Job Rotation Optimization using Mixed-Integer Nonlinear Programming (MINLP) Method to Balance Operator Workload in Automotive Industry

This research discussed balancing operator workload on the automotive industry assembly line by developing a novel job rotation mathematical model using mixed-integer nonlinear programming (MINLP) method that aims to obtain optimal job rotation design results by considering ergonomic aspects. The implementation of job rotation in workforce planning is carried out by manufacturing industry to reduce musculoskeletal disorder (MSD) risk factors. Ergonomic analysis was carried out at each workstation to evaluate the physical workload of various jobs, in which the results were used as the parameters of the job rotation mathematical model developed in this research to schedule optimal job rotation and achieve a balanced cumulative workload. Ergonomics aspect was considered in designing the job rotation model to prevent sequentially high workload exposure for an operator and also adjust the operator's capacity to do work at the workstation because it will be related to additional training costs and time consequences. The result of job rotation programming in this research is the optimal work order for each worker so that the global daily workload will be balanced. The job rotation strategy proposed in this research succeeded in reducing the spread and deviation of the cumulative daily workload among workers by decreasing the standard deviation from 10.73 to 0.32, proving that the physical workload is equally distributed among operators.