リアクティブスケジューリングにおける拡散共進化遺伝的アルゴリズムの拡張

Abstract

Most of the existing studies about production scheduling are carried out to find an optimum production schedule before starting production activities. However, unpredictable disturbances may occur in the actual manufacturing environment. In that case, the predetermined production schedule can't be applied to a manufacturing system. The disturbed production schedule needs to be modified for unpredictable condition changes flexibly. In the previous study, a reactive scheduling method have been proposed in order to modify the production schedule in parallel with the production activities. The proposed method uses a diffuse co-evolutionary genetic algorithm searching for the most suitable production schedule in the solution space with global scope. However, the previous method may not be able to search in the local area of the solution space. Therefore, this study adds a new method to the previous method for searching in a local area as well as a global area. The new method focuses on an idle time in a production schedule to shorten the makespan. A prototype of reactive scheduling system is developed in order to verify the effectiveness of the proposed method. Computational experiments demonstrate that the new method is superior to the previous method.