Gothenburg Mathematical Modelling Centre

Annual Report 2009

Editor: Setta Aspström Gothenburg 2010

Contents Introduction

4

People

6

Calendar

7

Optimisation and Modelling

9

Risk, Reliability, Quality

12

Bioinformatics

14

Publications

16

Financial report

26

Introduction Some highlights: • Publication of a Wiley book “Robust Design Methodology for Reliability” and a “Guide to load analysis for automotive applications” which together sets out the new philosophy and methods for reliability improvement which have been developed by GMMC. • A new cooperation with Albany International, EKA Chemicals, Stora Enso and Tetra Pak on dynamic numerical and stochastic simulation of paper fiber networks. • Multivariate regression chains have been developed as a framework for multivariate statistical analyses. They permits local modeling of development over time in single and joint responses and provide guidelines for planning and comparing follow-up studies. • We have made significant progress on numerical approximation of stochastic partial differential equations. Influx of good young people with enthusiasm for strategic research and the ability to combine frontline mathematics with substantial applications is central to GMMC. During 2009 we have hired two research associates/postdocs, Larissa Beilina and Carl Lindberg. Larissa’s area is inverse problems, earlier in signal analysis and now also as a partner in GMMC research on light ion radiation treatment. Carl works in our Risk, Reliability, Quality team. His current interest is strategies for optimal trading of financial instruments. Further, Thomas Gebäck has just begun in a GMMC postdoc position. He will also work on radiation treatment, and one aim is that he will be a link between GMMC and Anders Brame’s Radiation Physics group at Karolinska Institutet. The results of the midterm evaluation of the SSF strategic research centres were announced early in the year. GMMC received the summary grades Outstanding, Excellent, and Very Good, respectively, by the three external reviewers, and additional funding of SEK 4 million. At the beginning of the fall semester we arranged a “GMMC day” where all GMMC researchers gave talks on research problems and research plans for the future. It was quite a success and will be repeated. The Risk, Reliability, Quality team arranged the 2009 ENBISS (European Network for Business and Industrial Statistics) conference in Gothenburg. It attracted about 160 participants. Workshops continue to be one of the main GMMC tools. We use them to start new collaborations, intensify existing ones, learn the latest scientific and technical developments, and to disseminate GMMC results to industry and academia. This year we arranged workshops on Optimization of Maintenance Activities, Generalized Linear Models with Random Effects, Diffusion Measurement in Gel Networks, and Fatigue Design for Safety.

4

Throughout we use joint financing to amplify the effects of GMMC funding. Our new research associates/postdocs received additional funding by Mathematical Sciences in Gothenburg and by Radiation Physics at Karolinska Institutet. The Stochastic Centre in Gothenburg, the Departments of Applied Mechanics and of Shipping and Marine Technology at Chalmers, the Swedish Statistical Association, and the Swedish Testing and Research Institute in various ways were partners in our workshops. During the fall FCC ran an advertising campaign where GMMC was a part. Anette Runsteen, Astra-Zeneca, replaced Nibia Aires as GMMC board member since Nibia moved to a new position in Switzerland. Holger Rootzén, centre leader

In an advertising campaign (on page 3 of GöteborgsPosten, GP), FCC presented several examples of how mathematics enables technical advances through modelling, simulation, optimisation, and quality assurance.

5

People Leaders Holger Rootzén, centre leader Michael Patriksson, deputy centre leader Members Marina Alexanderson-Fisk, Mathematical Statistics Larisa Beilina, Mathematics Bo Bergman, Quality Sciences Fredrik Edelvik, Fraunhofer–Chalmers Centre Olle Häggström, Mathematical Statistics Peter Jagers, Mathematical Statistics Mats Jirstrand, Fraunhofer–Chalmers Centre Bo Johansson, Mathematics Stig Larsson, Mathematics Carl Lindberg, Mathematical Statistics Jacques de Maré, Mathematical Statistics Olle Nerman, Mathematical Statistics Uno Nävert, Fraunhofer–Chalmers Centre Mats Rudemo, Mathematical Statistics Igor Rychlik, Mathematical Statistics Aila Särkkä, Mathematical Statistics Bernt Wennberg, Mathematics Nanny Wermuth, Mathematical Statistics PostDocs Frida Abel, Mathematical Statistics Gunnar Cedersund, Mathematics Peter Gennemark, Mathematics Carl Lindberg, Mathematical Statistics Richard Wilson, Mathematical Statistics Board Jan S. Nilsson, chairman Lena Carlsson, University of Gothenburg Tomas Morsing, Second AP Fund Annette Runsteen, AstraZeneca Anders Ydergård, Volvo 3P

6

Scientific Board Gunnar Andersson, KP Pension & Försäkring Sir David Cox, University of Oxford Magnus Johansson, SKF Hilary Ockendon, University of Oxford Dieter Prätzel-Wolters, Fraunhofer ITWM Simon Tavaré, Cambridge Research Institute Secretary Setta Aspström, Mathematical Sciences Guests Alan Demlow, University of Kentucky at Lexington Gilles Guillot, Technical University of Denmark Sayan Gupta, Indien, Indian Institute of Technology Madras Miroslav Halas, Slovak University of Technology Fredrik Hellman, Uppsala universitet Jürg Hüsler, University of Bern Tomas Johansson, University of Birmingham Michael V. Klibanov, University of North Carolina at Charlotte Mihaly Kovacs, Otago University Omar Lakkis, University of Sussex Per-Gunnar Martinsson, University of Colorado at Boulder Madhuri Nukala, Mid Sweden University Natee Pantong, University of North Carolina at Charlotte Claudia Redenbach, University of Kaiserslautern Davood Rostamy, University of Tehran Andrew Stuart, University of Warwick Hermann Thorisson, Háskóli Íslands

