Benchmarking building energy in the multifamily industry: A data envelopment analysis (DEA) model

Abstract

A new data envelopment analysis (DEA) based approach for benchmarking energy efficiency in buildings in the multifamily sector was proposed in this dissertation. It addressed major limitations of existing DEA model. It provides a method that remediates missing or incorrect values for instances in the dataset, establishes a mechanism that accurately and effectively detects outliers in the dataset, selects appropriate variables to be included in the DEA model and provides justifications for the selection, creates a DEA model that differently handles controllable and non-controllable variables, and quantitatively measures the stability of efficiency scores of each decision making units across the entire period. Data was provided by a third utility management and energy service company in the multifamily housing industry. Research deliverables are expected to provide decision makers and facility managers with the crucial information for building energy improvement. The limitations of future work are also discussed at the end of this dissertation.