Assessing computer numerical control machines using data envelopment analysis

This paper reports on an application of data envelopment analysis (DEA) to evaluate computer numerical control (CNC) machines in terms of system specification and cost. The evaluation contributed to a study of advanced manufacturing system investments carried out in 2000 by the Taiwanese Combined Service Forces. The methodology proposed for the evaluation of 21 CNC machines is based on the combination of the Banker, Charnes, and Cooper (BCC) model (Banker et al. 1984) and cross-efficiency evaluation (Doyle and Green 1994). It aims at the identification of a homogeneous set of good systems, by measuring, for each machine, the pure technical efficiency through the BCC model. The use of cross-efficiency evaluation is to differentiate better between good systems and bad systems. These good systems can be further used for the selection of the best systems in the decision-making process. The analysis of results might help industrial and military buyers as a normative model of system excellence against which system purchasing behaviour could be compared. In addition, it might help present potential uses for machine manufacturers in competitor analysis and in determining unexplored market niches.