A bicriteria mathematical programming model for strategic road freight fleet renewal under environmental and economic objectives

Road transport is a major contributor to freight-related greenhouse gas emissions, and its relevance for efforts to decarbonize the transport sector as a whole is increasing. It is by now agreed that decarbonization of road freight will ultimately hinge on a transition away from oil-based fuels, mainly diesel and gasoline, which continue to dominate the sector. However, alternatives to oil span a multitude of technologies, ranging from electricity to biofuels, and entail varying levels and forms of investment. To support the development of informed decarbonization strategies in this context, we describe a bicriteria mathematical programming model for optimizing vehicle replacement decisions in a fleet of trucks to be operated over several years. Given a specification of the initial fleet, the model generates a set of renewal strategies that achieve different tradeoffs between aggregate well-to-wheel emissions and total investment and operation costs. The model accounts for multiple technologies, truck types, payloads and operational profiles, and explicitly considers the bearing of these characteristics on costs and emissions. The model also incorporates the installation costs of charging and alternative fueling infrastructure, of maintenance and insurance, as well as proceeds from salvage actions. Leveraging cost and emission data from French logistics companies, we investigate optimal renewal strategies across fleets representative of real-world operations. Our experiments reveal a correlation between emissions reductions and overall costs, with decarbonization pathways consistently incurring higher costs than a diesel-only strategy. Extensive computational experience comparing the effect of different energy price trajectories on the structure of optimal solutions is also reported. These results provide insights into the economic conditions under which each technology becomes competitive with respect to the others. They also highlight how expectations regarding the future of energy prices influence technology adoption and investment timing.