Simulation Optimization: Applications in Fish Farming—Theory vs. Practices

The aim of this chapter is to address these two problems. (a) To develop fish-farm simulation optimization equations and an application method, (b) to demonstrate the application of these equations in real life situations: 2,500 ton/year marine netcages and 1,000 ton/year recirculating aquaculture systems. The results. The model optimizes: (1) facility allocation, i.e., number and volume of netcages in each growing phase; (2) fish-batch arrival frequency; (3) number of fingerlings in a batch; (4) number of days in each culture netcage, and (5) grading criteria along the production lines. Compared with today existing management the optimized layout was superior, giving 1,687 vs. 981 ton/year). It is recommended that every aquaculture enterprise apply this concept in its design stage. Prefix Probably, one of the most beneficial links between operations research (OR) and computer science has been the development of discrete-event simulation software (the so-called, in this chapter, Simulation). The further development, the linkage of optimization techniques and simulation practice, has become nearly ubiquitous. Therefore, nearly every commercial simulation software packages have now included a sort of “optimization.” However, (a) contrary to the use of mathematical programming software packages, the simulation user has no way of knowing if a global optimum has actually been reached (hence, the quotations around optimization at the beginning of this paragraph). (b) In aquaculture, only few “simulation optimization” techniques have been applied in practices. The aim of this chapter is to address these two problems. (a) To develop fish-farm simulation optimization equations and an application method, (b) to demonstrate the application of these equations in real-life situations: 2,500 ton/year marine netcages and 1,000 ton/year recirculating aquaculture systems. The results. The model optimizes: (1) facility allocation, i.e., number and volume of netcages in each growing phase; (2) fish-batch arrival frequency; (3) number of fingerlings in a batch; (4) number of days in each culture netcage, and (5) grading criteria along the production lines. Compared with today existing management the optimized layout was superior, giving 1,687 vs. 981 ton/year). For the new system that is now under construction, the optimized layout was selected. Under our conditions: Optimal arrival frequency is a batch every month, and optimal retention times are 122 days in each successive growing phase (up to 62, 196, and 382 g, respectively). Use of these parameters did not violate the biomass-density criterion (15, 20, and 25 kg/m3, respectively) or the netcage utilization criterion (never below 99 %), which suggests that it is not feasible to have fewer culture netcages. The above numerical values reflect local conditions, but the concept is applicable anywhere. Take home message: Simulation optimization was developed and applied in aquaculture. It is recommended that every aquaculture enterprise apply this concept in its design stage. The onus now lies with the industry; to further apply the proposed simulation–optimization methodology, advancing to other aquaculture sites and other species.