Customer satisfaction assessment with fuzzy queries and ANFIS for an automotive industry

Measuring customer satisfaction is an important part of marketing research in enterprise modeling. It is the key to formulate customer value strategies and continuously improve them. This paper deals with the fuzzy querying language of regular relational databases called SQLf, and proposes an adaptive-network-based fuzzy inference system (ANFIS) based on Takagi-Sugeno-Kang (TSK) fuzzy controllers for this purpose. The system uses genetic algorithm (GA) for tuning the interface parameters of the proposed fuzzy model. Moreover, the parameters of the membership functions and weight of each effective factor in customer satisfaction are also optimized. The generated membership functions are used for processing fuzzy queries. Finally, the system is tested and verified in an automotive industry.