Establishing the DEA energy management system for individual departments within universities

摘要

Purpose - The purpose of this study is to present the theoretical framework of the "data envelopment analysis (DEA) Energy Management System (DEMS)" proposed to assist individual departments occupying the same buildings on university campus in assessing the energy efficiencies of their facilities, as well as to demonstrate the implementation results of the DEMS applied in the case of the Department of Architecture of NTUST in Taiwan. Design/methodology/approach - The proposed DEMS considers each "space" within a department in a given "time" (such as a month) as a decision-making unit (DMU). Then, regression analysis is performed on data of "existing environment", "occupancy" factors and "actual energy consumption EUI (energy usage intensity)" related variables. The regression equation derived is then used to calculate the "predicted EUI" for all DMUs. The "actual EUI" is further considered as the input data and the "predicted EUI" as the output data of the DEMS, on which data envelopment analysis is conducted to produce three types of energy-efficiency scores (overall efficiency, scale efficiency, pure technical efficiency) to indicate the energy efficiencies of all DMUs. Findings - The DEMS was developed and further implemented in the Department of Architecture of NTUST in Taiwan to illustrate how it can be used to assist individual departments within universities in assessing the energy management effectiveness of their spaces. Research limitations/implications - The accuracy of the energy-efficiency scores depends greatly on the accuracy of the predicted EUIs of spaces, and, therefore, it is critical to identify a better regression model with higher predictability (R~2). The relatively low actual EUIs of certain student spaces during winter and summer breaks may greatly affect the resulting energy-efficiency scores. Practical implications - The DEMS allows facility managers to assess and compare the energy-efficiency scores "among different spaces", to further review the energy efficiency of a space "over time" and to recognize the benchmark cases and pursue actions for energy improvement. Originality/value - This study explores the research concepts of "space type" and "internal benchmark" with an analytical method "data envelopment analysis" to assess the energy efficiency of an individual department which may only occupy certain floors of a building.