| This research studies the two-dimensional loading constrained vehicle routing problem with cross-docking under uncertain environment.The main contributions of our research can be summarized as follows:(1)This research studies the capacitated vehicle routing problem with two-dimensional loading constraints(2L-CVRP).The mathematical models of four variants of the 2L-CVRP with respect to first-in-last-out and rotation of items constraints are devoloped.According the characteristics of the problem,an improved adaptive tabu search algorithm incorporating the multi-oder bottom-left-skyline packing heuristic is proposed.Finally,the experimental results between our approach and the existing state-of-art methods demonstrate the correctiveness and effectiveness.(2)This research investigates a two-dimensional loading constrained vehicle routing problem with cross-docking(2L-VRPCD).A mixed integer linear programming(MILP)model of the 2L-VRPCD is proposed,aiming to minimize the total transportation cost.To efficiently solve large-scale instances,a hybrid optimization method incorporating an adaptive large neighborhood search(ALNS)algorithm and a multiorder bottom-fill-skyline(MOBFS)packing heuristic is proposed,where multiple problem-specific destroy and repair operators are introduced and implemented.The MOBFS is proposed to check the feasibility of each route.Afterwards,extensive computational experiments have proven the performance and versatility of the proposed algorithm for solving the problem.The comparison results indicate that our approach outperforms the MILP-based method with respect to computational efficiency and solution quality for solving 2L-VRPCD,especially large-scale instances.(3)In this research,a two-dimensional loading constrained vehicle routing problem with cross-docking and stochastic travel time(2LVRPCDSTT)is proposed and formulated.A stochastic programming model with recourse of 2L-VRPCDSTT is established based on Monte Carlo simulation and scenario analysis,to minimize the sum of the transporation cost,fixed cost of vehicles and the expected time window penalty cost.To achieve this,an enhanced adaptive tabu search(EATS)algorithm incorporating the tabu-based multi-order best-fit(TSMOBF)packing heuristic is proposed.A large number of computational experiments are conducted.Experimental results demonstrate the applicability of the proposed model and approach,which is capable of efficiently solving the real-life 2L-VRPCD with uncertain information.There are altogether 31 figures,30 tables and 124 items of bibliography in the thesis. |
