a method for target setting with share data

Data envelopment analysis (dea) is a mathematical programming technique for evaluating the relative efficiency of a set of decision making units (dmus) and can also be utilized for setting target. target setting is one of the important subjects since according to its results efficiency can be increased. an important issue to be currently discussed, is to set target while considering share data. these data for each individual indicate the share of the unit, which takes part in an activity, from the whole amount which is a predefined constant. it is obvious that the sum of units’ share is equal to the entire amount. thus, any changes in the magnitude of these data has to be dependent on the changes in data of other units. in this paper a two-stage procedure is developed to find benchmark units where share data ex ist. the fact that all dmus are jointly projected onto the new efficient frontier and simplicity, are the significant features of the proposed method. with a numerical example we demonstrate how this method works.