Data Envelopment Analysis for Management and Non-economists

This course is for those who are new to quantitative research and interested in productivity, benchmarking and optimisation.

Level 1 - runs over 5 days
Course dates: Monday 25 September 2017 - Friday 29 September 2017
Course status: Course completed (no new applicants)
Week 1
About this course: 

The main focus of this course is on learning basic data analysis tools through hands-on experience. Participants will systematically apply software tools to the tasks of developing a research project including: planning for collection of data, preparation of data for analysis and beginning analysis. They will also be introduced to advanced analysis tools including those for theory building, validation and presentation of findings.


Participants will explore applications of the software to their own research projects. Sample data will be provided but participants should bring their own data sets. The software allows up to 25 decision making units to be analysed.


The target audience for this course is researchers in the management, service operations management fields or researchers interested in productivity, operations research and optimisation. This course is suitable for postgraduate students and supervisors wanting to apply operations research methods without the underlying knowledge of mathematics.

Course syllabus: 

Day 1
Orientation and introduction to data envelopment analysis (DEA). DEA is a method used in economics and operations research to measure productivity and performance in complex multi-input, multi-output organisations.
DEA and its key terms are introduced, defined and applied. Frontier analysis and the concept of return to scale are explained. Examples of how DEA has been applied across a number of industries such as banking, health and tourism will also be discussed.


Day 2
A range of DEA models, such as input oriented and output oriented models, will be discussed. Advanced DEA models including multiplicative, additive, cone ration, assurance regions, super efficiency, two stage (network DEA) and dynamic DEA are also introduced and discussed.

Issues related to measures, data collection and data preparation techniques will be highlighted.


Day 3

An orientation of the DEA Frontier software will be discussed along with file formatting, data sets and applying DEA. Participants will run a number of simple DEA models, generating results.  In addition, an introduction to weight restrictions will be discussed.


Day 4
A range of different data files will be run and interpreting and presenting results will be discussed. The interpretation of efficiency scores, peer benchmarks, slacks and targets will be shown through a range of practical examples.


Day 5
Comparisons and robustness of results and advanced topics such as Malmquist index, bootstrapping and sensitivity analysis are introduced. One on one consultation will also be available.


Course format: 

This course will take place in a computer lab or a classroom with BYO laptops. 

You will be advised prior to the commencment of the course. An educational version of Solver and DEA Frontier software will be provided to those wishing to use their own laptops.

Recommended Background: 

Efficiency in using Windows based software and Excel is essential. No prior knowledge of DEA is required.

Recommended Texts: 

Further reading:

  • Avkiran, N 1999, “An application reference for data envelopment analysis in branch banking: helping the novice researcher”, International Journal of Bank Marketing, vol. 17, no. 5, pp. 206–220.
  • Sarkis, J 2007, “Preparing your data for DEA”, in J Zhu & W Cook (eds), Modeling data irregularities and structural complexities in data envelopment analysis, Springerlink, New York, pp. 305–320.
  • Scerri, M & Agarwal, R 2015, “Service enterprise productivity in action [SEPIA]”, in A Emrouznejad & E Cabanda (eds), Service productivity handbook, vol. 2015, Springer, London.
  • Zhu, J 2000, “Multi-factor performance measure model with an application to Fortune 500 companies”, European Journal of Operational Research, vol. 123, no. 1, pp. 105–124.
Course fees
Non Member: 
Full time student Member: 

Q: Do I have to have used Data Envelopment Analysis (DEA) before?

A: No prior knowledge of DEA is required


Q: Do I need a background in doing quantitative research?

A: No, the use of Excel means there is a familiar interface.


Q: Do I need to know mathematics or linear programming?

A: No we use DEA Frontier software and the course will explain the concepts and interpretation of results.

Spring Program 2017

The instructor's bound, book length course notes will serve as the course text.