Genetic Algorithm–Based Approach to Develop Driving Schedules to Evaluate Greenhouse Gas Emissions from Light-Duty Vehicles

Abstract

A city-specific driving schedule is developed for evaluating greenhouse gas emissions [specifically carbon dioxide (CO2)] from light-duty vehicles by using the data collected by a portable emission measurement system and a Global Positioning System device. A genetic algorithm is used to find the optimum driving schedules composed of a combination of microtrips selected from a pool of microtrips. Four types of assessment measures—Type I, driving activity; Type II, driving operating mode distribution; Type III, fuel consumption rate; and Type IV, the product of Types II and III—are adopted in developing the driving schedules. With the MOVES emission modeling approach, the developed driving schedules are used to predict CO2 emissions, which are then compared with and evaluated by the real-world CO2 emissions. Performance evaluation results show that the driving schedules developed by the driving operating mode distribution measure can result in more accurate CO2 emission estimates.