A multi response optimization approach toward optimizing operator allocation and machine dispatching rule in a multi stage injection process

researchers have studied manufacturing systems by considering deterministic parameters using mathematical modelling, heuristics, and meta-heuristics. However, it is important to consider the problem as close as possible to real world conditions which consist of uncertainty and stochastic parameters. In this paper we have presented a new approach for optimizing operator allocation and dispatching rule of injection process based on computer simulation. Because of the complicated nature of the problem and stochastic parameters, it is efficient to use discrete event simulation (DES) method to model the problem. Simulation model is validated by considering real data and statistical tests. Then controllable factors and response variables were determined and after that using design of experiments method, Taguchi, 18 different scenarios were developed. After running scenarios and gathering results, we have used a multi response optimization (MRO) approach based on artificial neural networks (ANN) to determine the most efficient level of operators and injection machine’s dispatching rule. At the end injection machine utilization increased by 6%, mean wait time of refrigerators decreased by 14.6 % and also operators cost decreased by 30%. Although the approach presented in this study presents a near optimal solutions but it is not time or cost consuming and is applicable for manufacturing systems that deal with erratic situations and demands.

In the last decades, many