Abstract
This paper describes an automatic scheduling system for steelmaking process from refining converter to continuous caster. Conventionaldaily scheduling system does not get an optimal solution automatically, but gets a feasible solution with no machine conflicts between processes. It is very difficult to get the optimal solution for the large-scale and complex problem of steelmaking process scheduling, because this scheduling involves particular constraints and varying production requirements of each process. The proposed scheduling procedure consists of optimal method by using modified genetic algorithm (GA). This method applies multi-constraints to the reduction of the search space obtaining feasible solutions. And it can suppress the calculation time. The system has two benefits. One is an extension of the weekly scheduling system from the daily one. The other is an improvement in the quality of the schedule. The cost effectiveness is low waiting time of secondary refining process and low changing of alternative secondary refining process. This algorithm was installed in the scheduling system of NKK Keihin Works and has contributed to improving its efficiency in the steelmaking process.
