Using fuzzy logic approach to predict work-related musculoskeletal disorders among automotive assembly workers

Musculoskeletal disorders (MSDs) are considered an important health concern, particularly in automotive assembly industries. Evaluation of the effects of all MSDs risk factors is difficult due to its multifactorial nature. In addition, the risk factors cannot be detected accurately when they are only based on individual opinions. Thus, in this study, fuzzy logic tool was used to evaluate the combined effects of all risk factors on MSDs. Methods: This cross sectional study was conducted on 100 male workers in an automotive industry. Job satisfaction, job stress, job fatigue, and body posture were evaluated by a self-reported questionnaire. Body posture was evaluated using Rapid Entire Body Assessment (REBA). Primary data analysis on extracting the input variables of MATLAB was performed by SPSS 22, with a significant level of 0.05. T test, one-way Anova, and Pearson correlation analysis were used to extract the input variables for the fuzzy logic model. The results obtained from the Nordic questionnaire was selected as the output of the fuzzy model. Fuzzy logic assessment was performed using MATLAB software version 7.0. Results: There were significant differences between WMSDs factors, including job fatigue, strain, working posture, and the REBA final score, and pain in all limbs of the body (p<0.05). A significant difference was also found between working posture with wrist score (p<0.05). The findings on defuzzification showed a strong correlation between real and modelling results. Conclusion: The results showed that many factors such as posture, fatigue, and strain affect MSDs. Based on the obtained results, all categories of risk factors, including personal, psychosocial, and occupational, should be considered to predict MSDs, which can be achieved by a modeling approach.