ASSESSMENT OF EFFICIENCY AND PRODUCTIVITY OF PUBLIC TRANSPORT SECTOR THROUGH DEA TECHNIQUES

Abstract

performing decision making units (DMUs) without requiring a prior specification of the technologies or information on input and output prices. It can handle multiple inputs and outputs and doesn't require any assumption of a functional form relating inputs to outputs. Thus, it is well suited for comparative performance analysis of public transport sector. The basic DEA models cannot assess the impact of slacks on efficiency scores. Also the results obtained by these models show that many multipliers have the zero value which indicate that the corresponding variables (input/output) have not been fully utilised in the assessment of the efficiency scores. To overcome this problem, DEA model has been extended. This study also attempts to make extension in the DEA models and apply them to measure the efficiency of the transport sector.. By a critical examination of the, available ,literature, it is found, that. DEA-based studies dealing with the relative efficiency of public transport sector in India are extremely limited. The present study attempts to assess the performance of the, public ,transport sector of Uttar Pradesh State through DEA technique. 'The . new slack, model is: developed and applied to determine the impact of slacks on efficiency. of the "regions" of the Uttar Pradesh State Road Transport Corporation (UPSRTC). The study also measures the impact of various background variables on the efficiency using the Tobit regression model. Sensitivity analysis is conducted to identify the outliers on the frontier, verify the robustness of the efficiency scores and ranking the efficient regions. Time series data for the period 1972-1973 to 2002-2003 have been used to assess the growth trends and efficiency patterns of the UPSRTC. The cross-sectional data required for efficiency analysis of regions of the UPSRTC are collected from the annual report of the UPSRTC for the year 2004-2005. Data for productivity measurement are obtained from the iv annual reports of the UPSRTC for the years 1990-1991 to 2002-2003. We estimate the production correspondence between three inputs and single output. Fleet-Size (FS), Total Staff (TS) and the Fuel Consumption (Fuel) are considered as input variables and Passenger-kilometers (Pass km) as output variable. An attempt is also made to incorporate a comprehensive list of inputs and outputs that reflect the general, informative and robust results. The study aims to: i. analyse the growth trends and efficiency pattern of the UPSRTC, ii. propose a new slack model in which impact of slack inefficiencies are included, iii. measure the relative efficiencies of the regions of the UPSRTC, ak iv. identify use of excess input and deficient output for the inefficient regions, v. measure the effect of several background variables on the performance of the regions, vi. verify the robustness of the efficiencies of the regions, and vii. assess the total factor productivity changes, technical efficiency and technical changes in the regions. The chapter-wise summary of the thesis is as follows Chapter 1 is introductory in nature. It deals with the aspects of performance measurement, statement of the problem, objectives, scope, and limitations of the study. This chapter also discusses briefly about the DEA techniques used to assess the efficiency and productivity. Chapter 2 is devoted to the review of relevant DEA and MPI based studies on the transport sector, other studies on the transport sector, and DEA-based foreign and Indian studies on the other sectors. Chapter 3 shows the advancements in DEA techniques from the basic models to the new DEA based models. This chapter presents the basic (CCR and BCC) model, AR model, DEA models based on slacks such as Additive model, Two Phase model, SBM model, SA model and DEA models with non-discretionary variables and categorical DMUs. Chapter 4 presents, in brief, the history, growth trends, partial productivity and efficiency pattern of the public transport sector of Uttar Pradesh. Time series study of the UPSRTC covers a period from 1972-73 to 2002-03. This chapter also examines the progress and changes in the performance of the UPSRTC during pre- and post-economic reform periods of India. In chapter 5, we propose a new slack model of DEA to evaluate the efficiency with the actual impact of slacks on the efficiency scores. The chapter also deals with the theoretical aspects of the new slack model with its empirical illustration for analysing the performance of the regions. Tobit regression and sensitivity analysis have been carried out in Chapter 6. In this chapter, we propose a new model for sensitivity analysis to determine the robustness of the efficiency scores obtained by the new slack model. This model is also useful for ranking of the efficient DMUs. Chapter 7 deals with the theoretical aspects of Malmquist Productivity Index (MPI) and assesses the total factor productivity change, technical efficiency and technical changes that have taken place in regions of the UPSRTC. It also examines the important reasons of low productivity in the public transport sector of the state. DEA based MPI approach is used