Calendar Workshops and courses Industry–Academy meeting: Mathematics and Statistics for Insurance and Finance, Sigtuna, February 10–11 Generaliserade lineära modeller med slumpeffekter, Vålådalen, March 9–12 Fatigue Design for Safety, Göteborg, May 25–27 9th European Network for Business and Industrial Statistics, theme Quantitative Process Analysis for Creation of Business Opportunities and Solutions, Gothenburg, September 20–24 Robust Design for Reliability, Cochlear, September 24, with the participation of invited guest researchers from National University of Singapore (Prof. Thong Ngee Goh, Prof. Loon Ching Tang) Chaos Expansions, Finite Elements, and Randomly Forced Equations, course held jointly with LTH Optimization of Maintenance Activities – Models, Methods and Applications, Gothenburg, December 10–11

Seminars GMMC meeting, February 18 Jacques de Maré: The Reliability Group Rebecka Jörnsten: Multi-level clustering GMMC meeting, May 26 Dmitrii Zholud: Quality Assurance for BioScreen Experiments Fredrik Edelvik/Stig Larsson: Innovative Simulation of Paper

GMMC Day, August 20 Joachim Almquist: System Identification from Spatiotemporal Cell Population Data Milena Anguelova: Combinatorial complexity and model reduction from invariant manifolds Martin Arvidsson: Identification of dispersion effects in replicated split-plot designs Larisa Beilina: A globally convergent numerical method and the adaptivity technique for a hyperbolic coefficient inverse problem Christoffer Cromvik: Stochastic mathematical programs with equilibrium constraints, with applications to optimization in IMRT Ottmar Cronie: TBA Fredrik Edelvik: EEG-based localization of the epileptic focus Thomas Galtier: Estimation of crossing intensity for Non-Gaussian Process Jonas Hagmar: Continuous-Discrete Identification Stefan Jakobsson: Approximation of antenna data based on rational radial basis function interpolation Alexandra Jauhiainen: Data analysis issues in connection with metabolic profiling Mats Jirstrand: Nonlinear Mixed effects Modeling using Stochastic Differential Equations Jenny Jonasson: Estimating diffusion coefficients from FRAP and NMRd measurements Karin Kraft: Adaptive finite element methods for optimal control of heavy vehicles Stig Larsson: Multiscale modelling of paper Carl Lindberg: Optimal trading in volatility Sara Lorén: Monte Carlo versus second moment evaluation of reliability, fatigue life example

7

Olle Nerman: Fundamental statistical problems with large population samples Thomas Svensson: A statistical view on possible errors in physical models – evaluation of a round robin exercise on fatigue life prediction Adam Wojciechowski: Optimization of maintenance operations

Christoffer Cromvik: Nonlinear Programming – Robust Models and Applications, September 25 Torben Hasenkamp: Designing for Robustness, March 13

Defences of licentiate theses GMMC meeting, November 26 Larisa Beilina: Verification of a globally convergent inverse algorithm on experimental data in the picosecond regime Uno Nävert: Mathematics as Technology

Defences of doctoral theses Fardin Saedpanah: The Finite Element Method for Fractional Order Viscoelasticity and the Stochastic Wave Equation May 26

8

Daniel Ahlberg: Asymptotics of firstpassage percolation on 1-dimensional graphs, January 14 Dmitrii Zholud: Contribution to Extreme Value Theory with Applications in Bioinformatics, March 30 Ali Mesforush: Finite element approximation of the linear stochastic CahnHilliard equation, May 27 Mao Wengang: Random Fatigue Analysis of Container Ship Structures, June 2

Optimisation and Modelling We describe below some achievements in the area of optimisation and modelling during 2009, in four active GMMC projects: three within mathematical modelling with medical applications; and one on the control of heavy vehicles.

Mathematical modelling with medical applications Team members: Mathematical Sciences, Chalmers University of Technology and University of Gothenburg: Mohammad Asadzadeh, Larisa Beilina, Christoffer Cromvik, Michael Patriksson Medical Radiophysics, Karolinska Institute: Anders Brahme 1. Picosecond scale experimental verification of a globally convergent algorithm for a coefficient inverse problem Recently (2008) Larisa Beilina and prof. M.V. Klibanov from University of North Carolina at Charlotte, Charlotte, USA, have developed a radically new globally convergent numerical method, whose general scheme works for a broad class of coefficient inverse problems with practical applications, for example in medical optical imaging and acoustic imaging, imaging of land mines, and airport security. In 2009 this method was combined with the adaptivity technique. In addition, this method was verified experimentally on the nanotechnology experiment at the Optical Center at University of North Carolina at Charlotte. The blind testing on real noisy data has demonstrated an excellent accuracy of results. To do this, an international interdisciplinary team of mathematicians of L. Beilina (Sweden) and M.V. Klibanov (USA), N. Pantong (USA), and physicists M.A. Fiddy (USA) and J. Schenk (USA) was arranged. 2. Mathematical modelling of particle transport with applications in radiation oncology In a collaboration with Karolinska Institue, mathematical modeling procedures were studied regarding charged particle transport in nonuniform media, studying in particular the energy deposition of high-energy (50-500 MeV) protons and high-energy electrons in energy ranges up to 50 MeV. This work is an extension of the bipartition model for highenergy electrons to the field of high-energy ions in inhomogeneous media with retained energy-loss straggling term. Convergence was established for semi-discrete and fully discrete approximations of such equations, for a broad beam model, using the standard Galerkin and streamline diffusion finite element methods. The analytic broad beam model of the light ion absorbed dose were compared with the results of the modified Monte Carlo (MC) code SHIELDHIT+ and those of Galerkin streamline diffusion approach. More specifically we compared the

9

1.6

SD 5.2 cm SD 6.5 cm SD 7.8 cm Bipartition 5.2 cm Bipartition 6.5 cm Bipartition 7.8 cm

Relative Energy Distribution

1.4

1.2

1

0.8

0.6

0.4

0.2

0

−0.2

0

5

10

15

20

25

30

MeV

1.2

SD 5.2 cm SD 6.5 cm SD 7.8 cm MC 5.2−5.6 cm MC 6.4−6.8 cm MC 7.6−8 cm

Relative Energy Distribution

1

0.8

0.6

0.4

0.2

0

−0.2

0

5

10

15

20

25

30

35

MeV

energy distribution in different depths, for 10-30 MeV ions and electrons, in water phantoms. Some comparison of the computed results in this part can be seen in Figs. 1 and 2. Related basic research work was also performed on elliptic PDEs relevant to diffusion transport processes; optimal finite element approximations were developed for neutral and charged (linear and nonlinear) particle transport equations. We also studied a globally convergent numerical method for a hyperbolic coefficient inverse problem with applications in image tracking in, e.g., radiation therapy. 3. Stochastic mathematical programs with equilibrium constraints, with applications to decentralized traffic control and intensity-modulated radiation therapy Modern engineering and medical optimisation problems often include the combination of simulation tools with efficient traditional engineering and medical techniques. This demands an efficient platform of optimisation tools that integrate the simulation models that together describe the system. Several GMMC projects are of this variety, where the core is in optimisation.

10

In many of these cases the simulation has an uncertain outcome because of inherent uncertainties in the phenomena modelled. Such is the case, in particular, with the clinical outcome of the treatment of an individual patient using intensity-modulated radiation therapy (IMRT); sources of uncertainty include the dose delivered by the machine to the patient’s organs-at-risk and the clinical targets, as well as the actual effect a given dose has upon a given individual. As the goals of the treatment are the cure of the patient as well as a high quality of life in the years following treatment, this uncertainty must be taken into account already in the construction of the treatment plan. A mathematical framework has been constructed for robust simulation-based optimization, where robust treatment plans in IMRT planning defines but one interesting problem instance. This framework, known as a stochastic mathematical program with equilibrium constraints (SMPEC), has recently been studied in detail by Cromvik and Pat-riksson. Among the results reached we mention (a) the robustness of globally optimal solutions as well as stationary solutions to changes in the probability distribution of the uncertain data; (b) the robustness of the corresponding model extensions arising when the standard min expected value objective is replaced by a risk objective, and when extending the problem to multiple objectives – the latter being of special interest in IMRT planning; and (c) the convergence of sample average approximation schemes for these problems. Algorithms have also been implemented and tested, in particular, in two IMRT applications.

Optimal control of heavy vehicles by means of an adaptive finite element method Team members: Mathematical Sciences, Chalmers University of Technology and University of Gothenburg: Karin Kraft, Stig Larsson Applied Mechanics, Chalmers University of Technology: Mathias Lidberg This research is concerned with numerical methods for optimal control of differential equations with applications in the automotive industry. This involves the development of theory and algorithms as well as application to anti-skid systems for heavy vehicles. The project is done in cooperation with Vehicle Safety at Chalmers. We develop numerical methods for solving the differential and algebraic equations that arise in connection with optimal control of the dynamics of heavy vehicles. The stiff boundary value problems that occur are usually solved by shooting methods. Our approach is to use adaptive finite element methods which allow error control and adaptive computational meshes. The algorithms are to be implemented in computer code and used in control systems for heavy vehicles. We have adapted the methodology of dual weighted residuals to optimal control problems. In our earlier work the algorithm was implemented for problems with quadratic a goal functional and linear state equation. During 2009 we have extended the theory and the implementation to fully nonlinear problems.

11

Risk, Reliability, Quality Core team members: Mathematical Sciences, Chalmers University of Technology and University of Gothenburg: Anastassia Baxevani, Carl Lindberg, Jacques de Maré, Holger Rootzén, Igor Rychlik Quality Sciences, Chalmers University of Technology: Bo Bergman University of Palermo and Chalmers: Stefano Barone Department of Shipping and Marine Technology, Chalmers: Jonas Ringsberg Department of Applied Mechanics, Chalmers: Jens Nielsen University of North Carolina at Chapel Hill: Ross Leadbetter Indian Institute of Technology, Madras: Sunetra Sarkar, Sayan Gupta Fraunhofer-Chalmers Centre for Industrial Mathematics: Pär Johannesson, Sara Lorén SP Technical Research Institute of Sweden: Thomas Svensson Volvo Car Corporation: Åke Lönnkvist Volvo Power Train: Per Johansson Coch-Lear: Martin Arvidsson Main application areas: – Reliability and Quality – Ocean Engineering – Management of Economic Risks

Reliability and Quality The book “Robust Design Methodology for Reliability” was completed and published by Wiley. A “Guide to load analysis for automotive applications” has been finished in cooperation with ITWM and distributed to our partners at DAF, Daimler, IVECO, MAN, SCANIA and VOLVO 3P. The guide will be published as a book within 18 months. These two books collect many of the main results of the four years of the GMMC reliability team. They give a comprehensive approach to all facets of a company’s work on reliability improvement. In order to discuss our ideas in the international community, we have organized the ENBIS conference at Chalmers and a work-shop in reliability in connection to the conference. We also organized a workshop “Fatigue design for safety” jointly with SP, Department of Applied Mechanics and Department of Shipping and Marine Technology at Chalmers. We have had a number of internal workshops at Varberg and Falkenberg where

12

we discussed and developed design practices for safety components. We worked with replacement strategies in maintenance, different reliability approaches and using interactions in robust design. To promote our ideas about variation and uncertainties as the root of unreliability and the remedy in robust design we have developed a course for industry which will be given in 2010.

Ocean Engineering A main direction is to give a basis for a ship routing program for minimizing risks of fatigue damage, excessive motion and fuel consumption. We have also started a study on ways to minimize CO2 emissions from ships. The work is in cooperation with Department of Shipping and Marine Technology, Chalmers and Den Norske Veritas and has recently been broadened to include also the Danish Technical University. A deeper investigation into fatigue processes in large container ships has been initiated. The two main questions are whether one has to employ multiaxial fatigue criteria in fatigue life prediction, and how to include models for material deterioration due to corrosion and fatigue into risk estimation. Our cooperation with the University of North Carolina and the US coast guard on prediction of risks of capsizing in heavy weather has been continued. Rychlik has become member of the committee on environment for the 18th International Ship and Offshore Structures Congress.

