Adaptive Large Neighborhood Search for Vehicle Routing Problem with Cross-Docking

Gunawan, Aldy; Widjaja, Audrey Tedja; Vansteenwegen, Pieter; Yu, Vincent F.

doi:10.1109/cec48606.2020.9185514

Cited by 10 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Operational problems of cross-docking are NP-complex tasks. Therefore, trying best to solve the problem, one should note that the exact algorithms can work effectively only for small problems [16].…”

Section: Introductionmentioning

confidence: 99%

Cross-docking cargo delivery routing for guaranteed minimum period

Oliskevych¹,

Danchuk²,

Mastykash³

2022

View full text Add to dashboard Cite

The article is devoted to the problem of effective use of cross-docking as a technology of cargo delivery with increased time requirements, which allows to resolve the contradictions of guaranteed delivery time ensuring and the efficiency of the existing fleet of trucks. The process of delivery organization is considered as the ordering on the transport network of many discrete freight flows in the form of their phases. If qualitative and / or quantitative changes do not occur from phase to phase with the flow, then the tact of such flow is constant. However, cross-docking flows change the size of the band of moving goods. Cargo can be moved as intended by any group size, which, however, is limited by the maximum and minimum values. A two-stage algorithm for solving the problem has been developed. The transport network is represented as a graph. The content of the route search problem is optimization, as it consists of multiple selections from the initial graph of arcs in the presence of restrictions on input and output flows. One needs to replace every each edge of the graph with an arc of the forward or reverse direction, or remove this edge. The criterion for the optimal solution of the problem, which is applied, is the minimum guaranteed duration of delivery of goods throughout the set of specified freight flows. At the first stage of the algorithm, the search for the shortest paths in the graph is performed, along which every given cargo flow can pass. The first stage of optimization is a linear problem of integer programming, the dimension of which is not too large. The initial data of the second stage is freight flows matrix, which is obtained as a result of optimization in the first stage. The content of the second stage of the algorithm is the solution of the equation of the balance of discrete goods flows. The balance equation means that all flows entering each peak including the sources of cargo flows of this peak have an average intensity equal to the intensity of the outgoing cargo flows from each source peak, including runoff. Due to the studied dependencies between the individual phases of the delivery process on the example of a cargo carrier on the transport network of Ukraine, the formulated restrictions and boundary conditions, the possibility of guaranteed accurate solution of a complex problem is obtained. At the same time, the shortest routes were found, reloading points were identified as well as time parameters of operation and the degree of loading of cars. According to the results of the research, a threefold increase in the productivity of the fleet of road trains with a reduction in the guaranteed delivery time by 30%.

show abstract

Section: Introductionmentioning

confidence: 99%

Cross-docking cargo delivery routing for guaranteed minimum period

Oliskevych¹,

Danchuk²,

Mastykash³

2022

View full text Add to dashboard Cite

show abstract

“…This study uses mixed-integer linear programming to model the proposed VRP with Cross-Docking. The problem in this study is based on the work of Lee et al (2006) and provides an extended and alternative model of the work done by Gunawan et al (2020), by allowing the use split pickups, split deliveries, available routes, and heterogeneous vehicles. Furthermore, our proposed model allows vehicles to start the delivery process without waiting for all vehicles to reach the cross-dock, provided that the inventory in the cross-dock can fulfill the demand in the vehicle's route.…”

Section: Methodsmentioning

confidence: 99%

“…The proposed algorithm gives competitive results while being significantly faster than the GAMS solver. Gunawan et al (2020) extends the work of Lee et al (2006) by modeling a VRPCD with multiple products and a set of homogenous vehicles, being called a VRP-MPCD. A mixed integer linear programming model was proposed and solved using the CPLEX solver.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The proposed model also aims to add multiple period deliveries, which means that vehicles don't need start the delivery at the same time, provided that they have enough inventory. Modified parameters to generate data proposed by Lee et al (2006) are used by multiple other studies as a benchmark dataset (Birim 2016;Gunawan et al 2020;Liao et al 2010;Vahdani et al 2012). Thus, this study also uses generated data with parameters based on Lee et al (2006).…”

Section: Objectivesmentioning

confidence: 99%

“…Several additional constraints further categorize the VRP, such as Capacitated VRP, Heterogeneous VRP, Muti-Depot VRP, Split Delivery VRP, and VRP with Cross-Docking (Archetti & Speranza 2008;Han & Wang, 2018;Mavi et al 2020). Previous studies on VRPCD linear models have provided focused on different factors such as split deliveries and multiple products (Gunawan et al 2020;Hasani-Goodarzi & Tavakkoli-Moghaddam 2012). Other studies on this topic use non-linear models to represent the problem and heuristics to generate solutions (Birim 2016;Lee et al 2006).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-Period Vehicle Routing Problem with Cross-Docking using Split Pickups and Deliveries

Proceedings of the International Conference on Industrial Engineering and Operations Management

View full text Add to dashboard Cite

Cross-docking provides benefits such as facilitating fast and direct unloading, consolidation, and reloading of goods from inbound to outbound trucks. The vehicle routing problem with cross-docking (VRPCD) is an extension of the vehicle routing problem with an additional constraint in which the depot has a role as cross-dock. The current paper provides a mixed-integer linear programming model for the VRPCD that considers multiple products, multiple vehicle types, and split pickups and delivery. Based on the literature, we generated two sets of instances, one with a homogenous vehicle and one with heterogeneous vehicles. The instances are then solved using the Gurobi Solver. Results show that instances with heterogeneous vehicles provide better results but require an increase in computation time (median of 37.3% for a 10-node instance with five types of vehicles). Split pickups and deliveries can produce lower objective values and generate solutions previously infeasible. The computational results also show that smaller instances of 10 nodes can be solved in a reasonable amount of time. However, larger instances of 30 nodes are not optimally solved given a limited computation time of 1 hour.

show abstract

Vehicle Routing Problem with Reverse Cross-Docking: An Adaptive Large Neighborhood Search Algorithm

Gunawan

Widjaja

Vansteenwegen

et al. 2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Cross-docking is a logistics strategy that aims at less transportation costs and fast customer deliveries. Incorporating an efficient vehicle routing could increase the benefits of the cross-docking. In this paper, the vehicle routing problem with reverse cross-docking (VRP-RCD) is studied. Reverse logistics has attracted more attention due to its ability to gain more profit and maintain the competitiveness of a company. VRP-RCD includes a four-level supply chain network: suppliers, crossdock, customers, and outlets, with the objective of minimizing vehicle operational and transportation costs. A two-phase heuristic employs an adaptive large neighborhood search (ALNS) with various destroy and repair operators is proposed to solve benchmark instances. The simulated annealing framework is embedded to discover a vast search space during the search process. Experimental results show that our proposed ALNS obtains optimal solutions for 24 out of 30 problems of the first set of benchmark instances while getting better results for all instances in the second set of benchmark instances compared to optimization software.

show abstract

Adaptive Large Neighborhood Search for Vehicle Routing Problem with Cross-Docking

Cited by 10 publications

References 12 publications

Cross-docking cargo delivery routing for guaranteed minimum period

Cross-docking cargo delivery routing for guaranteed minimum period

Multi-Period Vehicle Routing Problem with Cross-Docking using Split Pickups and Deliveries

Vehicle Routing Problem with Reverse Cross-Docking: An Adaptive Large Neighborhood Search Algorithm

Contact Info

Product

Resources

About