期刊名称：Relatórios de Pesquisa em Engenharia de Produção
电子版ISSN：1678-2399
出版年度：2004
卷号：4
出版社：Universidade Federal Fluminense (UFF)
摘要：Data Envelopment Analysis - DEA, developed originally by Charnes et al. (1978), is a method that uses linear programming for the evaluation of comparative efficiencies of the Decision-Making Units (DMUs). Classic multipliers DEA models allow each weight to assume any positive value, in order to maximize a DMU's efficiency, thus often resulting excessive weight variability and implausible weight values, under DM's viewpoint. This has led to the development of DEA models that incorporate weight restrictions, reflecting expert judgment, only the establishement of bounds is not a straightforward procedure. As an alternative, Thanassoulis, E. and Allen, R. (1998) have demonstrated that it is possible to find an unobserved DMU that is a substitute for a set of Weight Restrictions in DEA. However, they have not established the necessary conditions for an unobserved DMU to have an equivalent set of weight restrictions. This paper aims to provide a solution for this issue, presenting the necessary conditions to ensure this equivalence
