Human behaviour modelling in complex socio-technical systems : an agent based approach

For many years we have been striving to understand human behaviour and our interactions with our socio-technological environment. By advancing our knowledge in this area, we have helped the design of new or improved work processes and technologies. Historically, much of the work in analysing social interactions has been conducted within the social sciences. However, computer simulation has brought an extra tool in trying to understand and model human behaviours. Using an agent based approach this presentation describes my work in constructing computational models of human behaviour for informing design through simulation. With examples from projects in two main application areas of crisis and emergency management, and energy management I describe how my work addresses some main issues in agent based social simulation. The first concerns the process by which we develop these models. The second lies in the nature of socio-technical systems. Human societies are a perfect example of a complex system exhibiting characteristics of self-organisation, adaptability and showing emergent phenomena such as cooperation and robustness. I describe how complex systems theory may be applied to improve our understanding of socio-technical systems, and how our micro level interactions lead to emergent mutual awareness for problem-solving. From agent based simulation systems I show how context awareness may be modelled. Looking forward to the future, I discuss how the increasing prevalence of artificial agents in our society will cause us to re-examine the new types of interactions and cooperative behaviours that will emerge.