The application of software engineering principles towards developing an agent based economy climate model

This project attempts to apply the tools of Computer Science to the domain of integrated Economy-Climate modelling. As such the Agent Based Economy Climate (ABEC) model was built, creating an entirely agent based economic subsystem which was paired with an existing system dynamics climate model. Computer Science tools used include software engineering elements such as following a conceptual framework and using rigorous Unified Modelling Language (UML) to aide in model driven development. These conceptual elements were then implemented using design patterns such as an adapter pattern and a modified version of the command pattern to ensure that the developed model is easily expandable, extendable and reusable. The model also leverages a simple evolutionary algorithm in one of the subsections to demonstrate the power these techniques can bear in other domains. The developed simulation model was successfully validated against an established system dynamics model. Results also showed the evolu¬tionary algorithm had successfully improved the fitness of the evolved population. However the social aspect of the model was resulting in a level of homogeneity of consumers which was not credible, demon¬strating areas where improvements could be made to the simulation model