AI-Based Predictive Modeling and NSGA-II Optimization for Eco-Driving Route Planning in Electric Vehicles

Abro, Muhammad Arsalan Jalees; Larik, Abdul Sattar; Alsulami, Mohammad; Halepoto, Irfan Ahmed; Alyami, Sultan; Alqhtani, Samar M.; Alqazzaz, Ali; Shaikh, Asadullah · 2025 · OpenAlex-citations

DOI: 10.1109/access.2025.3613479

archive: archived pipeline: cataloged verified

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

Summary

This study addresses the challenge of limited driving range and energy inefficiency in Electric Vehicles (EVs), particularly in urban environments where frequent acceleration and deceleration degrade battery performance. Motivated by the need for sustainable energy solutions and the potential of eco-driving strategies to reduce energy consumption by up to 30.4%, the authors propose a multi-objective framework that integrates AI-based predictive modeling with route optimization. The research aims to balance energy efficiency and travel time by incorporating real-time traffic data, historical records, and vehicle-specific dynamics into a comprehensive mathematical model. The methodology employs a four-phase approach: developing an energy consumption model, collecting traffic data, applying machine learning for pattern recognition, and executing multi-objective optimization using the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The energy consumption model accounts for driving style (normal, moderate, fast), traffic intensity, tire rolling resistance, and auxiliary load usage. Data were sourced from the TomTom API for real-time traffic and weather conditions, NYC OpenData for historical traffic records, and specific energy consumption profiles from a Volkswagen e-Golf. The dataset was preprocessed using regression and classification machine learning techniques to refine feature extraction. The NSGA-II algorithm was then applied to generate Pareto-optimal route options, allowing for trade-offs between minimizing energy consumption and reducing travel time. Numerical simulations were conducted to validate the model’s effectiveness in predicting energy usage and optimizing routes. The study generated 100 predictive route options for a trip between Times Square and the Holland Tunnel in New York City. Among these, Route 2, spanning 5.17 km, was identified as the most efficient option, requiring only 1.03 kWh of total energy ($E_{Total}$). The results demonstrate that the integrated framework successfully balances energy use and travel time through Pareto-optimal decision-making. The model accurately forecasts energy consumption by incorporating traffic flow, road topology, and vehicle dynamics, confirming the efficacy of combining eco-driving strategies with intelligent route planning. The significance of this work lies in its contribution to extending EV range and enhancing driver-assistance systems through intelligent eco-routing. By bridging vehicle-level energy modeling with real-time traffic-aware route planning, the study provides a holistic approach that addresses gaps in existing research, which often focuses solely on behavioral analysis or isolated AI optimizations. The findings support the development of autonomous vehicle systems and sustainable mobility solutions by offering a validated method for reducing energy inefficiencies and greenhouse gas emissions. This framework enables drivers to make informed choices between energy-efficient and time-efficient paths, thereby promoting the adoption of eco-driving behaviors and improving overall EV performance.

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.

• eco driving