An Integrated Framework for Feature Recognition and Co-Evaluation of Process Planning Through Kinematics Configurations for Flexible and Changeable Manufacturing Systems

Abstract

In modern design, products are often designed and simulated in special software environments to test and see the effects in real world usage. That data is used to refine the design and make required changes to achieve objectives set forth beforehand. After a design is finalized, a computer file is generated which contains 3D data about the product or part. Then this file is analyzed by an operator and 2D drawings are generated from different views with dimensions. These 2-D drawings are ones which are sent to planning and then subsequently manufacturing facilities. The aim of this research is to provide a framework for data extraction from the 3D model of the newly designed part and feed it to the later stages of manufacturing. In addition, the framework will also let the operator enter Machine details (Specifications, Tools, DoF) right after the part data is entered. This will allow the framework to eliminate all process plans irrelevant to the available machining setup and avoid the hassle of doing so later. Weighted Genetic Algorithm (WGA) is proposed to be used for optimization and selection of best process plan(s). The priorities of different optimization objectives can be set using the weights assigned to each objective. A novel method of applying the genetic operators has been developed which allows the diversity to be ensured during the entirety of the life cycle of the algorithm. The framework has been developed in MATLAB® and has been designed form the ground up to be fully generic allowing the process to be seamless and highly intuitive.