Hybrid system of expert system and artificial neural networks for objective evaluation of product sensuous quality

Basic problems and the bottleneck of current approaches for objective assessment of product sensuous quality (PSQ) are discussed. As a solution, a new approach, an expert system (ES) based on artificial neural networks (ANNs) is proposed, in which the ES and ANNs co-operate in a superiority compensation way. T he knowledge base of the system can be effectively built and the evaluation of PSQ can be conducted on-line. As a case study, the new approach has been applied in leather handle test and it proves that the approach is capable of handling non-linear relationships among multiple measured PSQ parameters