Introduction to Decision-making: The Process This paper is extracted from the BE1172 Project Risk and Value Management module, Department of Architecture, Engineering and Construction, School of the Built and Natural Environment, Northumbria University.

Abstract PM have responsibilities for making sure they make sound judgements based on the acquisition of accurate, and sometimes, in complete data. This paper illustrates the fundamental aspects of decision making. Starting from an epistemic overview of the types of decisions that can be made and the associated decision making process, issues associated with the gathering of data and making alternative decisions will be covered. The paper should take you around 60 minutes to read and complete any directed tasks.

Types of Decision A useful way to characterise the various decision-making techniques, is by a number of variables that surround the nature of the required decision. This is illustrated by a simple decision-making model.

CONSTRAINTS

INPUTS information about the alternatives

Decider

Bounded rationality

OUTPUTS The Decision(s) made

RESOURCES Decision techniques

Thus the main variables can be considered to relate to:  

OUTPUTS: is the objective a single or multiple one? And, if multiple, what criteria should be used to ‘weight’ different results. INPUTS: how is information about the costs and benefits of the decision computed? Are the measures qualitative or quantitative? 1

Introduction to Decision Making: The Process

  

2

RESOURCES: what are the techniques available for making this type of decision? CONSTRAINTS: how much is known about the decision and its circumstances? Will there be a high degree of uncertainty that requires the use of probabilistic techniques? THE DECIDER: can the decision-maker be treated as a single entity, or is it a multiple decider? And, if the latter, is the decision likely to be co-operative (as in negotiation) or competitive (as in bidding and tendering)?

Based upon the variables above, decision theorists (see, for example Simon, 1977) have characterised different types of decision, by, for example, (1) the extent to which they can be solved ‘deterministically’ (i.e. they are programmable); (2) whether their measures of success are quantitative or qualitative; (3) whether their criteria are single or multiple; and by the number of participants involved (4), and (5) whether these participants are cooperative or competitive. (See the table, below)

Type Variable Measures

1

2

3

4

5

quantitative

qualitative or mixed

either

either

either

Certainty

certain

certain

uncertain

either

either

Objectives

single

single

single

multiple

either

Participants

single

single

single

single

multi

The following section deals with the decision-making process. Except for one element of the process – the calculation and evaluation of solutions – the process is fairly generic whatever the type of decision is in question. We will, therefore, deal with it in that way, and only consider these differences at that point in the process (as reflected in the Summary, above).

The Decision-making Process There are many models of the decision-making process. Classic examples include those of Dewey (1933) and Simon (1977). Dewey’s 5-phases of ‘reflective thought’ are:

1. 2.

Suggestion: the initial prompt, or instinctive reaction(s) to a problem; Intellectualisation: a definition of ‘the conditions that constitute the trouble’;

Introduction to Decision Making: The Process

3. 4. 5.

3

Guiding idea or hypothesis Reasoning: theorising about the feasibility of the idea(s) generated in stage 3; Testing: by putting the accepted idea into practice.

Simon (1977) describes four stages, namely: 1. 2. 3. 4.

Intelligence (i.e. gathering of information) Design (of alternative solutions) Choice (between them) Review (of the implemented solution).

It is now common to add the initial step of problem definition. In fact this was prompted by the recognition that one of the main barriers to the effectiveness of the decision-making or problemsolving process was the failure to understand the problem in the first place. The list of stages that we will work with, therefore is slightly longer than Dewey’s or Simon’s. The stages are as follows:       

defining the problem gathering data identifying alternative solutions evaluating their cost and effectiveness selecting the best implementing the selected option monitoring, evaluation and improvement (if possible).

Stage 1: Defining the Problem As we have already noted one of the main barriers to the effectiveness of the decision-making or problem-solving process can be the failure to understand the problem in the first place. There are various techniques for overcoming this, and most fall into one of two categories, namely: 

prompting and analysis techniques – methods that stimulate a logical approach to decision / problem causality (a classic example being the cause and effect or fishbone diagram). These techniques work on the inputs element of our simple decision-making model (shown above).



redefinition techniques – which differ in that they involve inducing a shift in the way the decision / problem is perceived. The concept of lateral thinking (see, De Bono, 1972) is a common example. These techniques try to overcome the constraints element of our simple decision-making model.

