Leveraging Advanced AI/ML Algorithms to Derive Actionable Insights from Fleet Vehicle In-Use Performance Ratio Monitoring (IUPRm) Data

In recent years, growing concerns over environmental pollution have led to the implementation of increasingly stringent emission regulations worldwide. These regulations require not only the reduction of pollutant emissions but also robust monitoring of critical emission control components under real-world driving conditions. The In-Use Performance Ratio Monitoring (IUPRm) framework quantifies how frequently onboard diagnostic systems evaluate these components within defined operational boundaries for each vehicle. IUPRm is organized into several monitoring groups, including catalyst monitoring, oxygen sensor monitoring, exhaust gas recirculation (EGR) monitoring, and gasoline particulate filter monitoring. The specific groups and monitoring requirements vary based on fuel type (gasoline or diesel), engine technologies (such as variable valve timing or EGR), and exhaust system configurations (single or dual bank). For automakers, analyzing these parameters across large vehicle fleets is a complex and data-intensive challenge. To address this, Tata Motors has developed advanced Artificial Intelligence (AI) and Machine Learning (ML) algorithms that integrate K-Means clustering with structured decision logic, significantly enhancing IUPRm analysis. This approach leverages centroid-based grouping and rule-based classification algorithms to extract actionable insights from extensive fleet data. Additionally, event-based data analysis enables more granular and detailed performance assessments. Collectively, these techniques automate and accelerate large-scale IUPRm analysis, substantially reducing manual effort and enabling faster, more reliable compliance with regulatory requirements. This paper presents the effective application of AI and ML algorithms for rapid and scalable IUPRm data analysis, demonstrating a significant reduction in manual workload while improving the accuracy and depth of emission performance monitoring.