Developing Methods for Modelling Procedures in System Analysis and System Dynamics

Detta är en avhandling från Department of Chemical Engineering, Lund University

Sammanfattning: System Thinking is the mindset of defining and confining a problem and its symptoms as well as a method for communicating system understanding. System Analysis is taking that problem apart to understand its causalities and structural arrangement. System Dynamics is the use of the results of System Analysis in order to reconstruct the system of causalities. Modelling is used for understanding complex problems and developing problem solving strategies. There is a need to use a consistent procedure when communicating and using models. This thesis shows that a modelling process with an aim to solve a problem always starts by asking a set of questions. The questions define how the system boundaries will be set and what level of details will be considered. A theoretical model is the conceptual analysis, a mental model, of the problem that is developed in the process from asking the questions to constructing Causal Loop Diagrams. A Causal Loop Diagram is a structural diagram that ex-plains how different variables/parameters are arranged through causalities and feedbacks. The numerical model is the translation of the mental model into a numerical equation that can ei-ther be created through a System Dynamic software tool or programmed in code. The conceptual or theoretical model always precedes the numerical model. These two combined form a Model Utilisation process which consists of; 1) all required functions needed to produce the results during the whole modelling procedure and, 2) the supporting modules of preparing input data and interpreting the outputs. Group Model building with stakeholders is essential for a successful modelling process. This thesis shows that group modelling involves four innovation phases; Definition, Clarification, Confirmation and Implementation. The Group Model building utilises an iterative learning process, a so-called Learning Loop, for finding the optimal adaptive method for solving a complex problem. Through the Learning Loop the System Dynamic approach becomes an adaptive learning process where the four innovation phases emerge as inherent parts of the process. The adaptive learning behaviour integrates individual skills and knowledge to make collective skills and knowledge operational through group modelling. Transparency in the modelling process which involves careful documentation through all steps will enhance the understanding of the behaviour of the problem thus lead to discoveries of mechanisms that might otherwise have been overlooked. The final aim of the group modelling is to build a model with optimal complexity which answers the question with the smallest amount of com-ponents but with sufficient performance. Papers included in the thesis PAPERS ON THEORY Haraldsson HV and Sverdrup HU. 2004. Finding Simplicity in complexity in biogeochemical modelling. In Wainwright J and Mulligan M. Environmental Modelling: Finding Simplicity in Complexity . Wiley: New York: 211-223. Haraldsson HV and Sverdrup HU. 2005. Making art into a science: Adaptive learning behaviour and managing group modelling innovation processes. (Submitted) Haraldsson HV and Sverdrup H. 2005. On aspects of System Dynamic workflow. (Manuscript) Haraldsson HV, Sverdrup HU, Belyazid S, Holmquist J and Gramstad R. 2005. The Tyranny of Small Steps I: Discovery of an archetypical behaviour. (Submitted) PAPERS ON CASE STUDIES Haraldsson HV, Sverdrup H and Rannhagen U. 2001. Is Eco-living more Sustainable than Conventional Living? Comparing Sustainability Performances between Two Townships in Southern Sweden . Journal of Environmental Planning and Management 44(5): 663-681. Ólafsdóttir R, Schlyter P and Haraldsson HV. 2001. Simulating Icelandic vegetation cover during the Holocene. Implications for long-term land degradation . Geografiska Annaler 83A(4): 203-215. Haraldsson HV and Ólafsdóttir R. 2003. Simulating the dynamics in vegetation cover with regards to long-term climatic variations in sub-arctic landscapes . Global and Planetary Change 38: 313-325. Haraldsson HV and Ólafsdóttir R. 2005. A modelling approach for evaluate the pre-industrial natural carrying capacity of human population in Iceland. (Manuscript) Haraldsson HV, Sverdrup H, Belyazid S, Sigurðsson BD and Halldórsson G. 2005. The System Analysis process preparing for assessment of effects of afforestation in Iceland. (Manuscript) Haraldsson HV. 2004. Introduction to Systems Thinking and Causal Loop Diagrams. Reports in Ecology and Environmental Engineering. Department of Chemical Engineering: Lund: 49. 3rd Edition