Multi-unit Restaurant-productivity Assessment: A Test of Data-envelopment Analysis
Reynolds, Dennis; Thompson, Gary M.
This report describes a three-step process for performing a data envelopment analysis (DEA) to compare restaurants’ efficiency and to examine their best practices. To start with, prospective efficiency factors must be analyzed to ensure that they are relevant. Secondly, to put restaurants on an equal footing the first DEA should consider only managerially uncontrollable (nondiscretionary) factors as inputs. With uncontrollable factors accounted for, managerially controllable factors can then be assessed in terms of their effect on productivity. Best practices can be isolated and assessed in this manner. To illustrate this three-step approach, data from 60 full-service restaurants are analyzed. From a large number of prospective input factors, the analysis considers a short list of uncontrollable inputs namely, hourly server wage, number of restaurant seats, and a coding variable representing whether the restaurant is a stand-alone facility. The output variables for this analysis were daily sales and tip percentage. Just over 20 percent of the restaurants operated with maximum efficiency, with the chain’s average efficiency hitting 82 percent-good, but leaving room for improvement. However, the two discretionary factors that were proposed as differentiating the restaurants’ efficiency-server hours and number of servers-proved not to be significant factors, inviting further analysis of the efficiency effects of additional discretionary factors.
data-envelopment analysis (DEA); efficacy; profit gains; increased productivity
Required Publisher Statement: © Cornell University. This report may not be reproduced or distributed without the express permission of the publisher