We will examine a few examples of each group. Four examples of prompting and analysis techniques are:  

questioning technique cause and effect diagrams

Introduction to Decision Making: The Process

 

4

root cause analysis cognitive maps

and two redefinition techniques are:  

progressive abstraction, and boundary examination.

Prompting and Analysis Techniques Perhaps the simplest, most common sense of these techniques involves a simple analysis of a problem driven by the questions - what, where, when, who, how and why.

Task At this point you should stop reading the paper and complete a self-reflection task during which you should think about a problem –possibly from your professional practice experience – and undertake an analysis of the problem you have identified using the questions outlined immediately above: what, where, when, who, how and why?

Category

Column 1

Column 2

Purpose

WHAT is being done?

WHY is it being done?

Place

WHERE is it done?

WHY there?

Sequence

WHEN is it done?

WHY then?

Person

WHO does it?

WHY them?

Means

HOW is it done?

WHY that way?

Column 3 What else could be done? Where else could it be done? When else could it be done? Who else could do it? How else could it be done?

Column 4 What should be done? Where should it be done? When should it be done? Who should do it? How should it be done?

Note that the matrix can be used in different ways: if a problem is a new one, and being tackled ab initio then only the fourth column need be used. However, where there is an existing system in place (and that system is not working) all four columns should be applied.

A more sophisticated tool is the ‘cause and effect’ diagram. There are a variety of variants of cause and effect diagrams, including Fishbone Diagrams, Cognitive maps, mind maps, influence diagrams and the like. A classic example is the Fishbone Diagram of Professor Kaoru Ishikawa of Tokyo University (see Majaro, 1991). A Fishbone Diagram is an analysis tool to display possible

Introduction to Decision Making: The Process

5

causes of a specific effect, problem, condition, or (as we will see later) risk. In essence it is the reverse of the decision tree (see later notes).

Cause 3

Cause 2

Cause 3.1

PROBLEM Cause 4.1

Cause 4

Cause 1.1

Cause 1

A similar, but more developed form of cause and effect mapping is used in the technique of Root Cause Analysis (RCA), a tool used for health, safety, environmental and quality investigations. Root causes are specific underlying causes that can be identified and over which the decision-maker has control. The RCA procedure (based upon a description by Rooney and Vanden Heuvel 2004) is as follows:

1. 2.

3. 4. 5.

collect relevant available data (this stage may have to be revisited and repeated as a picture emerges); organise and chart the data in a sequence diagram (very much the same as the cause and effect diagram shown above) showing the causal links that lead up to an effect or occurrence, and the conditions that surround it; identify the causal factors. Causal factors are ‘those contributors…that, if eliminated, would have either prevented the occurrence or reduced its severity’; identify the root cause(s). Root causes are the underlying reasons for each factor; and recommendations and their implementation.

Cognitive maps are 'a modelling technique which portrays ideas, beliefs, values and attitudes and their relationships to one another in a manner which facilitates examination and analysis' (Proctor, 1999: 237).

Such techniques can be more (influence diagrams) or less (mind maps) structured. According to Buzan (1994: 59), the Mind Map has four essential characteristics:    

The subject is crystallized in a central image The main themes radiate from it on branches Branches comprise a key image. Topics of lesser importance are on sub-branches The branches form a connected nodal structure.

Introduction to Decision Making: The Process

6

Redefinition Techniques Progressive abstraction is a technique, proposed by Greschka et al. (1973) that is one of a variety of redefinition techniques. It relies on identifying the essence of a problem through a series of abstractions. Take any particular perspective on a problem and imagine that it is on the rung of a ladder. The other rungs are different perspectives that are more abstract or less abstract than the original: asking ‘why?’ - moves up the ladder to a higher level of abstraction; asking ‘how?’ moves down the ladder, to a lower one. Such redefining of the problem can result in improved perspectives.

Boundary examination is an example of a redefinition approach closely connected with lateral thinking (de Bono, 1972). It encourages the problem-solver to take a fresh look at assumptions.

The process is:

1. 2. 3. 4.

