Heuristic Scheduling

Resource Scheduling

Use of AI techniques for various types of resource scheduling problems has been of substantial interest world-wide. This is partly because of the inadequacy of traditional Operations Research techniques in tackling the complexity of real-life problems. Also, the flexibility offered by AI techniques in providing explanations and modifications are often essential for a practical implementation of such systems. KBCS has worked on following projects related to resource scheduling on interaction with industry for real-life transportation problems:

• OCCPTS - A system for scheduling the movement of oil tankers for distributing petroleum products
• SARANI - A knowledge based scheduling support system for aircraft scheduling
• PIPES - A system for scheduling oil-movements through pipelines.

All these applications fall under the broad category of transportation problems. They can all be viewed as search problems. Though the nature of constraints are quite different for the different problems, at some level of abstraction they can all be viewed as similar. Therefore, we have tried to adopt a common methodology in order to tackle them. The systems we have developed use heuristic beam search as the underlying search strategy. We are trying to generalize the ideas we have used in order to create a framework for transportation problems.