All Issue

2026 Vol.25, Issue 2 Preview Page
Decomposing Energy Inefficiency in Non-Residential Buildings Using Hourly Electricity Data: A Frontier Analysis of Operational and Structural Efficiency

시간별 전기소비 자료를 이용한 비주거 건물의 에너지 비효율 분해: 운영효율과 구조효율의 프런티어 분석

Giseon Kim1
,
Jongwoo Kim2*
,
Sangkyu Park3*
김 기성1
,
김 종우2*
,
박 상규3*
1Assistant Research Fellow, Center for Energy Information and Statistics, Korea Energy Economics Institute (First author)
2Research Fellow, Climate and Energy Policy Research Group, Korea Energy Economics Institute (Co-author)
3Assistant Professor, Department of Business Administration, Hallym University (Corresponding author)
1에너지경제연구원 에너지조사통계연구실 부연구위원 (주저자)
2에너지경제연구원 에너지효율정책연구실 연구위원 (공동저자)
3한림대학교 경영대학/인문사회융합연구원 조교수 (교신저자)
*Corresponding Author
†These authors contributed equally to this work.
30 September 2026. pp. 223-267
PDF XML Citation
Abstract

Improving electricity efficiency in the building sector requires more than identifying high-consuming buildings, because observed consumption reflects not only unavoidable demand drivers—scale, use type, weather, and occupancy—but also the characteristics of the envelope and equipment and the way buildings are operated. Annual aggregate indicators therefore obscure the magnitude, source, and timing of inefficiency. This study assembles a balanced panel of 186 non-residential public buildings in Korea (office, education/research, assembly, and medical uses) for 2022, combining hourly electricity consumption, building-specific weather data, and the Energy Consumption Survey microdata (1,629,360 observations). We sequentially apply three frontier approaches—a conditional minimum-demand frontier based on change-point quantile regression, a four-component stochastic frontier, and a load-shape frontier—to decompose building energy inefficiency into operational and structural dimensions. Relative to the conditional minimum load attainable under identical weather, time-of-day, and operating states, excess consumption is estimated at 22.8% of total sample consumption and is concentrated in weekday daytime operating hours. The correlation between annual aggregate efficiency and hourly operational efficiency is only 0.19, confirming that annual indicators fail to capture time-varying operational inefficiency. Decomposing inefficiency into persistent (structural) efficiency (PE) and transient (operational) efficiency (TE), the two frontier approaches converge at a rank correlation of 0.833, and the realizable operational saving—after separating noise—is bounded within 181-301 GWh per year. An exploratory cluster analysis on the two efficiency scores classifies the sample into an efficient group (66%), an operation-improvement group (13%), and a structure- improvement group (21%); buildings that are simultaneously low on both dimensions are virtually absent, so a single prescription applies to most buildings. The groups are distinguished not by building size or vintage but by load shape and consumption intensity. These findings suggest a shift from uniform efficiency support toward diagnosis-based differentiated support, the institutionalization of operational improvement prior to capital investment in equipment, and the separation of aggregate energy saving and peak management as distinct policy objectives.

Keywords
Building energy efficiency
Stochastic frontier analysis
Change- point quantile regression
Hourly electricity consumption
Operational inefficiency
Structural inefficiency
Load shape

건물 부문의 전력효율 개선은 소비량이 많은 건물을 선별하는 것만으로는 이루어지기 어렵다. 관측된 소비에는 규모･용도･기상･재실과 같은 불가피한 수요요인과 함께 설비･외피의 특성 및 운영형태가 반영되어 있어, 연간 총량 지표만으로는 비효율의 크기와 원인, 발생 시점을 구분하기 어렵기 때문이다. 본 연구는 업무･교육연구･문화집회･의료 용도의 국내 비주거 건물 186개동을 대상으로 2022년 시간별 전력소비, 건물별 기상자료, 에너지총조사 자료를 결합한 균형패널을 구축하고, 조건부 최소수요 프런티어(변화점 분위수회귀), 4-성분 확률 프런티어, 부하형상 프런티어를 단계적으로 적용하여 건물 에너지 비효율을 운영적 차원과 구조적 차원으로 분해한다. 분석 결과, 동일한 기상･시간대･운영상태에서 달성 가능한 조건부 최소부하 대비 초과소비는 표본 전체 소비량의 22.8%로 추정되었으며, 주로 평일 주간 영업시간대에 집중되었다. 연간 총량효율과 시간 단위 운영효율의 상관은 0.19에 그쳐, 연간 지표만으로는 시간대별 운영 비효율을 포착하기 어려움을 확인하였다. 비효율을 구조효율(PE)과 운영효율(TE)로 분해한 결과 두 프런티어 접근의 순위상관은 0.833으로 나타났고, 잡음을 제거한 실현 가능 운영 절감 잠재량은 연간 약 181 GWh로 도출되었다. 두 효율지표를 이용한 탐색적 군집분석은 표본을 효율 양호군(66%), 운영개선 대상군(13%), 구조개선 대상군(21%)으로 유형화하였으며, 구조효율과 운영효율이 동시에 낮은 유형은 관측되지 않아 표본건물별로 우선 검토할 개선 방향을 구분할 수 있었다. 유형을 구분하는 요인은 건물의 규모나 연식이 아니라 부하형상과 소비강도였다. 이러한 결과는 일괄적 효율지원에서 비효율의 원인에 따른 진단 기반 차등지원으로의 전환, 설비투자에 선행하는 운영개선의 제도화와 그린버튼 플랫폼을 통한 지속적 관리가 필요함을 시사한다.

