We develop a procedure and the requisite theory for incorporating preference information in a novel way in the efficiency analysis of Decision Making Units. The efficiency of Decision Making Units is defined in the spirit of Data Envelopment Analysis (DEA), complemented with Decision Maker's preference information concerning the desirable structure of inputs and outputs. Our procedure begins by aiding the Decision Maker in searching for the most preferred combination of inputs and outputs of Decision Making Units (for short, Most Preferred Solution) which are efficient in DEA. Then, assuming that the Decision Maker's Most Preferred Solution maximizes his/her underlying (unknown) value function, we approximate the indifference contour of the value function at this point with its possible tangent hyperplanes. Value Efficiency scores are then calculated for each Decision Making Unit comparing the inefficient units to units having the same value as the Most Preferred Solution. The resulting Value Efficiency scores are optimistic approximations of the true scores. The procedure and the resulting efficiency scores are immediately applicable to solving practical problems.efficiency analysis, data envelopment analysis, multiple criteria decision making, value function
The concept of efficiency as it applies to Decision Making Units (DMUs), solutions, alternatives plays an important role both in Data Envelopment Analysis (DEA) and Multiple Objective Linear Programming (MOLP). Despite this and other apparent similarities, DEA and MOLP research has developed separately. We show that structurally the DEA formulation to identify efficient units is quite similar to the MOLP model based on the reference point or the reference direction approach to generate efficient solutions. DEA and MOLP should not be seen as substitutes, but rather as complements. We show that they cross-fertilize each other. MOLP provides interesting extensions to DEA and DEA provides new areas of application to MOLP.Data Envelopment Analysis, Multiple Objective Linear Programming, Efficiency, Performance Measurement
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.