An Assessment of the Performance of Australian Railways, 1990 to 1998

Supplement to Inquiry Report - Progress in Rail Reform

An Assessment of the Performance of Australian Railways, 1990 to 1998  was released on 13 April 2000. It provides a detailed quantitative analysis of the performance of Australian government-owned railways.

In a spirit of transparency and to facilitate further analysis by the research community, the data and program code used in the data envelopment analysis and a full set of the results of this analysis (presented in chapter 3 and appendices B and E) are made available in an electronic annex. The Commission encourages further use of these data. While every effort has been made to ensure the accuracy of the data, the Commission declines any responsibility for remaining errors and inconsistencies. Also see:

CONTENTS

Preliminaries
Cover, Copyright, Contents, Preface

Introduction

1 The Commission's Approach
1.1 The framework for analysis

2 Data
2.1 Data sources
2.2 Countries and agencies included in the study
2.3 Comparability of railways

3 Measurement of productivity
3.1 Concepts of efficiency
3.2 Data envelopment analysis

4 Performance outcomes for stakeholders
4.1 Consumers
4.2 Shareholders
4.3 Labour

5 Summary of performance
5.1 Productivity
5.2 Stakeholder outcomes
5.3 Conclusion

A Principles of data envelopment analysis
A.1 Measuring productivity
A.2 Selection of DEA method for this study
A.3 DEA methodology
A.4 DEA modelling issues

B Towards robust measures of productivity
B.1 Model selection
B.2 Comparing modelling results

C Railway characterstics
C.1 Railway profiles
C.2 Input-output relationships

D Studies of railway performance
D.1 Studies of railway productivity
D.2 Comparing results
D.3 Other studies of stakeholder outcomes

E A complete set of productivity results

F Independent reference panel reports

References

Electronic Annex

The electronic annex contains a set of program code designed for processing data and estimating DEA scores under different model specifications. The program is written in GAMS syntax. The GAMS software (in conjunction with a linear programming solver) is required to execute the program code. Reference to the GAMS software can be found in the comments attached to the program code.

Also presented here are the data used (tables 1 and 2) and results produced (tables 3 and 4) in the study. On the study results, both DEA scores and estimates of slack variables are tabulated.

CONTENTS

DEA program code

Table 1: Input and output data for models 1, 2 and 3

Table 2: Input and output data for models 4 and 5

Table 3: DEA results for models 1, 2 and 3

Table 4: DEA results for models 4 and 5

Printed copies

Printed copies of the report can be can be obtained from Media and Publications Section.

ISBN:1 74037 090 2

Have your say

We value your comments about this publication and encourage you to complete and submit the publications feedback form.