ÍÃ×ÓÏÈÉú´«Ã½ÎÄ»¯×÷Æ·

Abstract 

Energy conservation measures have been identified as the most cost effective way to reduce carbon emissions. However, a lack of available information regarding energy conservation prevents building owners from investing in energy efficiency. This research provides the groundwork for supplying building owners with a simple model to guide retrofit decisions in their buildings. Using readily available characteristics of buildings that have received a lighting retrofit, the achieved reduction in energy consumption was analyzed using a classification and regression tree. This statistical method determines which building attributes are most related to reduction in energy use. The results of this process show that simple building attributes, including square footage, business type, and vintage, are only responsible for a small portion of the variation in energy-use reduction following energy conservation measures. However, the classification and regression tree and the random forest methods provide insight into how building attributes can be used to explain energy reduction following a lighting retrofit. With more attributes, these simple visual tools may show how energy efficiency analysts can communicate potential savings to building owners, reducing costs to owners and carbon emissions in the atmosphere.