Mixed Integer Programming Models For Selecting Ground-level Ozone Control Strategies

Under the Clean Air Act (CAA) in 1990, the U.S Environmental Protection Agency (EPA) was required to set National Ambient Air Quality Standards (NAAQS) for six common air pollutants that are considered harmful to public health and the environment. Ground–level ozone was one of the six criteria pollutants monitored by the EPA and considered the most widespread health threat. Two precursors of ground level ozone are nitrogen oxides (NO_x) and volatile organic compounds (VOC) and the common sources of these two precursors include on–road vehicles, non–road engines, area sources, point sources, and biogenic and miscellaneous sources. An area where air pollution levels cannot meet the NAAQS persistently is designated as a “Nonattainment area”. A State Implementation Plan (SIP) describes how a state will reduce the emissions of pollutants to satisfy air quality standards in a timely manner. Recently, ozone standards have been revised from a 1–hour to an 8–hour standard by strengthening the threshold from 125 ppb for the previous standard to 85 ppb for the new standard. Due to SIP revision, a total of nine counties are designated as non-attainment for 8–hour ozone standard in the Dallas–Fort Worth (DFW) area, which differed from the original four counties in the earlier 1–hour standard. The main aim of this research is to study both linear and nonlinear mixed integer programming (MIP) models that seek to select targeted control strategies for the DFW region to reduce emissions, so as to achieve SIP requirements with minimum cost. The list of control strategies, along with the emission reduction and cost for each control strategy was obtained from the Texas Commission on Environmental Quality (TCEQ) and the North Central Texas Council of Governments (NCTCOG). Statistics, data mining, and optimization methods are used to determine a potential set of cost–effective control strategies for reducing ozone. These targeted control strategies are specified by different types of emission sources in various time periods and locations. Three MIP models, a static model, a sequential model, and a dynamic model are studied as both linear and nonlinear models. These different MIP models allow decision–makers to study how the targeted control strategies change under different circumstances. Two types of auxiliary variables are considered as supplemental control strategies in the optimization if the current set of control strategies is unable to reduce ozone to comply with the 8–hour ozone standard. Results from the different models can provide decision–makers with information on how the effectiveness of the control strategies vary with daily emission patterns and meteorology.