An Adaptive Network-based Fuzzy Inference System for predicting organizational commitment according to different levels of job satisfaction in growing economies

Abstract

The relationship between organizational commitment and job satisfaction has received plenty of attention in the literature. However, similar studies in growing economies are scarce. The objective of this study is to cover such a gap by introducing an intelligent algorithm for predicting organizational commitment considering job satisfaction as well as comparing its performance to conventional Multiple Linear Regression (MLR). In doing so, data was collected by distributing questionnaires among 200 employees from the food industry in Shiraz (Iran), which represents one of the most dynamic economies of the country. A 73% response rate was achieved. The respondents completed the questionnaire, which assessed six dimensions of job satisfaction (satisfaction with supervision, overall job, company policy and support, promotion and advancement, pay, and coworkers) and organizational commitment. Using MLR, the results indicated that workers' had higher satisfaction with overall job, company policy and support, and coworkers, bringing about significantly higher employees' organizational commitment level. An Adaptive Network-based Fuzzy Inference System (ANFIS) is also developed and tested for the purpose of this study to predict organizational commitment level based on different levels of job satisfaction. Comparing the results obtained from ANFIS and MLR shows that the proposed intelligent algorithm has better performance than conventional MLR and predicts organizational commitment more accurately, based on their root mean square error values (RMSE). A simulation model based on the rules learned by the ANFIS algorithm is also presented to simulate the organizational commitment level of employees by establishing their position on various indexes of job satisfaction. This model can help managers to achieve higher levels of employees' organizational commitment, since the main aspects of job satisfaction that need more focus are simulated. Different scenarios and situations could be simulated by this system, which is a main contribution of the current work. In terms of presenting an intelligent algorithm in order to predict organizational commitment level based on job satisfaction in food industrial companies, this study is pioneering among other studies.

Design and Development of a Genetic Algorithm Based on Fuzzy Inference Systems for Personnel Assignment Problem

“…Every day in our decision-making process, we face situations in which the obtained and available data are imprecise and vague in nature. In dealing with such conditions, using exact and precise modeling is not always an optimal choice (P. Rabiei & Arias-Aranda, 2018). In order to handle uncertainty and ambiguity, Zadeh (1965) introduced fuzzy logic theory to deal with situations in which boundaries are not exactly defined.…”