Growth model for white shrimp in semi-intensive farming using inductive reasoning methodology

Abstract A growth model for occidental white shrimp ( Penaeus vannamei ) in semi-intensive farming using Fuzzy Inductive Reasoning (FIR) methodology is presented. The model was developed using data from 17 cycles of culture (1990–1995) at the `El Remolino' shrimp farm located on the Northwest Pacific Coast in Sinaloa, Mexico. Due to the nature of the available data, special routines for handling missing data had to be used. Several qualitative relationships were found using the FIR methodology. The significant variables detected were: temperature, salinity, oxygen and weight. The model was validated with two cycles that had not been used in deriving the model. The forecast results exhibited an error ∼10%, which is a considerable improvement over the error of 20% obtained by classical statistical methods. The use of FIR methodology in aquaculture seems very promising. It can help farmers find good farming strategies for obtaining better profits.