Write an initial statement of the problem Highlight key words and analyse for hidden assumptions Concentrate on the impact of assumptions Record any new problem definitions that emerge

Stage 2: Gathering the Data Once a problem has been properly identified, the next step is to gather as much data as possible. To assist in doing this, there are a number of ‘prompting’ techniques. These include:   

Checklists SWOT analysis & PEST analysis Pareto analysis

Checklists Checklists are usually based on past experience of the problem in question. This accounts for both their strength and their weakness. A simple example would be the six-question matrix featured above. An example of a more detailed checklist (for project risks) is given below.

Introduction to Decision Making: The Process

Category

7

Sub-category

Identified Risk

Ground, Weather, Site

Collapse of sides of trench excavations

&

Security, Safety, Environment

resulting in delays, additional cost and

Technical

Scope, Technology / Interfaces

injury. Rigorous testing and

Commissioning / approval

commissioning required

Novelty / quality of design

Delays in design decisions and drawing

Completeness, Novation

issue caused by physical distance

Interfaces, location of designer

between project and designers' offices

Contractual

Contract / Insurance, Bonds, etc.

Strict 'fit for purpose' design liability

& Commercial

Liability / limitations/ exclusions

under the main contract, excluded by

Price, Cost, Payment, Cash-flow

subcontract designers' offers.

Exchange rates, Inflation

Causing excessive financial loss.

Physical

Design

Financial

Funding stability Client stability Managerial

External

Staff, Organisation, Resources

Non-availability of labour due to the

Subcontractors

construction of another nearby major

Communication

project which causes a regional shortage

Timescale

of specialist sub-contractors

Legislative & Regulatory

Disruption to a third party’s business

Political, Pressure groups

due to noise or construction traffic

Force majeure

resulting in financial loss / litigation

Environmental

SWOT Analysis The acronym SWOT is short for Strengths, Weaknesses, Opportunities, and Threats. The analysis is designed to assist with the simultaneous evaluation of something (an organisation, a business plan, a proposal…) against positive (Strengths) and negative (Weaknesses) internal factors and positive (Opportunities) and negative (Threats) factors that are external.

Strengths

………….

Factors that are internal to the organisation

………….

Factors that are external to the organisation

Weaknesses ………….

Opportunities …………

Threats

Introduction to Decision Making: The Process

8

PEST Analysis For exploring 'opportunities' and 'threats', the external factors, analysts often have recourse to a further simple device, the PEST diagram. The concept of ‘PEST’ (Political, Economic, Social, Technical) serves to prompt the recognition of external influences on an organisation.

political

economic

THE CORPORATE BUSINESS

technical

social

Pareto Analysis The Pareto principal (otherwise known as the ‘80-20 rule’) is named after the Italian Economist Vilfredo Pareto (1848-1923) and in particular his observation that 20% of the people in Italy owned 80% of its wealth. As a management tool Pareto Analysis relies on the same instinctive approach for decision-making, or more specifically for identifying those actions that will show the biggest returns.

“Example: Company A bids for a number of projects a year and needs to establish the probable cost of every activity in a project in order to establish a bid price. In a typical project bid there are 100 activities (and these tend to be similar from project to project) but it emerges that 20 of these normally represent 80% of the total project cost. How should this discovery affect the way Company A’s estimator spends his time, and the data to be gathered?”

Stage 3: Identifying Alternative Solutions The next stage in the classic problem-solving process is to identify alternative solutions. As with the previous section, the two main sources upon which to draw are derived from:

a. b.

experience, or creative thinking

Introduction to Decision Making: The Process

9

Experience, Tacit Knowledge and Organisational Learning It is an old dictum that people 'learn from experieOce', but research has suggested that this is not necessarily true. Kolb's theory of ‘Experiential Learning’ (see Kolb et al., 1995) proposes a cycle one part of which is experience. It is probably truer to say that people learn from 'reflecting on experience'. Individuals do this intuitively - though not always well. However most organisations are spectacularly bad at learning from their organisational experiences - or, put another way exploiting organisational learning.

Active Experimentation

Abstract Conceptualisation

Concrete Experience Reflective Observation

Creative Thinking: Brainstorming The alternative to acquiring solutions from the collective experience of the organisation is creative thinking. Brainstorming is a very old and well-known example of this type of technique. Brainstorming can be structured and unstructured. Its rules can be summarised as:

1. 2. 3. 4.

