information systems: experiences from the PlugIT project. In: Fieschi M., Coiera E., Li Y-C J., eds. MEDINFO 2004. Proceedings of the 11th World Congress on Medical Informatics, San Francisco, California, 7-11 September 2004, p. 1219-1222. Amsterdam: IOS Press. Nardi, B.A. (Ed.). (1996) Context and consciousness: activity theory and human-computer interaction. Mit Press. Nielsen, J. (1993) Usability engineering. Academic Press. Nuseibeh, B. & Easterbrook, S. (2000) Requirements engineering: a roadmap. In Proceedings of the Conference on the Future of Software Engineering (pp. 35-46). ACM. Offen, R. (2002) Domain Understanding is the key to successful system development, Requirements Eng 7: 172-175 Orlikowski, W. J., & Baroudi, J. J. (1991) Studying information technology in organizations: Research approaches and assumptions. Information systems research, 2(1), 1-28. Palmén, M., Korpela, M., & Saranto, K, (2012) Identifying health and wellbeing related information management activities and needs in families with young children. Finnish Journal of eHealth and eWelfare 4(3):165-176. Pärjäin. OmaHyvinvointi-hankkeen loppuraportti (2010) eds. Meristö, T., Muukkonen, P., Nurminen, M.I. & Tuohimaa, H. Available at: http://omahyvinvointi.utu.fi/assets/images/OHV_loppuraportti.pdf Pastor, O., Gómez, J., Insfrán, E., & Pelechano, V. (2001) The OO-Method approach for information systems modeling: from object-oriented conceptual modeling to automated programming. Information Systems, 26(7), 507-534. Pfleeger, S.L. &Atlee, J.M., (2010) Software engineering. Fourth Edition. Pearson. Pfleeger, S.L., & Hatton, L. (1997) Investigating the influence of formal methods. Computer, 30(2), 33-43. PlugIT (2004) Web-site of the PlugIT project. Available at: http://www.oppi.uef.fi/uku/plugit/english/ (22.10.2013) Pohl, K. (1994) The three dimensions of requirements engineering: a framework and its applications. Information systems, 19(3), 243-258. 132

Dissertations in Forestry and Natural Sciences No 153

References

Quek, A (2006) AcCent: An activity-theoretical framework for web-based systems development. PhD thesis, Staffordshire University. Quek, A. & Shah, H. (2004) A Comparative Survey of Activity-Based Methods for Information Systems Development. In ICEIS (5) (pp. 221-232). Riekkinen, A. (2004) Deriving software architecture from activity analysis. In Bertelsen, O.W., Korpela, M. & Mursu, A., eds. ATIT 2004. Proceedings of the First International Workshop on Activity Theory Based Practical Methods for IT-Design, 157-168. Robertson, S. & Robertson, J. (1999) Mastering the Requirement Process, Addison-Wesley. Rook, P. (1986) Controlling software projects. Software Engineering Journal, 1(1), 7. Saaren-Seppälä, K. (1997) Seinätekniikka prosessien kehittämisessä. Using the Wall-Chart Technique in Process Development. Kari Saaren-Seppälä Ltd., Finland. Scheer, A.W. & Nüttgens, M. (2000) ARIS architecture and reference models for business process management Lecture Notes in Computer Science, Volume 1806, Business Process Management, SpringerLink (pp. 376-389). Shaw, M. (1990) Prospects for an Engineering Discipline of Software. IEEE Software (November 1990), 15-24. Shlaer, S. & Mellor, S.J. (1988) Object-Oriented Systems Analysis. Modeling the World in Data. Yourdon Press computing series, Englewood Cliffs, NJ: Yourdon Press,| c1988, 1. Sommerville, I (2011) Software Engineering. Pearson. Ninth Edition. Song, X. (1995) A framework for understanding the integration of design methodologies. ACM SIGSOFT Software Engineering Notes, 20(1), 4654. Sparrow, L., Nguyen, L., & Cybulski, J. (2006) The slow wheels of requirements engineering research: responding to the challenge of societal change, in Eleventh Australian Workshop on Requirements Engineering: proceedings, University of South Australia, Adelaide, S. Aust., pp. 1-9. Szyperski, C. (2002) Component software: beyond object-oriented programming. Pearson Education.