Management of Economic Risks GMMC research on economic risk is aimed at both insurance companies and financial institutions. We developed methods to predict the impacts of catastrophic risks and applied the methods to wind storm insurance at Länsförsäkringar. A new project with Folksam on realistic models for dependent insurance risks was started during the year. It has received additional funding, for an industrial PhD student, from the Swedish Science Research Council. A second direction was methods for optimal trading of financial instruments, in particular for volatility related strategies. The aim is to provide research support for the buy-side of the derivative markets. The buy-side constitutes the actual market, i.e. are the market participants such as pension funds. They are the price takers, and take a position because they feel it is cheap. They do not hedge to offset risk, as they believe they get compensated for taking that risk. However, little research has been aimed at buy-side trading and decisions of when and how to execute and liquidate the positions have to be made ad hoc. The results of erroneous handling of derivatives can be disastrous, which the present financial crisis illustrates.

13

Bioinformatics In bioinformatics the PhD student Alexandra Jauhiainen and her supervisor Olle Nerman have had support from GMMC during the year. Alexandra has spent a second period in a brittish cancer research lab in Cambridge with Simon Tavaré. There she has collaborated with other bioinformaticians in several projects related to micro arrays and metabolic profiling using NMR measurements on cancer cell line cultures. On the local level Alexandra (and Olle) collaborated with Pierre Åman at Sahlgrenska Academy in another cancer screening project. At least three publications is expected to be ready for the dissertation of Alexandra in autumn 2010. Olle, Alexandra and Janeli Sarv also has several running projects with Anders Blomberg and Per Sunnerhagen at CMB, University of Gothenburg related to yeast genomics. During the year dr Rebecka Jörnsten, newly recruited associate professor in Mathematical Statisticians, initiated several new local research collaborations e.g. with Lena Carlssons research group in obesity related metabolic diseases, and with assistant professor Sven Nelanders reasech group centered around cancer systems biology.

Collaboration with CMR The collaboration with the Sahlgrenska Center for Cardiovascular and Metabolic Research (CMR) continues with the research student Martin Berglund, who is working on a nonlinear mixed effects model for data from a clinical study of the lipoprotein metabolism in human subjects. The work is concerned with the implementation of the technique, as well as its application on real data. One result of this study is that a compartmental model with a smaller number of compartments could represent the experimental data equally well.

Systems biology The research in systems biology within GMMC at the Fraunhofer-Chalmers Centre has been carried out in three areas: continuous-discrete identification with applications in systems biology, computational modeling of atrial fibrillation, and bioimaging software development with applications to quantitative yeast cell studies. The objective in continuous-discrete identification is to estimate parameters in a continuous time dynamic model using discrete time measurement data – a common situation found in many applied systems biology projects. We have developed a computational framework in terms of a Mathematica package, which features functions for model structure specification, measurement data management, sensitivity analysis, and parameter estimation. A major objective in this project is to employ so called prediction error minimization methodology to take uncertainties in both measurements and model structure into account when estimating parameters. The current computational framework has now

14

reached a level where it will give good support for method development along these lines. This project is partly carried out as a PhD-project and is a part of the EC project UNICELLSYS. We have continued our efforts in computational modeling of atrial fibrillation. Currently we simulate fibrillation and flutter like electro-dynamic activity in cell networks from simple sheets up to realistic atrial geometries. In addition, the effect of ion-channel modulation of this behavior has been investigated. The simulations are in good accordance with in vivo observations, have great potential to provide insights into the underlying mechanisms of atrial fibrillation and flutter, and can serve as a tool for prediction of drug effects. This project is part of a series of projects carried out in collaboration with AstraZeneca, which have mainly been funded by AstraZeneca and SSF. We has continued the development of the CellStat research software for quantitative yeast cell studies. CellStat is a tool equipped with graphical user-interface (GUI) for automated recognition and tracking of yeast cells from transmission microscope images, combined with quantification and localization of GFP-labeled proteins using fluorescence microscopy. The emphasis on the algorithms in CellStat lies on robust methodologies which enables long time-lapse studies of protein localization, migration, and inheritance over several cell cycles, as well as high through-put screening of protein functionality of a large number of gene-disrupted cells. The most recent study involves tracking of 50–100 cells in image sequences consisting of up to 1800 images for several different cell cultures. For each single cell, the shuttling of tagged proteins between cell nucleus and cytoplasm is measured and from this data, conclusions on cell stress levels can be drawn. Single cell studies like these are crucial for proper understanding of cell functionality and an automated image analysis tool like CellStat is a key component in extracting data from microscopy. This project is a collaboration with the Bionanophotonics group at Applied Physics at Chalmers and apart from the GMMC funding partly financed by a grant from the Swedish Research Council.

15

Publications Published or accepted journal articles M. Asadzadeh, A. Schatz and W. Wendland: A non-standard approach to Richardson extrapolation in the finite element method for second order elliptic problems. Math. Comp. 78, 1951–1973 M. Asadzadeh: On convergence of a h-p Streamline Diffusion and Discontinuous Galerkin Methods for the Vlasov-Poisson-Fokker-Planck System. Proceedings of 26th International Symposium on Rarefied Gas Dynamics (RGD26), Kyoto, Japan, July 20–25 2008. American Institute of Physics (AIP), 1084, 99–104 Marina Axelson-Fisk and Olle Häggström: Conditional percolation on onedimensional lattices. Advances in Applied Probability, 41 (4) Marina Axelson-Fisk and Olle Häggström: Biased random walk in a one-dimensional percolation model. Stochastic processes and their Applications, 119 (10), 3395–3415 Anastassia Baxevani, S. Caires and Igor Rychlik: Spatio-temporal statistical modelling of significant wave height. Environmetrics, 20 (1), 14–31 Anastassia Baxevani and Krzysztof Podgórski: Lamperti transform of fractional Brownian motion. Acta Physica Polonica B, 40 (5), 1395–1435 L. Beilina, M. P. Hatlo and H. E. Krogstad: Adaptive algorithm for an inverse electromagnetic scattering problem. J. Applicable Analysis, V.88 nr.1, 15–28

