ABSTRACT
The transition from Internal Combustion Engine (ICE) fleets to Electric Vehicles (EVs) has become a
corporate and national priority, yet most organizations struggle with identifying which vehicles should
be replaced first for maximum economic and environmental impact. This research proposes an intelligent
Hybrid Optimization–Simulation Framework (HOSF) that combines the analytical precision of
optimization with the predictive realism of simulation to produce a Replacement Priority Index (RPI) — a
quantitative metric that ranks fleet vehicles based on the potential benefit of electrification.
The proposed framework integrates multi-objective linear programming for cost, emission, and
utilization optimization with Monte Carlo and agent-based simulation models that account for real-world
uncertainties such as fluctuating fuel prices, maintenance variability, operational patterns, and policy
changes. This hybrid integration ensures that the model not only identifies the optimal replacement
candidates but also validates their performance under dynamic operating conditions.
The Replacement Priority Index (RPI) serves as the core decision metric, combining parameters like
vehicle age, mileage, emission intensity, maintenance cost, and route utilization frequency into a unified
score that drives prioritization. Vehicles with the highest RPI scores are recommended for immediate
replacement, yielding the most substantial economic and ecological returns. Over time, the model self-
updates using new cost and performance data, ensuring that recommendations remain relevant as
technology, policies, and operational demands evolve.
The proposed system dynamically recalibrates its recommendations under changing operational,
financial, and policy conditions, generating actionable insights that guide data-driven fleet transition
strategies. The framework was validated using open-source fleet and emission datasets from the U.S.
Department of Energy’s Alternative Fuels Data Center (AFDC), UK Department for Transport, India’s
Vahan Database, and the International Energy Agency’s (IEA) Global EV Data Explorer. Experimental
results demonstrate that the hybrid approach achieves a 25–30% improvement in cost-benefit efficiency
and up to a 35% gain in emission reduction effectiveness compared to traditional deterministic
optimization methods.
The research outcomes aim to contribute a data-driven framework for intelligent fleet electrification
planning aligned with India’s EV2030 and global decarbonization objectives. By bridging analytical
optimization and simulation-based prediction, the proposed HOSF establishes a scalable decision-
support system capable of transforming ICE-to-EV replacement from a reactive task into a proactive,
intelligence-led strategy for sustainable mobility management.
Keywords: Fleet Electrification, Optimization, Linear Programming, Machine Learning, Total Cost of
Ownership, Emission Reduction, Predictive Modeling, Sustainability.