Agent based model of fisher behavior as a dynamic constraint resolution problem

An agent-based approach for modelling fisher behavior as a dynamic constraint resolution problem is proposed. The fishers are modeled as agents tasked with optimizing different multi-objective utility functions over a search space subjected to ecological, social, and political constraints derived from existing ecological and social models. The agents search for a satisfactory strategy by using a guided local search algorithm modified to allow for competition or cooperation in varying degrees, and the utility function is modified to mimic perfect rationality, as well as to include well-known behavioral strategies such as repetition, imitation, and social comparison. The goal of the model is to allow analysis and comparison of fisher strategies and their impact on the environment under different ecological limitations, fishing policies and assumptions of rationality on the part of the fishers.