Repository | Book | Chapter
(2015) Artificial life and computational intelligence, Dordrecht, Springer.
Agent-based simulation of stakeholder behaviour through evolutionary game theory
Yngve Svalestuen , Pinar Öztürk , Axel Tidemann , Rachel Tiller
pp. 100-111
Aquaculture organizations establish facilities at the coast in Frøya, Norway. The facilities block the surrounding area from fishing and cause environmental damage to close natural resources. Fishers who depend on those natural resources get the opportunity to influence the aquaculture expansion through complaints about the municipality's coastal plan. Statistics show that fishers don't complain as much as expected. This work aims to investigate why. An agent-based simulation is developed in order to model the fishers as intelligent agents with complex interaction. Fishermen's decision making is simulated through an artificial neural network which adapts its behavior (i.e. weights) by "learning-by-imitation", a method in evolutionary game theory, from other stakeholders' behavior in the environment. The promising results show that with further development the simulation system may be part of a decision support system that promotes policies that are fair for the stakeholders.
Publication details
DOI: 10.1007/978-3-319-14803-8_8
Full citation:
Svalestuen, Y. , Öztürk, P. , Tidemann, A. , Tiller, R. (2015)., Agent-based simulation of stakeholder behaviour through evolutionary game theory, in M. Randall (ed.), Artificial life and computational intelligence, Dordrecht, Springer, pp. 100-111.
This document is unfortunately not available for download at the moment.