A short-trip combination method for reconstructing real-world NOx emission indicators from irregular OBD telematics of heavy-duty diesel vehicles

Heavy-duty diesel vehicles remain an important in-use source of atmospheric NOx, yet long-duration characterization of real-world emissions is constrained by the cost of Portable Emission Measurement Systems (PEMS) and by the irregular structure of Onboard Diagnostics (OBD) telematics. Using 13 months of 1 Hz telematics data from 502 China VI heavy-duty diesel vehicles, covering approximately 1.7 million trips, this study developed a Short-Trip Combination Method (STCM) to reconstruct a screening-oriented estimate of the 90th-percentile Moving Average Window (MAW) specific NOx emission indicator for hot-running operation. STCM segments thermally stabilized operation into short trips, classifies trips by driving mode, reconstructs regulation-constrained synthetic cycles through Monte Carlo resampling, and then calculates the MAW-based NOx indicator. The framework was benchmarked against 15 PEMS tests on four representative vehicles and further compared with a simple whole-trip averaging baseline. A 51-h evaluation window and 200 Monte Carlo iterations provided stable estimates. Median STCM estimates fell within the range of repeated PEMS measurements, with vehicle-level absolute errors of 0.11-0.182 g/kWh, whereas the supplementary baseline comparison showed lower agreement for whole-trip averaging. These results indicate that irregular long-duration OBD data can be converted into environmentally relevant hot-running NOx indicators that recover persistent elevated-emission behavior under real-world operation. Rather than replacing regulatory PEMS testing, the proposed framework provides a practical basis for continuous fleet-scale screening and prioritization of vehicles or operating periods requiring confirmatory inspection.