Host: The Japanese Society for Artificial Intelligence
Name : The 37th Annual Conference of the Japanese Society for Artificial Intelligence
Number : 37
Location : [in Japanese]
Date : June 06, 2023 - June 09, 2023
In recent years, Japan's declining birthrate and aging population have become serious issues, and the working population continues to decline. As a result, companies find it difficult to recruit new workers and must maintain the number of workers by considering the request of existing workers. Therefore, we propose a system that can generate schedules in real-time to reduce the mutual burden of all people involved in schedule creation. In this paper, considering a manufacturing site with a mix of workers with different work hours, we propose a scheduling system that divides conditions corresponding to a wide variety of needs into objective functions and constraints, formulates each of them as a linear equation, and attributes them to the linear programming method. In this paper, we propose a system that can create schedules based on multiple constraints such as rest alternation, and load balancing. In this paper, we will conduct a scheduling experiment in a problem set based on real data provided by a food manufacturer where the majority of hourly workers. The practicality of the proposed system will be verified by comparing the generated schedules with the work records of the food manufacturer.