Application of fuzzy set and neural network techniques in determining food process control set points

Abstract

Fuzzy set and neural network techniques were used to determine food process control set points for producing products of certain desirable sensory quality. Fuzzy sets were employed to interpret sensory responses while neural networks were applied to model the relationships between process and sensory variables. Rice cake production was used as a model process. Product sensory attributes were evaluated by a trained panel. Multi-judge responses were formulated as fuzzy membership vectors, which in turn were formed into fuzzy membership matrices of multiple sensory attributes. Neural networks were used to determine the sensory attribute controllability and the process control set points for achieving a given target of sensory quality. New products were made by using the process set points determined, and the product sensory attributes matched the desired sensory target values by less than 9% error. The results demonstrate the great potential of the fuzzy set concept and neural network techniques in sensory quality-based food process control with sensory evaluations quantified in a naturally fuzzy manner.