Building an OEM-agnostic framework for optimizing refueling tasks through ITS paradigms: The P.I.E.N.O. journey

The growing demand for sustainable mobility has highlighted the importance of intelligent solutions for reducing emissions and optimizing energy usage in transportation. Achieving this objective requires the integration of heterogeneous data sources, combining intra-vehicle information (e.g., fuel level and consumption) with external data (including fuel prices, station locations, and geospatial context). While existing approaches often address either intra-vehicle monitoring or inter-vehicle communication in isolation, integrated frameworks that are both practically deployable and adaptable across diverse mobility scenarios remain underdeveloped. The Petrol-Filling Itinerary Estimation aNd Optimization (PIENO) framework was designed to address this gap by merging vehicular sensor data with external information through cloud services, creating a digital twin of the vehicle. PIENO automates the search for optimal refueling stations, thereby reducing fuel consumption and costs while supporting eco-driving practices. Its modular architecture ensures adaptability across vehicle types and markets by incorporating microcontroller-based sensing, OEM interfaces, national fuel price datasets, AI-based price forecasting, and optimization algorithms. A revised and extended version, dubbed RI-PIENO (Revised and Improved \\PIENO), further advances this approach by transforming the static proof-of-concept into a continuously adaptive decision engine. RI-PIENO introduces temporal modeling of drivers’ habitual trips, integrates machine learning predictions of user routines into Operations Research-based optimization, and represents refueling as a time-evolving directed acyclic graph informed by real-time sensor and geospatial data. Evaluation through realistic multi-driver, multi-week simulations demonstrates that RI-PIENO achieves significant improvements in cost efficiency, travel time, and emission reduction compared to both PIENO and baseline strategies. By bridging intra- and inter-vehicular data while dynamically adapting to user mobility patterns, RI-PIENO demonstrates the potential of Intelligent Transportation Systems to provide scalable, eco-aware, and user-centric mobility solutions within next-generation IoT-enabled vehicular networks.