Extensions in efficiency measurement of alternate machine component grouping solutions via data envelopment analysis

This paper focuses on extending the analysis of an efficiency and productivity study on a cellular manufacturing system (CMS) as first introduced by Shafer and Bradford. Shafer and Bradford's initial study introduced a methodology that utilized simulation and data envelopment analysis (DEA) for a multiple performance measure analysis of a CMS. This paper presents an extension to their methodology by incorporating cross-efficiency measures and cluster analysis. One important contribution of this research is in identifying "overall" efficient and "false positive" layouts that are of high managerial significance, Another contribution is in selecting more appropriate targets for poorly performing layouts to use as a benchmark. An interesting managerial insight that emerged from this study is that, in general, cellular layouts with six cells have outperformed hybrid layouts and other cellular layouts. Also, this study concludes that "overall" efficient layouts must be considered by manufacturing engineers and managers in the design of a CMS. Last, the practical problems associated with the improvement of the "false positive" layouts are addressed.