16

L. Beilina and Michael V. Klibanov: Synthesis of global convergence and adaptivity for a hyperbolic coefficient inverse problem in 3D. Accepted for publication in J. Inverse and Ill-posed problems A. Bengtsson and I. Rychlik: Uncertainty in fatigue life prediction of structures subject to Gaussian loads. Prob. Eng. Mech., Vol. 24, 224–235 A. Bengtsson, K., Bogsjö and I. Rychlik: Uncertainty of estimated rainflow damage for random loads. Marine Structures, Vol. 22, 261–274 Bo Bergman: Conceptualistic Pragmatism: A framework for Bayesian analysis? IIE Transactions, 41:86–93 K. Bogsjö and I. Rychlik: Vehicle fatigue damage caused by road irregularities. Fatigue and Fracture of Engineering Materials and Structures, Vol. 32, 391–402 R. Butler, U. Machado and I. Rychlik: Distribution of wave crests in nonGaussian sea. Applied Ocean Research, Vol. 31, 57–64 Maria Daneva, Torbjörn Larsson, Michael Patriksson and Clas Rydergren: A comparison of feasible direction methods for the stochastic transportation problem. Computational Optimization and Applications, to appear Stefanie Eckel, Frank Fleischer, Pavel Grabarnik, Marian Kazda, Aila Särkkä and Volker Schmidt: Modelling tree roots in mixed forest stands by inhomogeneous marked Gibbs point processes. Biometrical Journal, 51, 522–539 L. Fernandez-Ricaud, Daniel Dalevi, A. Blomberg, Olle Nerman: Testing of

Chromosomal Clumping of Gene Properties. Statistical Applications in Genetics and Molecular Biology, 8 (1), 19 Matthias Geissert, Mihaly Kovacs and Stig Larsson: Rate of weak convergence of the finite element method for the stochastic heat equation with additive noise. BIT Numerical Mathematics, 49 (2), 343–356 Pavel Grabarnik and Aila Särkkä: Modelling the spatial structure of forest stands by multivariate point processes with hierarchical interactions. Ecological Modelling, 220, 1232–1240 Gilles Guillot, N. Lorén and Mats Rudemo: Spatial prediction of weed intensities from exact count data and imagebased estimates. Journal of the Royal Statistical Society Series: C (Applied Statistics), 58, 525–542 Torben Hasenkamp, Martin Arvidsson and Ida Gremyr: A review of practices for robust design methodology. Journal of Engineering Design, 20 (6), 645–657 Olle Häggström and Peter Mester: Some Two-dimensional Finite Energy Percolation Processes. Electronic Communications in Probability, 14, 42–54 Peter Jagers, Kais Hamza, Aidan Sudbury and Daniel Tokarev: The mixing advantage is less than 2. Extremes, 12 (1), 19–31 Peter Jagers: Extinction. Math. Pop Studies, 16 (3), 175–176 Jagers, Peter: Matematikens ord. LexicoNordica, 16, 315–318 Peter Jagers and Karin C. Hårding: Viability of small populations experiencing

recurring catastrophes. Math. Pop Studies, 16 (3), 177–198 S. Jakobsson, M. Patriksson, J. Rudholm and A. Wojciechowski: A method for simulation based optimization using radial basis functions, Optimization & Engineering, to appear S. Jakobsson, M. Saif-Ul-Hasnain, R. Rundqvist, F. Edelvik, B. Andersson, M. Patriksson, M. Ljungqvist, D. Lortet and K. Wallesten: Combustion engine optimization: A multiobjective approach, Optimization & Engineering, to appear S. Jakobsson, B. Andersson and F. Edelvik: Rational radial basis function interpolation with applications to antenna design. Journal of Computational and Applied Mathematics, Vol. 233(4), 889–904 Margareta Jernås, Bob Olsson, Kajsa Sjöholm, Anders Sjögren, Mats Rudemo, Bengt Nellgård, Lena Carlsson and Carl David Sjöström: Changes in adipose tissue gene expression and plasma levels of adipokines and acute-phase proteins in patients with critical illness. Metabolism: clinical and experimental, 58 (1), 102–8 Pär Johannesson, Thomas Svensson, Leif Samuelsson, Bo Bergman and Jacques de Maré: Variation Mode and Effect Analysis: An Application to Fatigue Life Prediction, Quality and Reliability Engineering International, Volume 25, Issue 2, 167–179 Erik Kristiansson, Michael Thorsen, Markus J. Tamás and Olle Nerman: Evolutionary Forces Act on Promoter Length: Identification of Enriched CisRegulatory Elements. Molecular Biology and Evolution, 26 (6), 1299–1307

17

C. Kühn, K. V. S. Prasad, E. Klipp, Peter Gennemark: Formal Representation of the High Osmolarity Glycerol Pathway in Yeast Genome Informatics, 22, 69–83 Mats Kvarnström, N. Lorén, Magnus Nydén and A. Westergard: Brownian dynamics simulations in hydrogels using an adaptive time-stepping algorithm. Physical Review E, 79 (1), Article Number: 016102 Stig Larsson and Fardin Saedpanah: The continuous Galerkin method for an integro-differential equation modeling dynamic fractional order viscoelasticity. IMA J. Numer. Anal. Carl Lindberg: Portfolio optimization when expected stock returns are determined by exposure to risk. Bernoulli, 15 (2), 464–474 Christina Mauléon and Bo Bergman: Exploring the epistemological origins of Shewhart’s and Deming’s theory of quality: influences from C.I. Lewis’ conceptualistic pragmatism. International Journal of Quality and Service Sciences Vol. 4, No 1, 77–89 Claes Molin, Alexandra Jauhiainen, Jonas Warringer, Olle Nerman, Per Sunnerhagen: mRNA stability changes precede changes in steady-state mRNA amounts during hyperosmotic stress. RNA, 15, 600–614 Hendry Raharjo, Aarnout Brombacher, Thong-Ngee Goh and Bo Bergman: On integrating Kano’s model dynamics into QFD for multiple product design. Quality and Reliability Engineering International (in-press) Claudia Redenbach, Aila Särkkä, Katja Schladitz and Johannes Freitag: Anisot-

