Stochastic fleet mix optimization: Evaluating electromobility in urban logistics. (February 2022)

• Record Type: Journal Article
• Authors: Malladi, Satya S.; Christensen, Jonas M.; Ramírez, David; Larsen, Allan; Pacino, Dario
• Is Part Of: Transportation research. Volume 158(2022)
• Publication Date: 2022-02
• Subjects: Fleet size and mix -- Vehicle routing problem -- Stochastic fleet sizing -- Sample average approximation -- Adaptive large neighborhood search -- Case studies; Logistics -- Periodicals; Transportation -- Periodicals; 388.011
• DOI: 10.1016/j.tre.2021.102554
• Languages: English
• ISSNs: 1366-5545

Abstract

Abstract: In this paper, we study the problem of optimizing the size and mix of a mixed fleet of electric and conventional vehicles owned by firms providing urban freight logistics services. Uncertain customer requests are considered at the strategic planning stage. These requests are revealed before operations commence in each operational period. At the operational level, a new model for vehicle power consumption is suggested. In addition to mechanical power consumption, this model accounts for cabin climate control power, which is dependent on ambient temperature, and auxiliary power, which accounts for energy drawn by external devices. We formulate the problem of stochastic fleet size and mix optimization as a two-stage stochastic program and propose a sample average approximation based heuristic method to solve it. An adaptive large neighborhood search algorithm is used for each operational period to determine the operational decisions and associated costs. The applicability of the approach is demonstrated through two case studies within urban logistics services. Highlights: We propose SFSMP, a novel stochastic optimization problem for strategic fleet size and mix decision-making in urban freight logistics. Improved energy consumption model including the energy required for auxiliary usage. We propose a generalization of the fleet size and mix VRP for the evaluation of the operational problem appearing in the SFSMP. We propose a heuristic variation of the sample average approximation method to solve the SFSMP. Finally, we demonstrate the applicability of the approach on two different case studies of service operations in urban freight logistics.