A TWO STAGE SOLUTION APPROACH FOR THE SHIFT-BASED SCHEDULING PROBLEM ON WELDING MACHINES

In this study, one of the special version of n product, m unrelated parallel machine scheduling problem is considered. This problem occursin a firm which produces white goods and it is determining that which job produces at which welding machine and in what order. Generally, alignment of workload is determined on scheduling problems but the shifts of the workload are not considered. However, in real-life, there are some restrictions that needs attention, in shift basis. Therefore, this study proposes a model both scheduling the jobs on which machine needs to be used and align it correctly, also designates the shift of production. With this model, special circumstances such as specific capacity for a shift could be considered. In this problem, there are preparations periods depending on alignment. These periods are changing depending on the weight, height and brand of product. In this study first, a coefficient of variation has been identified that could represent their preparation time and an approach has been developed to calculate these coefficients. In this way, measurement of setup time will consist of in case of sequential production of all products requirements can be eliminated. Purpose of this problem is minimizing the sum of finishing period of the last assignment and discrepancy factor. A goal programming model is developed for this multi-objective problem which includes special process constraints. This suggested mathematical model is tested with small sized and real-life problems. A solution is obtained for the small sized problem but real-life problem could not be solved with GAMS/CPLEX. Therefore, the two-stage approach which will work successively has been proposed instead of integrated models for the solution of major problems. In the first stage of the procedure, Works are assigned to shifts. In the second stage, a mathematical model is solved for each shifts. The real life problem can be solved by proposed solution approach