Predicting Green Building Performance Over Time: Data Mining Untapped Information In LEED
Free (open access)
355 - 365
A. Jeerage, B. Erwine, S. Mallory & V. Agarwal
This paper will present a method for data mining building performance information available in the LEEDTM rating system project applications to show predicted environmental impact and improved performance over time. A study initiated by the City of Seattle analyzed data from over 50 LEED projects within the city limits to quantify the climate change impact of LEED buildings, developing aggregate performance predictions for water, energy and solid waste measures and savings. This paper demonstrates the efficacy of mining LEED derived performance data to make predictions about environmental impact, including carbon reduction, of green building programs using the City of Seattle Performance Evaluation Program as an example. This paper probes key findings regarding the most common green building measures, aggregated savings, the market penetration of specific strategies, and opportunities lost. The LEED Documentation Data Mining Methodology® magnifies its value if used to compare varying building cohorts to identify regional, programmatic, and ownerrelated performance patterns and to inform continuous improvement of sustainable building approaches. The paper explores how this data mining methodology could be adapted and developed for any rating system and any cohort of buildings, and thus utilized to predict sustainable design and related climate impacts of building programs. Keywords: data-mining, sustainability, LEED.
data-mining, sustainability, LEED