This paper proposes an alternative methodology for the selection of industrial robots using data envelopment analysis (DEA). It aims at the identification, in a cost/benefit perspective, of the optimal robot, by measuring, for each robot, the relative efficiency through the resolution of linear programming problems. The methodology adopted is based on a sequential dual use of DEA with restricted weights. This approach increases the discriminatory power of standard DEA and makes it possible to achieve a better balancing of robot performances. Further benefits refer to the possibility of extending the use of this approach to various multi-attribute decision-making problems where each performance may depend on a number of factors. An empirical application of the methodology, using data from 12 robot manufacturers, confirms the applicability of revised DEA to advanced manufacturing technology selection, and reinforces its use as a tactical/operational tool in the area of production/operations. In order to evaluate the overall balancing of robot performance indicators, a sensitivity analysis (with variable weight restrictions) is also carried out. The comparison of the results with those obtained by applying cross-efficiency, another DEA-based methodology, is also addressed and discussed. Finally, the dual model of DEA has helped to provide a useful economical and technological analysis of the inefficient robots.
Evaluating and selecting investments in industrial robots
BRAGLIA, MARCELLO;
1999-01-01
Abstract
This paper proposes an alternative methodology for the selection of industrial robots using data envelopment analysis (DEA). It aims at the identification, in a cost/benefit perspective, of the optimal robot, by measuring, for each robot, the relative efficiency through the resolution of linear programming problems. The methodology adopted is based on a sequential dual use of DEA with restricted weights. This approach increases the discriminatory power of standard DEA and makes it possible to achieve a better balancing of robot performances. Further benefits refer to the possibility of extending the use of this approach to various multi-attribute decision-making problems where each performance may depend on a number of factors. An empirical application of the methodology, using data from 12 robot manufacturers, confirms the applicability of revised DEA to advanced manufacturing technology selection, and reinforces its use as a tactical/operational tool in the area of production/operations. In order to evaluate the overall balancing of robot performance indicators, a sensitivity analysis (with variable weight restrictions) is also carried out. The comparison of the results with those obtained by applying cross-efficiency, another DEA-based methodology, is also addressed and discussed. Finally, the dual model of DEA has helped to provide a useful economical and technological analysis of the inefficient robots.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