키워드
건물 에너지효율
확률 프런티어 분석
변화점 분위수회귀
시간별 전력소비
운영 비효율
구조 비효율
부하형상
References
1

김한주･주형빈･김덕우･허연숙. (2024). 교육시설 에너지 벤치마킹을 위한 연간 기저 및 난방 에너지 영향 인자 분석. 한국건축친환경설비학회논문집, 18(6), 491-501.

2

신동현･임형우･조하현. (2019). 확률경계모형(SFA)을 이용한 국내 주택용 난방에너지 소비효율 비교 분석. 한국경제연구, 37(2), 151-186.

10.46665/jkes.2019.06.37.2.151
3

신동현･조하현･임형우. (2020). 메타경계모형을 이용한 개별 가구의 난방 효율성 분석. 한국경제연구, 38(4), 121-142.

10.46665/jkes.2020.12.38.4.121
4

오지현･김선숙. (2026). 의료시설 에너지 벤치마킹을 위한 사용량과 영향 변수 분석. 대한건축학회논문집, 42(3), 273-282.

5

주형빈･허연숙. (2024). 도서관 건물의 에너지 벤치마킹을 위한 SHAP 기반 주요 인자 분석. 대한건축학회 학술발표대회논문집, 44(2), 519-522.

6

Aigner, D. J., Lovell, C. A. K., Schmidt, P. (1977). Formulation and estimation of stochastic frontier production function models. Journal of Econometrics, 6(1), 21-37.

10.1016/0304-4076(77)90052-5
7

Alberini, A., Filippini, M. (2018). Transient and persistent energy efficiency in the US residential sector: Evidence from household-level data. Energy Efficiency, 11(3), 589-601.

10.1007/s12053-017-9599-z
8

Andrews, A., Jain, R. K. (2022). Beyond energy efficiency: A clustering approach to embed demand flexibility into building energy benchmarking. Applied Energy, 327, 119989.

10.1016/j.apenergy.2022.119989
9

Arjunan, P., Poolla, K., Miller, C. (2020). EnergyStar++: Towards more accurate and explanatory building energy benchmarking. Applied Energy, 276, 115413.

10.1016/j.apenergy.2020.115413
10

American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE). (2014). ASHRAE Guideline 14-2014: Measurement of Energy, Demand, and Water Savings. ASHRAE.

11

Azzalini, A. (1985). A class of distributions which includes the normal ones. Scandinavian Journal of Statistics, 12(2), 171-178.

12

Borgstein, E. H., Lamberts, R., Hensen, J. L. M. (2016). Evaluating energy performance in non-domestic buildings: A review. Energy and Buildings, 128, 734-755.

10.1016/j.enbuild.2016.07.018
13

Buck, J., Young, D. (2007). The potential for energy efficiency gains in the Canadian commercial building sector: A stochastic frontier study. Energy, 32(9), 1769-1780.

10.1016/j.energy.2006.11.008
14

Burlig, F., Knittel, C., Rapson, D., Reguant, M., Wolfram, C. (2020). Machine learning from schools about energy efficiency. Journal of the Association of Environmental and Resource Economists, 7(6), 1181-1217.

10.1086/710606
15

Chung, W. (2011). Review of building energy-use performance benchmarking methodologies. Applied Energy, 88(5), 1470-1479.

10.1016/j.apenergy.2010.11.022
16

Colombi, R., Kumbhakar, S. C., Martini, G., Vittadini, G. (2014). Closed-skew normality in stochastic frontiers with individual effects and long/short-run efficiency. Journal of Productivity Analysis, 42(2), 123-136.

10.1007/s11123-014-0386-y
17

Fels, M. F. (1986). PRISM: An introduction. Energy and Buildings, 9(1-2), 5-18.

