PROCESS PLANNER'S ASSISTANT: AN INTERACTIVE AND ITERATIVE APPROACH TO AUTOMATING PROCESS PLANNING

摘要

Most existing approaches to computer aided process plan-rnning aim for full automation by searching for a plan that is opti-rnmal with respect to a pre-specifed objective function. Such fullrnautomation is often infeasible in practice for three reasons: (i)rnthe search space of potential plans is huge, (ii) optimality metricsrnare often context sensitive and can only be elicited through userrninteraction, (iii) because of the importance of process planning,rnorganizations are more interested in process planning assistantsrnthat support human expert process planners rather than stand-rnalone process planners.rnIn this paper, we present a process planner's assistant",rnwhich helps the human process planning experts in coming uprnwith process plans. In order to achieve this, the process plan-rnner's assistant must have access to the full search space of pro-rncess plans, and the ability to modify plans in response to humanrncriticism. The former is provided by basing our system on ASUrnFeatures Testbed, a comprehensive and systematic frameworkrnfor recognizing and reasoning with features in machinable parts.rnTo support the latter, the system is equipped with an incre-rnmental and interactive search mechanism. We will discuss thernoperational details of the resultant system, called ASUPPA.