Seyed Saeed Hosseinian - Department of Industrial Engineering, Shahed University, Tehran, Iran.
Abas Hajfathaliha - Department of Industrial Engineering, Shahed University, Tehran, Iran.
Masoud Khalili - Department of Mathematics, Shahed University, Tehran, Iran.

The use of weight restrictions is a well-known method for preventing DMUs to ignore or overestimating some inputs or outputs the efficiency assessment. The most popular weight restrictions are absolute weight restrictions and assurance regions which the first one imposes lower and upper bounds on the inputs and outputs weights separately and the second one imposes the lower and upper bound on the ratios of the weights. Assurance regions are divided to two sets of type I and type II. Assurance regions type I determine the lower and upper bounds on the ratio of weights between inputs and outputs separately. The values of lower and upper bounds of weight restrictions are related to the relative importance of the inputs and outputs. This paper presents a method, based on AHP, to derive the lower and upper bounds of absolute weight restrictions and assurance regions type-I from the pair-wise comparison matrices of a group of decision makers. This method decreases the effect of the inconsistent comparison matrices in assessing the lower and upper bounds. In other words, the decision makers with lower inconsistency have the stronger effect in determining the weights bounds.

Data envelopment analysis (DEA); Analytic hierarchy process (AHP); Weight restrictions; Inconsistency