Brainstorm first, evaluate later Go for quantity rather than quality Encourage wild ideas Build on and combine ideas.

There exist a number of barriers to creative thinking, and these can include being 'captives of experience' as well as sheer lack of perception. Many of the theories of perception provide an insight into why we fail to think creatively when attempting to solve problems. In fact it appears that the majority of the theories about how we perceive are derived from observations upon how we fail to perceive!

Introduction to Decision Making: The Process

10

Stage 4: Evaluation of Alternatives There are a number of techniques that are available for the evaluation and selection of alternative solutions. These include:    

programmable (deterministic) decision techniques techniques for decisions with mixed qualitative/quantitative data multi-criteria decision techniques probabilistic or stochastic techniques

However, since some of these techniques are complex, and require worked examples, they are treated in a separate paper ‘Evaluation Techniques’ while the present paper remains focused on the process of decision-making. The techniques covered in the ‘Evaluation Techniques’ paper, are: 

 

Programmable (deterministic) decision techniques o Force-field (Pro and Con) analysis o Investment appraisal  Simplistic methods (Rate of return and Payback period)  Discounted cash-flow methods  Equivalent Annual Cost (EAC)  Net Present Value (NPV)  Internal Rate of Return (IRR) o Pay-off tables o Methods from Operational Research  linear programming;  assignment / allocation/ transportation;  scheduling, sequencing, queuing. o Decision trees Decisions with mixed qualitative/quantitative data o Cost-benefit analysis Multi-criteria decision techniques o Best outcome o Mean best outcome o Expected Value Best Outcome o Maximin Criterion o Minimax Regret Criterion (Savage’s Regret Criterion1)

Stage 5: Implementing the Selected Alternative Two aspects of implementation are considered here: 

1

Communication

Savage (1951)

Introduction to Decision Making: The Process



11

Resistance to change

Communication is a complex subject and communication theories offer many models describing the process. An example is offered by Rogers (1962). It involves 5 stages:

AWARENESS - INTEREST - EVALUATION - TRIAL - ADOPTION

One of the main barriers to implementation of decisions is resistance to change. Lawrence and Greiner (1970) give five basic causes for this:

1. 2. 3. 4. 5.

Fear of the unknown Lack of information Threat to status Fear of failure Lack of perceived benefits.

Stage 6: Monitoring and Evaluation In management it is important that, once decisions have been implemented, that they are monitored and evaluated (and further action taken if the results are unsatisfactory). The following simple diagram illustrates the way the elements of monitoring and evaluation fit together with other parts of the problem-solving process.

PLAN ACT

DO CHECK

The monitoring and evaluation process

Introduction to Decision Making: The Process

12

Bibliography and References Allais, M. (1979) The So-Called Allais Paradox and Rational Decisions Under Uncertainty. Dordrecht: Reidel. Bazerman, M.H., Curhan, J.R., Moore, D.A., and Valley, K.L. (2000) ‘Negotiation’. Annual Review of Psychology Vol. 51: 279-314. Buzan T. (1993) The Mind Map Book. BBC Active, Harlow, England. Byrne, P. (1996) Risk, uncertainty and decision-making in property development. E. & F.N. Spon, London. De Bono, E. (1972) The five-day course in thinking. Penguin Books. Dewey, J. (1933) How we think. A restatement of the relation of reflective thinking to the educative process (Revised edn.), Boston: D. C. Heath. Kolb D., Rubin I., and Osland J. (1995) Organizational Behaviour, an experiential approach (6th ed), Prentice-Hall Majaro, S. (1991) Creative Gap: Managing Ideas for Profit. McGraw-Hill Book Company, Maidenhead, UK. Pilcher, R. (1992) Principles of Construction Management (3rd edition) McGraw-Hill Book Company, Maidenhead, UK. Rooney, J. and Vanden Heuvel, L. (2004) ‘Root cause analysis for beginners’. Quality Progress 4553. Savage, L.J. (1951) ‘The theory of statistical decision’. Journal of the American Statistical Association, vol. 46, pp. 55-67. Simon, H. (1977) Models of Discovery: and other topics in the methods of science. Dordrecht, Holland: Reidel. Walker, P. and Greenwood, D.J. (2002) Construction Companion to Risk and Value. RIBA Publications.