In late 2019, ASHRAE hosted the Great Energy Predictor III (GEPIII) machine learning competition on the Kaggle platform. This launch marked the third energy prediction competition from ASHRAE and the first since the mid-1990s. In this updated version, the competitors were provided with over 20 million points of training data from 2,380 energy meters collected for 1,448 buildings from 16 sources. This competition’s overall objective was to find the most accurate modeling solutions for the prediction of over 41 million private and public test data points. The competition had 4,370 participants, split across 3,614 teams from 94 countries who submitted 39,403 predictions. In addition to the top five winning workflows, the competitors publicly shared 415 reproducible online machine learning workflow examples (notebooks), including over 40 additional, full solutions. This paper gives a high-level overview of the competition preparation and dataset, competitors and their discussions, machine learning workflows and models generated, winners and their submissions, discussion of lessons learned, and competition outputs and next steps. The most popular and accurate machine learning workflows used large ensembles of mostly gradient boosting tree models, such as LightGBM. Similar to the first predictor competition, preprocessing of the data sets emerged as a key differentiator.

Research is needed to explore the limitations and potential for improvement of machine learning for building energy prediction. With this aim, the ASHRAE Great Energy Predictor III (GEPIII) Kaggle competition was launched in 2019. This effort was the largest building energy meter machine learning competition of its kind, with 4,370 participants who submitted 39,403 predictions. The test data set included two years of hourly whole building readings from 2,380 meters in 1,448 buildings at 16 locations. This paper analyzes the various sources and types of residual model error from an aggregation of the competition’s top 50 solutions. This analysis reveals the limitations for machine learning using the standard model inputs of historical meter, weather, and basic building metadata. The errors are classified according to timeframe, behavior, magnitude, and incidence in single buildings or across a campus. The results show machine learning models have errors within a range of acceptability (RMSLE_scaled =< 0.1) on 79.1% of the test data. Lower magnitude (in-range) model errors (0.1 < RMSLE_scaled =< 0.3) occur in 16.1% of the test data. These errors could be remedied using innovative training data from onsite and web-based sources. Higher magnitude (out-of-range) errors (RMSLE_scaled > 0.3) occur in 4.8% of the test data and are unlikely to be accurately predicted.