18

ropy analysis of pressed point processes. Advances in Statistical Analysis, 93 (3), 237–261 P. Rohde, D. Drobin, R. G. Hahn, B. Wennberg, C. Lindahl, F. Sjöstrand, C. Svensen: Modelling of peripheral fluid accumulation after a crystalloid bolus in female volunteers – a mathematical study. Computational and Mathematical Methods in Medicine, accepted for publication Holger Rootzén and Erik Brodin: Univariate and Bivariate GPD Methods for Predicting Extreme Wind Storm Losses. Insurance: Mathematics and Economics, 44 (2), 345–356 Holger Rootzén, Anne-Laure Fougeres and Johan Nolan: Models for Dependent Extremes Using Stable Mixtures. Scandinavinan Journal of Statistics, 36, 42–59 Holger Rootzén: Weak convergence of the tail empirical function for dependent sequences. Stoch. Proc. Appl., 119 (2), 468–490 Selpi, Christopher H. Bryant, Graham J. L. Kemp, Janeli Sarv, Erik Kristiansson, Per Sunnerhagen: Predicting functional upstream open reading frames in Saccharomyces cerevisiae. BMC Bioinformatics, 10, 451 Elin Svensson, Thore Berntsson, Michael Patriksson and Ann-Brith Strömberg: An optimization methodology for identifying robust process integration investments under uncertainty, Energy Policy, vol. 37, no. 2, 680–685 Elin Svensson, Thore Berntsson and Ann-Brith Strömberg: Benefits of using an optimization methodology for identi-

fying robust process integration investments under uncertainty – A pulp mill example. Energy Policy, 37 (3), 813–824 Kaisa Thorell, Annika Bergman, Helena Carén, Staffan Nilsson, Per Kogner, Tommy Martinsson and Frida Abel: Verification of genes differentially expressed in neuroblastoma tumours: a study of potential tumour suppressor genes. BMC medical genomics, 2 (1), 53 Michael Thorsen, Gabriel G. Perrone, Erik Kristiansson, Mathew Traini, Tian Ye, Ian W. Dawes, Olle Nerman and Markus J. Tamás: Genetic basis of arsenite and cadmium tolerance in Saccharomyces cerevisiae. BMC genomics, 10, 105 Stefano Trazzi, Lorenzo Pareschi and Bernt Wennberg: Adaptive and recursive time relaxed Monte Carlo methods for rarefied gas dynamics. SIAM Journal on Scientific Computing, 31 (2), 1379–1398 Dag Wedelin and Peter Gennemark: Benchmarks for identification of ordinary differential equations from time series data. Bioinformatics, 25 (6), 780–786 Bernt Wennberg and Mattias Sundén: The Kac master equation with unbounded collision rate. Markov Processes and Related Fields, 15 (2), 125–148 G. M. Marchetti and Nanny Wermuth: Matrix representations and independencies in directed acyclic graphs. Annals of Statistics, 37, 961–978 Nanny Wermuth, G. M. Marchetti and David R. Cox: Triangular systems for symmetric binary variables. Electronic Journal of Statistics, 3, 932–955

Annica Wilzén, Staffan Nilsson, RoseMarie Sjöberg, Per Kogner, Tommy Martinsson and Frida Abel: The Phox2 pathway is differentially expressed in neuroblastoma tumors, but no mutations were found in the candidate tumor suppressor gene PHOX2A. International journal of oncology, 34 (3), 697–705 Adam Wojciechowski, Michael Patriksson, Johan Rudholm and Stefan Jakobsson: A method for simulation based optimization using radial basis functions. Optimization & Engineering J. Xin, L. Beilina and Michael V. Klibanov: Globally convergent numerical methods for coefficient inverse problems for imaging inhomogeneities. J. Computing in Science and Engineering (CISE), accepted for publication S. Åberg, K. Podgórski and I. Rychlik: Fatigue damage assessment for a spectral model of non-Gaussian random loads, Prob. Eng. Mech., Vol 24, 608–617

Published or accepted books and book articles S. Barone, M. Giorgio, M. Guida and G. Pulcini: Stochastic modeling and prediction of catalytic converters degradation. In: Safety, Reliability and Risk Analysis: Theory, Methods and Applications. Martorell et al. (eds), Taylor & Francis S. Barone and A. Lanzotti A: Robust Ergonomic Virtual Design. In: Statistics for Innovation – Statistical Design of “continuous” product innovation. Ed. P. Erto, Springer A. Bengtsson, K. Bogsjö and I. Rychlik: Fatigue Damage Uncertainty. In: Robust

19

Design Methodology for Reliability, Wiley, 151–171 B. Bergman, S. Lorén, J. de Maré and T. Svensson, eds: Robust Design Methodology for Reliability: Exploring the Effects of Variation and Uncertainty, Wiley Alexander Chakhunashvili, Stefano Barone, Per Johansson and Bo Bergman: Robust Product Development Using Variation Mode and Effect Analysis. In: Robust Design Methodology for Reliability: Exploring the Effects of Variation and Uncertainty, Wiley, 57–70 Olle Häggström: Vem kan man lita på? In: Osäkrat klimat – laddad utmaning, 149–162

Published or accepted conference proceedings papers M. Asadzadeh and P. Kowalczyk: Convergence of a mixed discontinuous Galerkin and finite volume scheme for the 3 dimensional Vlasov–Poisson–Fokker–Planck system. Proceedings of 8th Enumath, Uppsala University M. Asadzadeh and L. Thevenot: On discontinuous Galerkin and discrete ordinates approximations for neutron transport equation and the critical eigenvalue. Proceedings of ICTT21, Polytecnico di Torino S. Barone and B. Bergman: Statisticalbased tolerance setting by eliciting the loss function reasoning. Proceedings 9th ENBIS conference, Gothenburg, September 20–24 S. Barone, B. Bergman, A. Chakhunashvili and P. Johansson: Focus on failure

