part recognition algorithm is independent from different commercial expert system applications. A knowledge base is also constructed using a Windows based text editor and a rule writing format was developed. From this view point, part recognition algorithm is original. With optional and automatic rule writing module integrated into the part recognition algorithm, complex parts with free form surfaces are easily recognized without any limitation regarding number of faces, topology, face type and geometry. The part recognition algorithm can be used for manufacturing activities such as group technology (GT) and process planning. However, when number of faces on the part increases, reasoning time for part recognition increases due to computational complexity. In this study part recognition algorithm is applied to computer aided assembly system. Computer aided automatic assembly is quite simplified with this study. Part recognition based assembly may be a new approach for CAD systems which comprises of computer aided assembly. For implementation of automatic assembly, a diesel automobile engine is selected because the parts of it are very complex.

Limitation of the assembly system is that assembly information of any product to assemble except the diesel engine should be located into database in the system. Future works should be focused on the effective use of the part recognition and FORM different CAD/CAM activities, and on obtaining the flexibility on the part recognition based computer aided assembly system. When they are performed, part recognition based automatic assembly may be a good progress for CAD systems. Part recognition and FORM can also be used as input for CAD/CAM applications.

References

[1] A. C¸ ic¸ek, M. Gu¨lesin, Reconstruction of 3D models from 2D orthographic views using solid extrusion and revolution, Journal of Materials Proces-sing Technology 152 (2004) 291–298.

[2] M.H. Kuo, Automatic extraction of quadratic faces from wire-frame models, Computers & Graphics 25 (2001) 109–119.

 746 A. C¸ ic¸ek, M. Gu¨lesin / Computers in Industry 58 (2007) 733–746

 [3] M. Gulesin, R.M. Jones, Face oriented neighboring graph (FONG): a part representing scheme for process planning, Computer Integrated Manu-facturing Systems 7 (3) (1994) 213–218.

[4] M.P. Bhandarkar, R. Nagi, STEP-based feature extraction from STEP geometry for agile manufacturing, Computers in Industry 41 (2000) 3–24.

[5] A.J.C. Trappey, C.S. Lai, A data representation scheme for sheet metal parts: expressing manufacturing features and tolerance requirements, Journal of Manufacturing Systems 14 (6) (1995) 393–405.

[6] M.T. Wang, M.A. Chamberlain, A. Joneja, T.C. Chang, Manufacturing feature extraction and machined volume decomposition in a computer integrated feature based design and manufacturing planning environment, Computers in Industry 23 (1993) 75–86.

[7] T.N. Wong, C.B. Leung, Conversion of box shaped features for manu-facturing applications, Computers & Industrial Engineering 34 (2) (1997) 501–514.

[8] P. Pal, A. Kumar, A hybrid approach for identification of 3D features from CAD database for manufacturing support, International Journal of Machine Tools & Manufacture 42 (2002) 221–228.

[9] P. Pal, A.M. Tigga, A. Kumar, A strategy for machining interacting features using spatial reasoning, International Journal of Machine Tools & Manufacture 45 (2005) 269–278.

[10] S. Meeran, A.H. Zulkifli, Recognition of simple and complex interacting non-orthogonal features, Pattern Recognition 35 (2002) 2341–2353.