Performance Evaluation of ATC based Greedy Heuristic Algorithms in Scheduling Diffusion Furnace in Wafer Fabrication

This study addresses the scheduling of a Batch Processing Machine (BPM) with incompatible job-families, dynamic job-arrivals, and non-agreeable release times & due dates with the objective of minimizing total weighted tardiness (TWT). First we propose a mixed integer linear programming model (MILP) for the research problem. Due to the computational intractability in getting optimal solution for large scale real-life sized problems, we propose nine variants of Apparent Tardiness Cost (ATC) based greedy heuristic algorithms (GHA). From the series of computational experiments carried out it is observed that out of nine ATC based GHA, two proposed variants of ATC based GHA consistently performing better than other variants of ATC based GHA in comparison with an estimated optimal solution.