Intelligent system for fish stock prediction and allowable catch evaluation

The paper describes an intelligent system for evaluation and prediction of marine fish stocks and determination of allowable quotas. The system can forecast population abundance and biomass, possible total allowable catch, determine the quota allocation to subjects of fishery, based on various criteria. As knowledge sources, the system uses the results of long-term observations and research, regular dependencies established by experts, and human experts themselves. As initial data, monitoring and sampling results are used. Practical use of the system in 1995–1997 demonstrated satisfactory agreement between the values predicted by the system and the actual population characteristics. The system has been developed using SIMER+MIR intelligent techniques (Osipov, G.S., Gukova, S.M., Komarov, S.I., Kurshev, E.P., 1987. Expert system tools for badly structured fields. In: Plander, I. (Ed.), Artificial Intelligence and Information-Control Systems of Robots-87. North-Holland, Amsterdam).