An intelligent fuzzy set-up planner for manufacturability and fixturability evaluations

Abstract

Manufacturability evaluation of the features on a part for machining is a popular research topic due to the rising concern over effective product design in the Concurrent Engineering and CIM concepts. Many of the currently available manufacturability evaluation systems, however, require specialized expertise and inference and/or analysis methods, such as statistical tolerance analysis, cutting force analysis, etc., which may be time-consuming and would require a considerable amount of specialized knowledge that is not directly available to the designers. Many of these systems that are rule-based, formalize the ill-defined boundaries of the operations' capabilities using precise production rules. In addition, the manufacturability evaluation of a part is usually performed after the part has been designed. However, manufacturability evaluation must become an integral part of the design process to truly integrate design and manufacturing. Set-up planning, which is a macro-planning process in both process planning and fixture planning, plays a key role in achieving this integration. The requirements in both process planning and fixturing are considered, as a set-up plan is formulated in set-up planning. Through the consideration of these requirements, manufacturability evaluation can be performed and design elements that might cause problems in the machining and work-holding of a part can be identified. In this paper, a practical and useful intelligent fuzzy set-up planner that is capable of performing manufacturability and fixturability evaluations is presented. This system employs machinability data and machining heuristics that are commonly practised on the shop floor for assessing the manufacturability of features. Fuzzy set representations and reasoning are used in the representation of the operations and features relations, and in the operations selection and set-up planning procedures in this system. The fuzzy membership functions are reflective of the capability of the current manufacturing system as they are parametric and can be constructed during the program execution.