Two-stage fleet size and mix electric vehicle routing problem with energy estimation

Source abstract

Decarbonization of road transportation in Europe is slow, while interest in emerging technologies and operating strategies with environmental and operational benefits is increasing. To assess their impacts, vehicle routing models support strategic fleet assessment, yet many rely on simplified energy models and neglect key vehicle- and system-level factors, limiting evaluation of technologies and operating strategies. We develop a two-stage optimization framework that determines fleet size, mix, and routes for simultaneous pickup and delivery, estimating realistic energy use on road networks. The energy model is evaluated against electric-truck data, and the framework quantifies trade-offs among driverless multipurpose, human-driven combustion, and battery-electric fleets in combined heavy-goods freight-and-waste city logistics. Although the energy model overestimates absolute energy use, it reproduces the main subroute ranking and captures fleet- and network-specific effects. The framework supports energy-informed strategic, tactical, and operational decisions; key limitations are scalability and that results reflect automation rather than full multipurpose capability.