Scheduling System for Multi-party Arrangement Work by Utiling Mathematical Optimization and Machine Learning Techniques

Abstract

In this paper, we propose a new scheduling method for the whole system in which sub-systems are linked to each other. The sub-systems are independent and make its schedule by itself. All the sub-system's schedule must be consistent in the total system (System of Systems). Proposed method is intended to be used at a certain system among the subsystems, and propose and ask the schedule to be changed to another system. Though simple asking may sometimes rejected, our method can build more acceptable schedule for other systems, by predicting the other system's response and select the most acceptable one. This scheme is done by machine-learning technology. With the experiment, we have verified the feasibility of our proposed method.