An Analysis of Synergy Effects in a Flow Refueling Location Problem with Capacity Constraints

In this research, we investigate several instances of the Flow Refueling Location Problem (FRLP). First, we examine the formulation of De Vries & Duijzer (2017) and their extensions of it (the Expected FRLP and Chance Constrained FRLP) that model the driving range as a stochastic parameter. Next, we treat a new FRLP formulation of Boujelben & Gicquel (2019) and use it as a basis for our newly introduced Unit-Capacitated FRLP, in which capacity is represented as interchangeable modular units. We show that the models of De Vries and Duijzer outperform the deterministic FRLP in terms of expected coverage and chance constrained coverage, and that they inhibit synergy effects. Moreover, it is shown that the flexibility of a modular unit structure for capacity yields similar efficiency gains, especially when the driving range is relatively low compared to the size of the network and units are split up into smaller pieces multiple times. Finally, using the capacitated model as opposed to the uncapacitated deterministic model results in substantial coverage gains.