PC master production scheduling system with task/resource/quantity selection heuristics

摘要

Master production scheduling systems with numerical tables have been in great demand due to reduced scheduling costs. This paper presents a knowledge representation framework for task, resource, and quantity selection heuristics which model human scheduling expertise. A PC master production scheduling system was developed on this framework moreover its usefulness has been evaluated by two actual PC production factory groups. It is shown that the time required to mate a schedule can be reduced to 1/3-1/20, compared to conventional methods using spreadsheet tools. When the system was applied to other factory groups under considerably different conditions, the knowledge representation framework helped to reduce the customization costs. Consequently, the proposed PC master production scheduling system has been operated daily by true actual PC production factory groups, and it resulted in greatly reduced scheduling costs.