20

avoidance and risk reduction through Variation Mode and Effect Analysis (VMEA). Proceedings 2009 Spring Research Conference on Statistics in Industry and Technology, Coquitlam, BC, Canada, May 27–29 F. Besnard, M. Patriksson, A.-B. Strömberg, A. Wojciechowski and L. Bertling: An optimization framework for opportunistic maintenance of offshore wind power systems. Proceedings of 2009 IEEE Bucharest PowerTech Conference, June 28–July 2 K. Bogsjö and I. Rychlik: Uncertainty of estimated damage for random loads. Presented at Second Int. Conf. on Variable Amplitude Loading, Darmstadt, Germany, March Fredrik Edelvik, Björn Andersson, Stefan Jakobsson, Stig Larsson, Mikael Persson and Yazdan Shirvany: An improved method for dipole modeling in EEG-based source localization. World Congress of Medical Physics and Biomedical Engineering, Vol. 25/IX, 146– 149, München, Germany, September Peter Gennemark and Dag Wedelin: Improved Parameter Estimation for Completely Observed Ordinary Differential Equations with Application to Biological Systems. Computational Methods in Systems Biology, LNCS 5688, 205–217 S. Jakobsson, B. Andersson and F. Edelvik: Approximation of Antenna Data with Rational Radial Basis Function Interpolation. International Conference of Numerical Analysis and Applied Mathematics, Vol. 1, 473–477, Crete, Greece, September

Svante Lifvergren, Alexander Chakhunashvili, Ida Gremyr, Bo Bergman, Andreas Hellström: A Large Scale Implementation of Six Sigma at the Skaraborg Hospital Group. 16th International Annual EurOMA Conference Svante Lifvergren, Alexander Chakhunashvili, Bo Bergman, Peter Docherty: Online statistical monitoring of critical patient data increases patient safety. International PICMET Conference, July 28–30, Cape Town, RSA P. Johannesson and T. Svensson: A loadstrength model for fatigue applications. Presented at the international conference ENBIS9, arranged by the European Network for Business and Industrial Statistics, Gothenburg, September 20–24 S. Lorén and T. Svensson: Monte Carlo versus second moment evaluation in reliability, a fatigue life example. Presented at the international conference ENBIS9, arranged by the European Network for Business and Industrial Statistics, Gothenburg, September 20–24 Å. Lönnqvist and T. Svensson: Chance, assignable causes, and statistical control in reliability. Presented at the international conference ENBIS9, arranged by the European Network for Business and Industrial Statistics, Gothenburg, September W. Mao, J. Ringsberg, I. Rychlik and G. Storhaug: Comparison between a fatigue model for voyage planning and measurements of a container vessel. Proceedings of the ASME 28th International Conference on Ocean, Offshore and Arctic Engineering (OMAE2009), Hawaii, USA, May 31–June 5

J. Nilsson, M. Patriksson, A.-B. Strömberg, A. Wojciechowski and L. Bertling: An opportunistic maintenance optimization model for shaft seals in feed-water pump systems in nuclear power plants. Proceedings of the 2009 IEEE Bucharest PowerTech Conference, June 28–July 2 Johan Sandström and Jacques de Maré: Probability of subsurface fatigue initiation in rolling contact. Proceedings of the 8th International Conference on Contact Mechanics and Wear of Rail/Wheel Systems, 2, 419–424 A. Stjernman, A. Derneryd, S. Jakobsson, B. Andersson and F. Edelvik: Multi-objective Optimization of MIMO Antenna System. Proceedings from 3rd European Conference on Antennas and Propagation, Berlin, Germany, March T. Svensson, M. Karlsson, B. Johannesson, P. Johannesson and J. de Maré: Predictive Safety Index for Variable Amplitude Fatigue Life. Presented at the conference Material and Component Performance under Variable Amplitude Loading, Darmstadt, Germany, March

Submitted articles or reports Pierre Ailliot, Anastassia Baxevani, Anne Cuzol, Valerie Monbet and Nicolas Raillard: Space-time models for moving fields with an application to significant wave height fields. Preprint 2009:39, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg T. Almgren, N. Andreasson, M. Patriksson, A.-B. Strömberg and A. Wojciechowski: The replacement problem: A polyhedral and complexity analysis, submitted to Mathematical Programming

21

M. Asadzadeh, A. Brahme and J. Kempe: Ion transport in inhomogeneous media ased on the bipartition model for primary ions. Preprint 2009:12, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Mohammad Asadzadeh, Anders Brahme and J. Xin: Ion Transport in Inhomogeneous Media, II. Galerkin Methods for Primary Ions. Preprint 2009:14, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Mohammad Asadzadeh and Piotr Kowalczyk: Convergence of a mixed discontinuous Galerkin and finite volume scheme for the 3 dimensional Vlasov-PoissonFokker-Planck system. Preprint 2009:6, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Mohammad Asadzadeh and L. Thevenot: On discontinuous Galerkin and discrete ordinates approximations for neutron transport equation and the critical eigenvalue. Preprint 2009:48, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Anastassia Baxevani, Oskar Hagberg and Igor Rychlik: Note on the distribution of extreme wave crests. Preprint 2009:41, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Larisa Beilina and Michael V. Klibanov: A Globally Convergent Numerical Method and Adaptivity for a Hyperbolic Coefficient Inverse Problem. Preprint 2009:6, Department of Mathematical Sciences,

22

Chalmers University of Technology and University of Gothenburg Larisa Beilina and Michael V. Klibanov: A globally convergent numerical method and the adaptivity technique for a hyperbolic coefficient inverse problem. Part I: analytical study. Preprint 2009:32, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Larisa Beilina and Michael V. Klibanov: A globally convergent numerical method and the adaptivity technique for a hyperbolic coefficient inverse problem. Part II: numerical studies. Preprint 2009:33, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Larisa Beilina and Michael V. Klibanov: Synthesis of global convergence and adaptivity for a hyperbolic coefficient inverse problem in 3D. Preprint 2009:11, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg L. Beilina, M. V. Klibanov and M. Yu. Kokurin: Adaptivity with relaxation for ill-posed problems and global convergence for a coefficient inverse problem. Preprint 2009:47, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Larisa Beilina and Michael V. Klibanov: A posteriori error estimates for the adaptivity technique for the Tikhonov functional and global convergence for a coefficient inverse problem. Preprint 2009:49, Department of Mathematical Sciences, Chalmers University of Technology and Gothenburg University

