Weight Restrictions on Geography Variables in the DEA Benchmarking Model for Norwegian Electricity Distribution Companies
Type/no
R33/08
Author
Endre Bjørndal, Mette Bjørndal og Ana Camanho
The DEA model for the distribution networks is designed to take into account the diverse operating conditions of the companies through so-called “geography” variables. Our analyses show that companies with difficult operating conditions tend to be rewarded with relatively high efficiency scores, and this is the reason for introducing weight restrictions.
We discuss the relative price restrictions suggested for geography and high voltage variables by NVE (2008), and we compare these to an alternative approach by which the total (virtual) weight of the geography variables is restricted. The main difference between the two approaches is that the former tends to affect more companies, but to a lesser extent, than the latter. We also discuss how to set the restriction limits. Since the virtual restrictions are at a more aggregated level than the relative ones, it may be easier to establish the limits with this approach. Finally, we discuss implementation issues, and give a short overview of available software.
We discuss the relative price restrictions suggested for geography and high voltage variables by NVE (2008), and we compare these to an alternative approach by which the total (virtual) weight of the geography variables is restricted. The main difference between the two approaches is that the former tends to affect more companies, but to a lesser extent, than the latter. We also discuss how to set the restriction limits. Since the virtual restrictions are at a more aggregated level than the relative ones, it may be easier to establish the limits with this approach. Finally, we discuss implementation issues, and give a short overview of available software.
Language
Written in english