CASE-BASED REASONING AND LEARNING IN MANUFACTURING WITH THE TOLTEC PLANNER

The research presented concentrates on bringing together case-based re asoning and expert knowledge-based reasoning in a planner called TOLTE C. The expert's domain-specific knowledge is modeled as dynamic memory structures and this representation is used to help the planner reason and control its planning process. TOLTEC uses a complex indexing of i ts cases, so as to allow incremental retrieval. The TOLTEC planner is applied to a highly constrained domain, and it is shown how the final plan is created- by adding memory- and situation-selected chunks of su btask expansion to each subtask, until the problem is reduced to primi tive (non-expandable) tasks. It is also shown how the use of dynamic m emory structures and dynamic, user-directed backtracking allows the pl anner to predict and discover failures, recover from them, and modify its knowledge according to them. Finally, it is shown how in a domain of multiple possible solutions for each goal the methodology developed allows the planner to slowly model itself to the preferences of the u ser. The paper also discusses some of the application domains where TO LTEC has been used, including process planning of cylindrical and pris matic parts, design checking, material selection, and design of commun ications systems.