ASPECTS OF DESIGN RETRIEVAL PERFORMANCE USING AUTOMATIC GT CODING OF 2 D ENGINEERING DRAWINGS

Many companies have the aim of reducing part variety as it confers on it a range of significant benefits. Achieving this requires a retrieval system that is effective, that designers will use and that is cost effective to implement and maintain. Coding and classification/Group Technology(GT) codes are widely used in systems for retrieval of existing parts but involve expensive and time consuming manual coding. A system called Camac has been developed which uses a novel approach of automated coding of 2D engineering drawing to deliver a cost effective solution with the added benefit that searching is based on a CAD sketch of the part and not text descriptors. Testing has already shown that the performance of Camac to be excellent but that it may be affected by the detail and quality of the seed sketch. Testing is describe which investigates the impact of sketch detail and quality as well as database size. It is concluded that an acceptable performance with poor quality sketches should be attainable with databases in excess of 20000 items and that for good quality sketches databases of significantly greater size will not be a problem.

[1]  J. P. Eakins Design criteria for a shape retrieval system , 1993 .

[2]  Nallan C. Suresh,et al.  A neural network system for shape-based classification and coding of rotational parts , 1991 .

[3]  Arlo L. Ames,et al.  Production ready feature recognition based automatic group technology part coding , 1991, SMA '91.

[4]  J. Barton,et al.  Retrieving designs from a sketch using an automated GT coding and classification system , 2005 .

[5]  Youcef Nadir,et al.  PROCODE—automated coding system in group technology for rotational parts , 1993 .

[6]  Subodh Chandra Sharca A Critical Study Of The Classification And Coding Syntons In Manufacturlns Companies. , 1978 .

[7]  Mohan V. Tatikonda,et al.  Adoption and implementation of group technology classification and coding systems : insights from seven case studies , 1992 .

[8]  Joaquim A. Jorge,et al.  Towards content-based retrieval of technical drawings through high-dimensional indexing , 2003, Comput. Graph..

[9]  Michael Anderson,et al.  NIRS: Large scale ART-1 neural architectures for engineering design retrieval , 1994, Neural Networks.

[10]  Nancy Lea Hyer,et al.  Group technology in the US manufacturing industry: A survey of current practices , 1989 .

[11]  Andrew Kusiak,et al.  Group technology , 1987 .

[12]  William F. Hyde Improving productivity by classification, coding and data base standardization , 1981 .

[13]  S. Dowlatshahi,et al.  Application of group technology for design data management , 1995 .

[14]  H. Johnson,et al.  Profit Beyond Measure: Extraordinary Results Through Attention to Work and People , 2000 .