We are glad to announce that Professor Tone will present two talks, as detailed below.
Dynamic DEA with network structure: A slacks-based measure approach
We propose a dynamic DEA model involving network structure in each period within the framework of a slacks-based measure approach. We have previously published the network SBM (NSBM) and the dynamic SBM (DSBM) models separately. Hence, this article is a composite of these two models. Vertically, we deal with multiple divisions connected by links of network structure within each period and, horizontally, we combine the network structure by means of carry-over activities between two succeeding periods. This model can evaluate (1) the overall efficiency over the entire observed period, (2) dynamic change of period efficiency and (3) dynamic change of divisional efficiency. The model can be implemented in input-, output- or non-(both) oriented forms under the CRS or VRS assumptions on the production possibility set. Finally, we applied this model to a dataset of US electric utilities and compared the result with that of DSBM.
A scale and cluster adjusted DEA model that permits both convex and non-convex efficient frontiers
In DEA we are often puzzled by the big difference in CRS and VRS scores, and by the convexity production set syndrome in spite of the S-shaped curve often observed in many real data. In this paper we perform a challenge to these subjects. At the first step, we evaluate CRS and VRS scores for all DMUs by means of some techniques, e.g. by the conventional methods. We obtain the scale-efficiency for each DMU, e.g. CRS/VRS. Using the scale-efficiency, we decompose the CRS slacks into scale-independent and scale-dependent parts for each DMU. At the second step, we eliminate scale-dependent slacks from the data set and thus obtain a scale-independent data set. At the third step, we classify DMUs into several clusters depending on the degree of scale-efficiency or on some predetermined characteristics. We evaluate slacks of scale-independent DMUs within the same cluster by the CRS model and obtain in-cluster slacks. Adding scale-dependent and in-cluster slacks, we define the total slacks for each DMU. And at the fourth step, we evaluate the efficiency score of DMU by means of the total slacks via the SBM model and project the DMU onto the efficient frontiers which are no more guaranteed to be convex and usually are non-convex. Lastly, we define the scale-dependent data set by which we can find the scale elasticity of each DMU. We apply this model to a data set of Japanese universities.
Event organized by University of Edinburgh Business School and sponsored by OR Group of Scotland.