Data-Driven Estimation of Heavy-Truck Residual Value at the Buy-Back

Cerquitelli, Tania; Regalia, Andrea; Manfredi, Emanuele; Conicella, Fabrizio; Bethaz, Paolo; Liore, Elena · 2020 · DOAJ

DOI: 10.1109/ACCESS.2020.2998940

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of accurately estimating the residual value of heavy trucks at the end of leasing contracts (buy-back). Traditional methods rely on coarse benchmarks like mileage and age, which fail to capture the actual wear and tear of high-cost commercial vehicles. Motivated by the rise of connected vehicles and telematics, the authors propose a data-driven, unsupervised methodology to assess vehicle degradation using real-time usage data. This approach aims to provide leasing companies with objective, interpretable insights to optimize marketing strategies, maintenance decisions, and pricing for re-sold vehicles. The study utilizes a dataset comprising telematics data from over 10,000 heavy trucks monitored over three years. The methodology involves five steps: data exploration, correlation analysis, residual value estimation, outlier detection, and dashboard creation. Initially, 38 features were reduced to nine significant numeric attributes through Pearson correlation analysis and domain expert input, eliminating redundant variables. The core contribution is the development of two estimation formulas: a "data-driven" formula that calculates degradation based on the ratio of measured parameter values to their mechanical asymptotic limits, and a "business-driven" formula that applies weighted scores to parameters based on their perceived impact on wear (e.g., fuel consumption and hours of travel are weighted higher than engine brake usage). Both formulas account for vehicle inactivity depreciation. Additionally, the DBSCAN clustering algorithm is employed to identify outliers, representing trucks with the lowest residual values requiring immediate attention. Experimental evaluation demonstrates that the proposed methodology effectively differentiates trucks that traditional metrics might classify similarly. The analysis revealed significant variation in truck usage patterns, validating the need for granular telematics data. The resulting scores allow for the categorization of trucks into three classes (basic, comfort, premium), enabling targeted business strategies. For instance, "premium" trucks with low wear can undergo aesthetic improvements for higher resale value, while "basic" trucks require only minimal functional interventions. The system was deployed as an interactive Power BI dashboard, allowing users to visualize residual value scores, compare specific trucks against peers with similar market and model years, and identify worst-performing vehicles. The significance of this work lies in its transition from static, benchmark-based valuation to dynamic, usage-based assessment for heavy commercial vehicles. By leveraging unsupervised learning and telematics, the approach provides a more accurate reflection of vehicle condition, supporting better financial decisions for leasing companies. The study confirms that integrating internal usage parameters with external conditions yields superior differentiation compared to traditional methods, offering a scalable solution for the automotive industry’s shift toward data-driven business models.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• exposure measurement