A Multiobjective Genetic Algorithm for Assessing Service Quality Factors in Banking Environment

This study proposes a multiobjective genetic algorithm-based framework to optimize key determinants of service quality in banking environments, addressing the limitations of conventional SERVQUAL and BSQ model. This study uses Nigerian banks as a case study to provide a trustworthy scale of measurement for service quality in banks. The BSQ model served as the foundation for the adopted SERVQUAL model. A 5-point Likert scale was used in a standardized questionnaire, and 136 valid answers were gathered. Based on the BSQ model, a total of 22 items were taken into consideration, and the importance of these factors to customer satisfaction was examined. The most important parameters for assessing service quality were extracted using factor analysis. According to principal factor analysis, eight (8) items were chosen to measure how well services were perceived in banks. Genetic algorithm in MATLAB was used to formulate the objective function for multiple regression analysis on the selected items. Goodness of fit of the developed service quality model was evaluated using the F-Test. To determine whether the chosen individual item for service quality were significant, the t-test was also employed. The results showed statistical significance and were perfectly suited to the advanced service quality model. Finding the most accurate variables to assess the perceived level of service quality in banks was done using a genetic algorithm. The results demonstrated that the best items for service quality in banks are “they deliver when promised”, “accuracy in account records”, “services offered align with the most recent advancements in financial services”, and “you can be certain that you will receive value for the money you paid”. The results suggest that tangibles, service portfolio, and effectiveness and assurance are the best ways to assess the quality of bank services.