On-road heavy-duty diesel exhaust particulate mass emissions

New speed correction factor (SCF) and operational correction factor (OCF) models of gram per mile diesel particulate emissions from on-road heavy duty vehicles are developed. These models adjust base emission rates to different driving conditions. The OCF model is derived first using a mixed log linear regression model fit to 531 chassis dynamometer tests, with covariates derived from a measure of high power transient driving, time spent idling, and average speed. Emissions rates increase with high power transient driving and decrease as average speed or time idling increases. The model is capable of predicting relative changes in on-road heavy-duty diesel PM emission rates on driving cycles that are different than the driving cycles used to fit the model.; SCF models are then estimated using a cubic polynomial function of speed, both with and without facility type as a covariate. The models are estimated on vehicle activity/operation data from GPS instrumented trucks, with normalized emission rates estimated using the OCF model. Facility type is a significant variable at the 95% confidence level, however the variation attributed to facility classifications is of little practical significance. The new link level SCF curves have a different shape than the trip based SCF curves assumed in emission inventory models. The new SCF model recreates observed laboratory data better than existing trip based SCFs, and compares well to field data. The new link level SCF model is an improved tool for transportation planning applications.