• Explain the differences between different benchmarking model specifications and justify the choices of specific models for a given problem context.
• Calculate efficiency scores and identify the corresponding benchmarks, peers & weights in simple examples.
• Interpret the results from large empirical studies and discuss their implications for management.
• Explain the differences between the envelopment and multiplier formulations of the so-called DEA models and their respective uses.
• Argue for the relevance of different model extensions for specific scenarios, including but not limited to the use of weight restrictions and different projections onto the efficiency frontier.

Being able to quantify the efficiency of resource utilisation is important, both in the private sector, due to the intensity of competition, and in the public sector, where budget cuts and demands for efficiency improvments are common.

This course covers specific benchmarking methods that are very useful in practice for assessing the performance of different organisational units, particularly the socalled Data Envelopment Analysis (DEA) approach. As well as providing measures for the extent of efficiency, DEA also identifies role model units which less efficient units can emulate and performance targets at which inefficient units might aim. Benchmarking analysis through the use of DEA thus supports the identification and adoption of operating practices conducive to the efficient utilisation of resources. The approaches are flexible enough to be able to incorporate e.g. ESG indicators in the analysis, and are therefore applicable for efficiency assessments in many different practical settings. They are also used in practice for regulation of natural monopolies within e.g. the utillity sectors (electricity distribution, water and wastewater companies) in Denmark and many other (especially European) countries.

Besides covering both theoretical and practical aspects of the use of DEA, the course also looks at some real life applications, amongst which are studies within e.g. the financial services industries undertaken by the lecturer on the course.

The course will involve combinations of lectures, small examples and exercises done in class and potentially some more realistic problems for students to practice on at home.

The approaches covered in the course are data driven, so the course is highly  quantitative - but utilising linear programming based approaches, so it is mathematical rather than statistical in nature.

Bogetoft and Otto: Benchmarking with DEA, SFA and R. Forlag: Springer-Verlag New York Inc.

Thanassoulis: Introduction to the Theory and Application of Data Envelopment Analysis: A foundation text with integrated software. Forlag: Springer-Verlag New York Inc.