Evaluating and managing congestion in Chinese production

Abstract

This dissertation is devoted to the analysis of congestion in the context of its uses in Chinese production. Data Envelopment Analysis is the major tool be employed for this purpose. Applications to Chinese industry provide concrete examples of potential uses that can result from the research that will be reported in this dissertation. Congestion is a severe form of inefficiency in which increases in inputs result in decreases in outputs. Western approaches are directed to simply eliminating congestion by reducing inputs but this need not apply in non-western contexts. China, for instance, must accommodate 16,000,000 to 18,000,000 new entrants into its labor force each year. For perspective, we might note that former President Clinton points to the creation of some 10,000,000-12,000,000 new jobs as a major accomplishment during his 8 years as president. Given the need for dealing with this huge influx of labor, the Chinese government seems to be willing to experience some lessening in the output of its industries in exchange for increased employment of labor. In this dissertation we develop methods to identify ways in which the management of congestion can be improved. This makes it possible to reduce the output decreases that would otherwise occur without reducing the labor inputs that are identified as congesting. We thus accommodate the twin objectives. (i) improving the methodologies of DEA and (ii) improving the management of congestion so that the output reductions which would otherwise be experienced are diminished. This leads to further possible ways to deal with congestion and the employment to which it is related. For instance, we show how to further improve the management of congestion by identifying possible tradeoffs between input reductions (or increases) and output increases (or decreases) so that better decisions can be made in coordinating the two. This leads to still further possibilities since input reduction in one company or plant may be traded off with input augmentations in another company or plant in ways that can simultaneously improve both the input and output performances. All of the methods and concepts developed in this dissertation are also tested by actual data in selected Chinese industries with significance of their utilizations.