10.1016/0378-7788(86)90003-4
18

Filippini, M., Greene, W. (2016). Persistent and transient productive inefficiency: A maximum simulated likelihood approach. Journal of Productivity Analysis, 45(2), 187-196.

10.1007/s11123-015-0446-y
19

Greene, W. (2005). Reconsidering heterogeneity in panel data estimators of the stochastic frontier model. Journal of Econometrics, 126(2), 269-303.

10.1016/j.jeconom.2004.05.003
20

Jondrow, J., Lovell, C. A. K., Materov, I. S., Schmidt, P. (1982). On the estimation of technical inefficiency in the stochastic frontier production function model. Journal of Econometrics, 19(2-3), 233-238.

10.1016/0304-4076(82)90004-5
21

Kavousian, A., Rajagopal, R. (2014). Data-driven benchmarking of building energy efficiency utilizing statistical frontier models. Journal of Computing in Civil Engineering, 28(1), 79-88.

10.1061/(ASCE)CP.1943-5487.0000327
22

Koenker, R., Bassett, G. (1978). Regression quantiles. Econometrica, 46(1), 33-50.

10.2307/1913643
23

Kumbhakar, S. C., Lien, G., Hardaker, J. B. (2014).Technical efficiency in competing panel data models: A study of Norwegian grain farming. Journal of Productivity Analysis, 41(2), 321-337.

10.1007/s11123-012-0303-1
24

Kwac, J., Flora, J., Rajagopal, R. (2014). Household energy consumption segmentation using hourly data. IEEE Transactions on Smart Grid, 5(1), 420-430.

10.1109/TSG.2013.2278477
25

Luo, X., Hong, T., Chen, Y., Piette, M. A. (2017). Electric load shape benchmarking for small-and medium-sized commercial buildings. Applied Energy, 204, 715-725.

10.1016/j.apenergy.2017.07.108
26

Meeusen, W., van den Broeck, J. (1977). Efficiency estimation from Cobb-Douglas production functions with composed error. International Economic Review, 18(2), 435-444.

10.2307/2525757
27

Meng, Q., Mourshed, M., Wei, S. (2018). Going beyond the mean: Distributional degree-day base temperatures for building energy analytics using change point quantile regression. IEEE Access, 6, 39532-39540.

10.1109/ACCESS.2018.2852478
28

Park, J. Y., Yang, X., Miller, C., Arjunan, P., Nagy, Z. (2019). Apples or oranges? Identification of fundamental load shape profiles for benchmarking buildings using a large and diverse dataset. Applied Energy, 236, 1280-1295.

10.1016/j.apenergy.2018.12.025
29

Paulus, M. T., Claridge, D. E., Culp, C. (2015). Algorithm for automating the selection of a temperature dependent change point model. Energy and Buildings, 87, 95-104.

10.1016/j.enbuild.2014.11.033
30

Piscitelli, M. S., Giudice, R., Capozzoli, A. (2024). A holistic time series-based energy benchmarking framework for applications in large stocks of buildings. Applied Energy, 357, 122550.

10.1016/j.apenergy.2023.122550
31

Roth, J., Rajagopal, R. (2018). Benchmarking building energy efficiency using quantile regression. Energy, 152, 866-876.

10.1016/j.energy.2018.02.108
32

Taylor and Letham. (2017). Forecasting at scale. Meta Github. https://facebook.github.io/prophet/static/prophet_paper_20170113.pdf

33

Waldman, D. M. (1982). A stationary point for the stochastic frontier likelihood. Journal of Econometrics, 18(2), 275-279.

10.1016/0304-4076(82)90041-0
34

Yang, Z., Roth, J., Jain, R. K. (2018). DUE-B: Data-driven urban energy benchmarking of buildings using recursive partitioning and stochastic frontier analysis. Energy and Buildings, 163, 58-69.

10.1016/j.enbuild.2017.12.040
Information
  • Publisher :Korea Energy Economic Institute·Korea Resource Economics Association
  • Publisher(Ko) :에너지경제연구원·한국자원경제학회
  • Journal Title :Korean Energy Economic Review
  • Journal Title(Ko) :에너지경제연구
  • Volume : 25
  • No :2
  • Pages :223-267
  • Received Date : 2026-08-12
  • Revised Date : 2026-09-21
  • Accepted Date : 2026-09-22
  • DOI :https://doi.org/10.22794/keer.2026.25.2.007
Journal Informaiton Korean Energy Economic Review Korean Energy Economic Review
Online Submission
  • NRF
  • KISTI Current Status
  • KISTI Cited-by
  • ccl
Journal Informaiton Journal Informaiton - close