W. Chen and W. Duan: Computational Aspects of Mathematical Models in Image Compression. Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Christoffer Cromvik and Michael Patriksson: On the robustness of global optima and stationary solutions to stochastic mathematical programs with equilibrium constraints. Preprint 2009:2, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Christoffer Cromvik and Peter Lindroth: Robust multi-objective optimization based on a user perspective. Preprint 2009:31, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg M. Daneva, M. Göthe-Lundgren, T. Larsson, M. Patriksson and C. Rydergren: A sequential linear programming algorithm with multi-dimensional search-Derivation and convergence, submitted to Optimization Methods and Software Thomas Galtier: Note on the Estimation of Crossing Intensity for Laplace Moving Average. Preprint 2009:26, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg T. Galtier, S. Gupta and I. Rychlik: Crossings of Second-order Response Processes Subjected to LMA Loadings. Preprint 2009:40, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Alexander Herbertsson, Jiwook Jang and Torsten Schmidt: Pricing basket default

swaps in a tractable shot-noise model submitted Mihaly Kovacs, Stig Larsson and Fardin Saedpanah: Finite Element Approximation of the Linear Stochastic Wave Equation with Additive Noise. Preprint 2009:18, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Mihaly Kovacs, Stig Larsson and Fredrik Lindgren: Weak convergence of finite element approximations of linear stochastic evolution equations with additive noise. Preprint 2009:37, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg Sig Larsson and Ali Mesforush: Finite Element Approximation of the Linearized Cahn-Hilliard-Cook Equation. Preprint 2009:22, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg P. Lindroth, M. Patriksson and A.-B. Strömberg: Approximating the Pareto optimal set using a reduced set of objective functions, revised for European Journal of Operational Research I. Rychlik, J. Rydén and C. Andersson: Estimation of return values for significant wave height from satellite data, to appear in Extremes Fardin Saedpanah: Existence and Uniqueness of the Solution of an Integro-Differential Equation with Weakly Singular Kernel. Preprint 2009:16, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg

23

Fardin Saedpanah: A Continuous SpaceTime Finite Element Method for an Integro-Differential Equation Modeling Dynamic Fractional Order Viscoelasticity. Preprint 2009:17, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg E. Svensson, A.-B. Strömberg and M. Patriksson: Optimal investment planning of industrial energy efficiency measures under uncertainty, submitted to OR Spektrum J. Xin, L.Beilina and Michael V. Klibanov: Globally convergent numerical methods for coefficient inverse problems for imaging inhomogeneities. Preprint 2009:34, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg

ticity and the Stochastic Wave Equation Supervisor: Stig Larsson, Mathematical Sciences

Licentiate theses Daniel Ahlberg: Asymptotics of firstpassage percolation on 1-dimensional graphs Supervisor: Olle Häggström, Mathematical Sciences Wengang Mao: Random Fatigue Analysis of Container Ship Structures Supervisor: Igor Rychlik, Mathematical Sciences Ali Mesforush: Finite element approximation of the linear stochastic CahnHilliard equation Supervisor: Stig Larsson, Mathematical Sciences

Dmitrii Zholud: Contribution to Extreme Value Theory with Applications in Bioinformatics. Preprint 2009:8, Department of Mathematical Sciences, Chalmers University of Technology and University of Gothenburg

Dmitrii Zholud: Contribution to Extreme Value Theory with Applications in Bioinformatics Supervisor: Holger Rootzén, Mathematical Sciences

PhD theses

Master theses

Christoffer Cromvik: Nonlinear Programming – Robust Models and Applications Supervisor: Michael Patriksson, Mathematical Sciences

Fredrik Andersson, Engineering Mathematics: Surface interpolation for detail restoration Supervisor: Bo Johansson

Torben Hasenkamp: Designing for Robustness Supervisor: Bo Bergman, Quality Sciences

Erik Holmgren, Engineering Mathematics: Risk indices for the estimation of uncertainty in wind power predictions based on ensembles of numerical weather predictions Supervisor: Igor Rychlik

Fardin Saedpanah: The Finite Element Method for Fractional Order Viscoelas-

24

Roland Jakobsson, Industrial ecology – for a sustainable society and Mattias Olsson, Complex adaptive systems: Real World Fuel Consumption of Buses – Prediction for Auxiliary Units Using Multivariate Regression Methods Supervisor: Jacques de Maré Zeynep Kalender, Bioinformatics and Systems Biology: Dicovery of commonly expressed genes in Ewing’s Sarcoma and Myxoid Liposarcoma. A metaanalysis study of publicly available gene expression data Supervisor: Olle Nerman Samuel Lorin, Engineering Mathematics: Geometrisk variation inom formsprutning Supervisor: Igor Rychlik

Yun Niu, Engineering Mathematics: How much are Whippings Contributing to Fatigue and Extreme Responses in Ship Structure Details Supervisor: Igor Rychlik José Sanches Lopez, Engineering Mathematics: Vehicle Damage Prediction from Advanced and Simple Systems Measurements Supervisor: Jacques de Maré David Sjögren, Engineering Mathematics: Statistical methods for improving surrogate models in antenna optimization Supervisor: Stefan Jakobsson, FCC Examiner: Tommy Norberg, Mathematical Sciences Tobias Österlund, Bioinformatics and Systems Biology: A novel framework for cross-species gene expression analysis Supervisor: Olle Nerman

25

Financial Report - GMMC 2009 in 1000 SEK

Income from: The Swedish Foundation for Strategic Research Chalmers University of Gothenburg

Total income

5644 600 71

1)

6315

Expenses Personnel expenses: Reliability and risk Biomathematics Optimisation and modelling Management and coordination Project initiation Rendering our results Fellowships Total personnel expenses Travelling and accomodation for guests Workshops in Gothenburg Travelling, Conference attendance Equipment Miscellaneous Total other expenses

University overhead for university administration, computer support, premises etc.

Total expenses

1170 969 1142 504 50 175 50 4060 86 358 135 25 88 692

1563

6315

1) University of Gothenburg has supported associated professor with 327 000 SEK during 2009.

 

GMMC Gothenburg Mathematical Modelling Centre Department of Mathematical Sciences Chalmers University of Technology and University of Gothenburg SE-412 96 Gothenburg Sweden http://www.chalmers.se/math/EN/research/gmmc