The service network design problem with fleet and emissions management

Abstract

We propose the Service Network Design Problem with Fleet and Emissions Management (SND-FEM). This problem considers fleet management decisions regarding how many vehicles of each type (diesel, electric, hydrogen) a carrier should acquire as well as in what regions they should operate. The impact of these fleet-level decisions on customer service is captured by explicitly modeling the routing of shipments and vehicles while recognizing the consumption of limited onboard resources (diesel fuel, electrical energy, hydrogen). Thus, the problem also prescribes resource replenishment decisions, with an explicit recognition that the emissions and costs associated with resource replenishment can vary by region. The environmental impact of problem decisions is captured with a detailed accounting of Well-to-Wheel (WtW) emissions based on the framework proposed by the Global Logistics Emissions Council, a methodology that has become the industry standard. We propose a Mixed Integer Programming (MIP) formulation of the problem and consider two objectives: (i) transportation cost, and (ii) WtW emissions. With a case study based on carefully collected data from government and industry sources, we elaborate the potential of including novel power-train technologies in heavy-duty truck fleets to reduce both cost and emissions-based fleet performance measures.