5 th International Advanced Technologies Symposium (IATS’09), May 13-15, 2009, Karabuk, Turkey FEATURE RECOGNITION AND DESIGN ADVISORY SYSTEM FOR SHEET METAL COMPONENTS Mohammad Ali Farsi a,* and Behrooz Arezoo b a, * Islamic Azad University- Eghlid branch, Iran, mohamadalifarsi@yahoo.com b CAD/CAPP/CAM Research centre, Amirkabir University of Technology, Tehran, Iran, arezoo@aut.ac.ir Abstract The present paper explains a feature recognition module and design advisor system for sheet metal components. In this system features of sheet metal components are extracted automatically from 3-D model. These data are modeled based on object oriented. The proposed system will aid the designer right at the design stage with useful design and manufacturability recommendations that have been gathered through years of experience. The system has been implemented in solid works 2008 using visual basic. Keywords: Feature recognition, Advisory system, Sheet metal. 1. Introduction Today, many products are made with sheet metal components. These products can be simple components requiring simple fabrication steps while others may involve very complicated processes requiring several steps to fabricate. For the complicated parts the time to market from conceptualization stage can be significantly shortened at various levels from design to manufacturing by understanding the product, its features and implications of the features [1]. It is widely accepted that about 70 percent of the final product cost is determined during the design stage [2]. Systems that automatically provide inputs on manufacturing guidelines to designers during the initial design process that lower costs and reduce cycle times have proven highly effective in achieving their goals. The sheet metal industry is largely an empirical industry. The design and process planning of sheet metal products need human involvement more than other areas of manufacturing industry. Design of complex parts are based heavily on human experience. In the design stage, if some of the design considerations are ignored, the cost of production may increase rapidly. To prevent modification attempts of the initial design from becoming a trial and error process, some knowledge- based assistance is necessary. If the traditional knowledge gained from years of experience is integrated into a solid modeler, it will provide valuable input to the designer. The designer can then attempt changes while adhering to the functional requirements of the design for achieving easier manufacture and reduced costs at the same time. Although, there have been attempts ([1], [3-[5]) to create a solution for these problems, most of them have been design rule checkers using a few design rules. Also, these systems only check for violations of certain constraints and do not suggest recommended practices where appropriate. A comprehensive design advisory system has not been attempted yet. The comprehensive approach should successfully encompass the design rules, standard manufacturing practices and design recommendations across all sheet metal processes and also investigate the interaction between these elements in the creation of a final sheet metal part. 2. Related works Nnaji et al. [1] have developed a set of principles for extracting features from sheet metal parts. In order to make the system generalized, the system developed requires B-Rep information about the sheet metal part in IGES neutral file format. Mantripragada et al. [5] have developed a feature based design system, which acts like an interactive design tool and can be used to alert designers for potential production problems, defects and failures. It also provides them with information that can be used to explore alternative designs, evaluate trade-offs and arrive at optimal design for the given process conditions. Jagirdar et al. [6] have proposed a feature recognition methodology for recognizing shearing operations for 2-D sheet metal components created by a wire-frame model. Yeh, et al. [7] described a feature based product modeler, Promod-S, which includes a rule-based design advisor among several other modules. In recent years, Shashikiran [8] designed an Advisory system for sheet metal components. This paper describes a new package for feature recognition and advisory system. The system uses 3-D solid model of sheet metal component as input. It also provides an advisory system to control the part and warn the user of any incorrect features in the component. The system also presents some comments for necessary modifications. 3. Feature Recognition The features that are basic to sheet metal parts are stamped and formed features. Stamped features are categorized as internal and external features. Internal features include features like round holes, square holes, rectangular holes, oblong holes, etc. External stamped features include different types of notches. The most widely formed features are bends and flanges. In the feature recognition module, all data used for the advisory system and a die designer system are extracted [9]. Data modeling methodology in this work is based on object oriented approach. For example, the following data are extracted from a bend feature: Bend radius Bend angle Length of each bend Bend height Bending direction © IATS’09, Karabük University, Karabük, Turkey