Date of Graduation
Statler College of Engineering and Mineral Resources
Civil and Environmental Engineering
A great deal of attention is being paid worldwide to particulate matter (PM), which is now considered a significant component of air pollution. Specifically, in this thesis, road dust is a primary source of PM that is having a significant impact on human health and air quality. For example, impaired visibility due to road dust can cause more vehicle accidents. Hence, in order to efficiently develop PM control strategies, it is critical to improve the estimation of PM concentration levels generating from paved and unpaved roads. Since 1979, the U.S. Environmental Protection Agency (EPA) has developed emission factor equations to quantify the magnitude of PM for paved and unpaved roads based on multiple linear regression (MLR) models. However, the MLR models are not suitable for PM data that exhibit the characteristics of complexity and non-linearity, thereby limiting the predictive accuracy of MLR to estimate PM. The objective of this thesis is to present a method to improve the quality of the existing EPA emission factor equations for paved and unpaved roads by employing an artificial neural network (ANN). The proposed method consists of the following steps: data processing for outliers, data normalization, data classification, ANN model training to determine the weights of emission factors identified, and method validation through additional data testing. This thesis included a case study using the data retrieved from the database used by the EPA to generate their emission factor equations for paved and unpaved roads. The proposed method was evaluated by demonstrating its improved performance as shown in the coefficient of determination (R2) and the root mean square error (RMSE) values compared to the values obtained with the existing EPA emission equations. The empirical findings of the case study verified that the proposed method using the ANN model is capable of improving the quality of the EPA emission equations, resulting in higher R 2 and lower RMSE values for both paved and unpaved roads. The expected significance of this thesis is that the proposed method improves the ability to develop more reliable emission factors for predictable PM levels that can help agencies establish enhanced PM control strategies. In addition, the method may have application in other fields that require a selection process to identify an optimal combination of input variables.
Liu, Tse-Huai, "Development of Enhanced Emission Factor Through the Identification of an Optimal Combination of Input Variables Using Artificial Neural Network" (2018). Graduate Theses, Dissertations, and Problem Reports. 6104.