Energy Efficiency Improvement in Office Buildings by Benchmarking Models

Benchmarking energy efficiency is an important tool to promote the efficient use of energy in buildings that can be considered energy consumers of municipal energy systems. Benchmarking models are mostly constructed in a simple benchmark table (percentile table) of energy-use intensities, which can be developed by means of simple normalization (simple) with floor area, regression analysis (RA) to normalize not only floor area but also some explanatory factors (e.g., operating hours), stochastic frontier analysis (SFA), and data envelopment analysis (DEA). Because of the properties of these four mathematical methods, we can have two kinds of benchmarking systems, public and internal benchmarking systems. Simple, RA, and SFA can be used to develop the public benchmarking systems by a set of reference buildings, which can function as a public yardstick of energy-use performance in buildings. That is, other nonreference building owners (including reference one) can use the resulting yardstick to benchmark their performance. On the other hand, DEA can be applied to develop the internal benchmarking system that cannot be used by public users, and only the reference buildings can be benchmarked. Depending on the benchmarking purposes, we can find that both types of benchmarking models can provide the benchmarking results to encourage poor reference performers to improve their performance. For example, energy use of public service area of office buildings can be improved by applying the internal benchmarking system to the chiller system coefficient of performance, while tenants' energy-use performance by the public benchmarking systems. © 2015 John Wiley & Sons, Ltd. All rights reserved.