A Batch Optimization Solver for diffusion area scheduling in semiconductor manufacturing
Résumé
This paper presents a method and a software for solving a batching and scheduling problem in the diffusion area of a semiconductor plant, the ATMEL fabrication unit in Rousset, France. The diffusion area is one of the most complex area in the fab. A significant number of lots has to be processed while satisfying complex equipment process and line management constraints. The purpose of this study is to investigate approaches to group lots in batches, to assign the batches on the equipment and to sequence these batches. Three indicators are used to evaluate the quality of a solution: the total number of moves, the batching coefficient and the X-factor. The problem is modeled through the disjunctive graph formulation. A constructive algorithm is proposed and improvement procedures based on iterative sampling and simulated annealing are developed. Computational experiments, carried out on actual industrial problem instances, show the ability of iterative sampling to improve significantly the initial solution. The proposed simulated annealing method brings in turn important enhancements to the results of iterative sampling. The software based on these methods, named Batch Optimization Solver (BOS), is currently used in the diffusion area of ATMEL. The disjunctive graph model allows in addition a high level of interactivity with the decision makers.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...