| This thesis presents a formal definition of a new resource-oriented operations planning architecture for a set of shop floor activities that integrates material processing and material movement planning for discrete part manufacturing. This integration of heterogeneous manufacturing planning tasks that incorporates processing and material handling tolerance and kinematic constraints has not been accomplished before. The definition includes a formal specification of what the operations planner does and a description of the algorithms used to accomplish the operations planning. The architecture, developed for this research, includes an input domain consisting of resource and part models that are object-oriented and designed to facilitate the integration between operations planning and other manufacturing activities such as product design, production cost estimation, scheduling, and production control. The resource model defines how attributes and behaviors of resources may change the part's location on a shop floor or add new features to the part. New algorithms, which are incorporated in this object-oriented resource model, reason about the capabilities of resources to manufacture parts using tolerance stacking, transformations, standard machining rules, and robot trajectory planning. These algorithms also generate instruction sets for material processing and handling activities. The effects and preconditions of actions by resources on a single part are modeled to facilitate operations planning. Thus, the operations planner is driven by data defining the resources and parts so that it may easily handle changes in either. The operations planner incorporates these resource and part models with a problem-reducing search algorithm to automatically generate sequences of operations that result in the finished part. These sequences of operations are converted into an operations plan that can be directly applied to discrete event simulation, scheduling, manufacturability analysis, and production control. |