Dissertations in Forestry and Natural Sciences No 153

133

Pentikäinen M.: Co-Development of Work and Information Systems

Toivanen, M. (2004) IT needs exploration [online]. In: Bertelsen O.W., Korpela M., Mursu A., eds. ATIT 2004. Proceedings of the First International Workshop on Activity Theory Based Practical Methods for IT-Design, Copenhagen, Denmark, 2-3 Septemper, 2004, p. 147-151. 2004. Available at: http://www.daimi.au.dk./PB/574/PB-574.pdf Toivanen, M., Häkkinen, H., Laitinen, P., Minkkinen, I., Röppänen, P., & Tuomainen, T. (2004a) Kotihoidon tiedon tarpeet. Kuopion yliopisto, 2004. PlugIT-hankkeen selvityksiä ja raportteja 12. Toivanen, M., Häkkinen, H., Minkkinen, I., Riekkinen, A., Ikävalko, P., & Röppänen, P. (2004b) Toimintalähtöisyys tiedon tarpeiden, tiedonkulun ja ohjelmistovaatimusten selvittämisessä. Kuopion yliopisto. PlugIT-hankkeen selvityksiä ja raportteja 11. Toivanen, M., Luukkonen, I., Ensio, A., Häkkinen, H., Ikävalko, P., Jaatinen, J., Klemola, L., Korhonen, M., Martikainen, S., Miettinen, M., Mursu, A., Röppänen, P., Silvennoinen, R., Tuomainen, T., and Palmén, M. (2007) Kohti suunnitelmallisia muutoksia. Opas terveydenhuollon tietojärjestelmien toimintalähtöiseen kehittämiseen. Kuopion yliopiston selvityksiä E. Yhteiskuntatieteet 39. 80 s. Available at: http://www.oppi.uef.fi/uku/zipit/julkaisut/978-95127-0880-2.pdf Toroi, T. (2009) Testing Component-based Systems: Towards Conformance Testing and Better Interoperability. University of Kuopio. Truex, D., Baskerville, R., & Travis, J. (2000) Amethodical systems development: the deferred meaning of systems development methods. Accounting, management and information technologies, 10(1), 53-79. Tuomainen, M., Mykkänen, J., Luukkonen, I., Itälä, T., Kotinurmi, P., Han, Y., Sommer, N., Nykänen, P., & Toivanen, M. (2010a) OmaHyvinvointi-hanke. PHR-katsaus. Tuomainen, M., Toivanen, M., Mykkänen, J., Palmén, M., Luukkonen, I., Itälä, T., Tarkkanen, K., Reijonen, P. & Kouri, P. (2010b). Pärjäimen suunnitteluperiaatteet–käyttökonteksti, tiedot ja arkkitehtuuri. Turner, P., Turner, S., & Horton, J. (1999). From description to requirements: an activity theoretic perspective. In Proceedings of the international ACM SIGGROUP conference on Supporting group work (pp. 286-295). ACM.

134

Dissertations in Forestry and Natural Sciences No 153

References

Van Der Aalst, W.M., Ter Hofstede, A.H., & Weske, M. (2003) Business process management: A survey. In Business process management (pp. 1-12). Business Process Management, Lecture Notes in Computer Science Volume 2678, 2003, pp 1-12, Springer Berlin Heidelberg. Van Lamsweerde, A. (2000) Requirements engineering in the year 00: A research perspective. In Proceedings of the 22nd international conference on Software engineering (pp. 5-19). ACM. Virkkunen, J. & Ahonen, H. (2011) Supporting expansive learning through theoretical-genetic reflection in the Change Laboratory. Journal of Organizational Change Management, 24(2), 229-243. Virkkunen, J. & Kuutti, K. (1999) Questioning the unit of analysis in BPR. In Khoong, C. M. (ed.) Reengineering in action – The Quest for WorldClass Excellence. London: Imperial College Press, 156-172 Virkkunen, J. & Newnham, DS. (2013) The Change Laboratory A Tool for Collaborative Development of Work and Education. Sense Publishers. Vygotsky, L.S. (1978) Mind in society: The development of higher mental process. Compiled from several sources and edited by Cole M., JohnSteiner V. and Scribner S., Harvard MA: Harvard University Press. Walls, J.G., Widmeyer, G.R., & El Sawy, O.A. (1992) Building an information system design theory for vigilant EIS. Information systems research, 3(1), 36-59. White Baker, E. (2011) Why situational method engineering is useful to information systems development. Information Systems Journal, 21(2), 155-174. Wiegers, K.E. (2003) Software requirements. Microsoft Press. Zave, P. & Jackson, M. (1997) Four dark corners of requirements engineering. ACM Transactions on Software Engineering and Methodology (TOSEM), 6(1), 1-30. ZipIT (2007) project website: http://www.oppi.uef.fi/uku/zipit/ Zündorf, A. (2001) Tutorial: From use cases to code-rigorous software development with UML. In Proceedings of the 23rd International Conference on Software Engineering (pp. 711-712). IEEE Computer Society.

Dissertations in Forestry and Natural Sciences No 153

135

Marika Pentikäinen Co-Development of Work and Information Systems

This thesis is about attempts to

An Analysis of the Construction of the

improve work and information

Activity-Driven ISD Methodology in 2001-2013

systems together. The ActivityDriven Information System Development (ADISD) methodology was constructed for that purpose during the years 2001-2013. ADISD was developed and applied in several ISD projects in the context of health care in Finland. This research takes a look back and analyzes the methodology development activities to identify possibilities for further development and to contribute to other methodology developers.

Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences isbn 978-952-61-1558-0