An associative memory approach for assembly planning systems

A case associative assembly planning system (CAAPS) which integrates associative memory organization and neural computing techniques has been developed. The purpose of the CAAPS is to provide an environment in which an engineer can think of assembly in terms of high-level features and synthesize such an assembly rapidly. At all stages of the design process he can consult the episodal associative memory (EAM) to see what 'experience' knows of a similar assembly. Efficient use of prior experiences is emphasized. The EAM was implemented using an unsupervised learning algorithm to dynamically accumulate experiences and recall entire assembly designs or devise a new design on demand for a given new task in the design environment. The EAM can efficiently support a designer in creating plans and in indexing and retrieval of assembly design in the CAAPS.<>