TireFuture: Development of an ML-based software tool for determining the ideal truck tire configuration based on vehicle data and driving profiles.

The project focuses on developing an intelligent ML-based software tool that automatically determines the ideal tire configuration for trucks, based on vehicle data and actual driving profiles. By optimizing both tire selection and tire pressure, fuel consumption can be reduced by up to 10%. This enables logistics companies and fleet operators to significantly cut operating costs while measurably lowering their CO₂ emissions.

At the core of the system is a multi-stage analysis of the available tire types. First, the different tire categories are organized using heuristic methods in a hierarchical classification. Subsequently, an ML model – most likely a neural network with heteroscedastic regression – compares tires with similar profiles in terms of their impact on fuel consumption under specific driving profiles.

Based on these analyses, the tool generates a fuel-consumption-optimized tire recommendation for each individual driving profile. The system leverages existing sensor and vehicle data and combines them with state-of-the-art ML techniques to enable robust, data-driven decisions. In this way, the solution helps logistics companies optimize their fleets from both an economic and environmental perspective.