Identifying specifications of in-use vehicles failing the inspection/ maintenance emission test

The objective of this study is to illustrate the vehicle deterioration by usage condition, and identify characteristics of vehicles that fail the Inspection/Maintenance emission test. Using the data sampled from the I/M program of Nanjing, China in 2018, we found vehicle mileage is a better indicator for the deterioration than vehicle age, and newer vehicles have higher deterioration rates than older ones. A random forest model was developed to predict the result of the emission test using vehicle specifications as independent variables. The specifications of vehicles that potentially violate the emission limit were also identified from the model. The result shows vehicles with engine displacement lower than 1.5L, less stringent emission standards, 10-year-old to 17-year-old, or cumulative mileage longer than 120,000 km have a higher probability to fail the test. The proposed method can be used to target on potential high emitters, which improves the efficiency of vehicle inspection.