Scheduling

Smart Engineering Systems Architectures

Summary

Over the past three decades, a great deal of research has been done in scheduling. Unfortunately, most of the research has been theoretical in nature and problem specific. Hence, the impact of scheduling research in manufacturing companies was limited. Companies adopted more practical solutions to the problem due to their extensive knowledge of their facilities, customers, and the dynamics of the scheduling objectives. This creates a big gap between the application and theoretical developments. This is basically due to fundamental characteristics of scheduling problems; effective scheduling is a knowledge intensive activity requiring a comprehensive model of the factory and its environment at all times in the schedule generation process. There is a need for effective intelligent scheduling systems that can interact with their environment and adapt to changes in time. Emerging technologies, artificial neural networks, fuzzy logic, and evolutionary programming provide essential tools for designing such systems. Scheduling research is focussed towards building such models.

Recent research in this area integrated genetic algorithms and artificial neural networks for generating job shop schedules in a discrete manufacturing environment based on a non-linear, multi-criteria function. A genetic algorithm is used as a search technique, and a artificial neural network is used as a multi-criteria function evaluator.

Graduate Students Involved in Research

Sinchai Sittisathonchai

Last update: December 20, 1995.