Since there is no literature available on the MT-VRPB, we provide
brief reviews for the two related routing problems namely the
MT-VRP and the VRPB.
2.1. MT-VRP
The multi-trip vehicle routing was first studied in Salhi [29] where
multiple trips were conducted in the context of vehicle fleet mix.
Limited to double trips, a matching algorithm is proposed to assign
routes to vehicles within a refinement process. Taillard et al. [35]
introduced the MT-VRP model based on the classical VRP and proposed
a three-phase heuristic algorithm. In the first phase, tabu search
is used to generate a population of routes satisfying the capacity
constraint; a set of different VRP solutions is then obtained in phase
two. Routes are then assigned to the vehicles by solving the binpacking
problem (BPP) in the last phase. Moreover, a set of classical
MT-VRP instances are generated in their study which are widely used
in the literature as benchmarks. Brandao and Mercer [4] studied a real
world application of MT-VRP with time windows and heterogeneous
fleet, and used a tabu search algorithm to solve the problem. The
methodology developed in this study is adapted in Brandao and
Mercer [5] where classical MT-VRP instances were solved and compared.
Petch and Salhi [27] developed a multi-phase constructive
heuristic algorithm with an objective of minimising the overtime used
in multi-trips. The algorithm obtained an MT-VRP solution by solving
BPP which is improved further using the 2-Opt and 3-Opt exchange
heuristic procedures. Salhi and Petch [31] revisited their previous