Predicting emissions rates for the Atlanta on-road light duty vehicular fleet as a function of operating modes, control technologies, and engine characteristics

In 1996, the USEPA declared the Atlanta Metropolitan Region a non-attainment area for ozone. As a result, federal funding for new road construction projects has since been withheld until the region's metropolitan planning organization (MPO) is able to show with concrete actions, how it plans to bring the region back into compliance with EPA mandated air quality standards.; The air quality problem in Atlanta is not unique; across the United States many large urban centers are confronted with similar health effects from poor air quality, and the economic fallout from suspension of needed federal dollars for development and growth. When urban areas exceed air quality standards, transportation and air quality planners must adopt specific and documentable transportation plans, projects and programs that will lead to a reduction in harmful emissions and comply with set standards. To make the right kind of decisions, policy makers and planners have to depend on accurate estimates of the emission inventory in any given region. Wrong estimates might have two results: costly or unnecessary strategies if overestimated and, serious health implications, if underestimated. Unfortunately, recent research by the US EPA and others have determined that current motor vehicle emission rate predicting models are deficient in large part because they do not account for “modal” vehicle activities like acceleration, deceleration, and engine load.; Georgia Tech's MEASURE model is one of a handful of emerging modal emissions models aimed at addressing these deficiencies. This dissertation has focused on developing more accurate algorithms that will be embedded within MEASURE for use in predicting emissions from motor vehicles. Using state of the art statistical techniques that combine the best aspects Hierarchical Tree Based Regression (HTBR) and Ordinary Least Squares Regression (OLSR), as well as other modeling techniques, three models were developed for use in forecasting emissions of CO, HC, and NOx from the current light duty vehicle fleet in Atlanta. The new modal emission rates indicate that vehicle technologies and vehicle operating profiles (modal activities) have significant impacts on emission rates. By using statistical weights in the modeling process, it is shown that the models are applicable in any region nationwide.; Validation with an external data set, and comparative analysis show these new models to be superior to current models on the basis of three statistical criteria. To demonstrate applicability, the models were used to assess the air quality benefit of an intelligent transportation system (ITS) technology at an intersection in Mid-Town Atlanta.