A weighted goal programming model for maintenance workforce optimisation for a process industry

摘要

The recent upsurge in economic distress of organisations, and particularly the sustainability challenges faced by them raises new concerns that strongly motivate maintenance workforce structural re-modelling. Maintenance workforce planning is an interdisciplinary area spanning maintenance, industrial engineering, and human resource planning. Various analytical models in the literature have been developed, re-modelled and implemented for maintenance workforce planning. However, new research insights focusing on budgets, worker distribution and performance metrics (availability and quality of work done) as well as hiring and firing costs are keenly needed. By responding to this call, the current communication adopts a case-study approach in the optimisation of maintenance workforce variables based on weighted goal programming, genetic algorithms (GA) and Euclidean distances with these parameters treated in a unique manner. An optimisation model selected from the literature was used to formulate a model for a brewery plant maintenance system. The formulated model used a genetic algorithm (GA), particle swarm optimisation and a differential evolution algorithm. The results obtained were compared. It was observed that GA was the most suitable solution method. The GA results showed that the maximum number of full-time workers hired or fired for the different worker categories were the same (one worker). Worker efficiency and availability were both above 80%, while the quality of work done was above 70%. The results showed that the solutions from the weighted goal programming, GA and Euclidean distance were satisfactory.