Doctoral Abstracts (1990-2001)

CALA, MARTIN
(December 1990), Associate Professor, Youngstown State University, Youngstown, OH

Email: mcala@cc.ysu.edu

An Intelligent Model Based Process Planning Methodology for Surface Mount Technology

Process planning in Printed Circuit Board assembly using Surface Mount Technology is scheduling intensive. Artificial Intelligence based techniques that have been successfully applied to computerized planning and scheduling systems in other industries can be extended to printed circuit board operations. These techniques which include the use of Blackboard Architectures, Object-Oriented Programming Systems, and Non monotonic Reasoning Systems have made real- time, integrated systems viable. In this research, the aforementioned techniques have been blended in a unique architecture of blackboard systems that communicate via object oriented messages to arrive at the set of decisions that constitutes a shop-floor process plan and schedule. The message passing protocol includes a mechanism for revision of decisions based upon inputs that have henceforth changed. The result of revision to the decision set is a bi-directional, or non monotonic, path to the final decision set. The methodology was specialized to the assembly of printed circuit boards using surface mounted components in a high-variety/low-volume product mix. The assembly facility was represented as a hierarchical object of product, process, and organizational knowledge. The ultimate objective of this research was to develop an integrated process planning and scheduling system that surmounts the limitations of previous Computer Aided Process Planning (CAPP) systems for electronic assembly. To this end, this methodology extends the degree of integration within the surface mount technology facility while both assisting in the computation of various process planning parameters, and remaining effective and applicable under the dynamic conditions of frequent technological advances or under the conditions and constraints of different sites. The prototype system illustrates the required interface with Computer Aided Design/Engineering systems and Computer Aided Manufacturing systems to reduce design-to-manufacture times, reduce product development and support costs, improve quality conformance, improve delivery conformance, and provide information for feedback to planning and design of new products. The methodology was validated by comparing the responses of the knowledge-based system with the responses of practicing printed circuit board process engineers when both are provided with equivalent design, facility, and production demand information. The verification studies indicate that this computer based methodology provides the flexibility and adaptability of the human process planner with an added degree of consistency across process plans.