Answers are grounded in this paper and cite their sources.
Fleet environmental management needs scalable ways to identify fuel-intensive operation in dense telematics streams. We analyzed 464,936 hot-engine observations from 14 heavy-duty diesel tractor units monitored by OBD-interface networked terminals and GNSS through cooperation with a Nanjing-based transport company. We asked whether patterns defined in ten development vehicles could support screening in four unseen vehicles. Maximal persistent episodes occupied 10.62% of support but carried 29.17% of the time-integrated OBD fuel-rate signal. Development-defined spatial cells retained excess holdout burden after exposure adjustment, and relative-history profiles remained recognizable in the unseen vehicles. At natural prevalence, operational and fuel histories provided similar ranking information, while their combination yielded a modest improvement. A daily rank cap translated scores into a fixed review workload without overwhelming users with repeated alerts. The findings position remote OBD as a vehicle-external screening tool for prioritizing vehicle–time–location contexts for fleet review and targeted confirmatory measurement.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: