A capacity scaling heuristic for the multicommodity capacitated network design problem

Katayama, Naoto; Chen, M.; Kubo, Mikio

doi:10.1016/j.cam.2008.10.055

Cited by 39 publications

(14 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, SACG2 produces results that are better by 0.04%. We also note that we are unable to consider the result of SACG2 for instance 30,520,400FT as this value is lower than the lower bound 150009 reported by Katayama et al (2009), and any comparison for this instance would therefore be misleading.…”

Section: Comparative Analysismentioning

confidence: 54%

“…In this section, we report comparative computational results of the proposed algorithm with the Cycle-based Tabu Search (CTS) of Ghamlouche et al (2003), Path Relinking (PR) by Ghamlouche et al (2004), Multilevel Cooperative Algorithm (MCA) by Crainic et al (2006), Capacity Scaling Heuristic (CSH) by Katayama et al (2009), IP Search (IPS) by Hewitt et al (2010), the two algorithms based on Simulated Annealing and Column Generation (SACG1 and SACG2) described by Yaghini et al (2013) the results for which are reported with time limits 600 and a 18000 seconds, respectively, and Local Branching (LocalB) by Rodríguez-Martín and Salazar-González (2010). The algorithm described by Alvarez et al (2005) could not be included in the comparisons as the authors do not report any results with the instances tested here; instead they use their own benchmark instances.…”

Section: Comparative Analysismentioning

confidence: 99%

“…The focus of their algorithm is on the specification of the problem instance solved at each level and the definition of the cooperation operators. Katayama et al (2009) propose a column and row generation heuristic for solving the MCNDP. The authors relax the arcs' capacity constraints, while a column and row generation technique is developed to solve the relaxed problem.…”

Section: Literaturementioning

confidence: 99%

See 2 more Smart Citations

A cycle-based evolutionary algorithm for the fixed-charge capacitated multi-commodity network design problem

Paraskevopoulos

Bektaş

Crainic

et al. 2016

European Journal of Operational Research

View full text Add to dashboard Cite

This paper presents an evolutionary algorithm for the fixed-charge multicommodity network design problem (MCNDP), which concerns routing multiple commodities from origins to destinations by designing a network through selecting arcs, with an objective of minimizing the fixed costs of the selected arcs plus the variable costs of the flows on each arc. The proposed algorithm evolves a pool of solutions using principles of scatter search, interlinked with an iterated local search as an improvement method. New cycle-based neighbourhood operators are presented which enable complete or partial re-routing of multiple commodities. An efficient perturbation strategy, inspired by ejection chains, is introduced to perform local compound cycle-based moves to explore different parts of the solution space. The algorithm also allows infeasible solutions violating arc capacities while forming the "ejection cycles", and subsequently restores feasibility by systematically applying correction moves. Computational experiments on benchmark MCNDP instances show that the proposed solution method consistently produces high-quality solutions in reasonable computational times.

show abstract

Section: Comparative Analysismentioning

confidence: 54%

Section: Comparative Analysismentioning

confidence: 99%

Section: Literaturementioning

confidence: 99%

See 1 more Smart Citation

A cycle-based evolutionary algorithm for the fixed-charge capacitated multi-commodity network design problem

Paraskevopoulos

Bektaş

Crainic

et al. 2016

European Journal of Operational Research

View full text Add to dashboard Cite

show abstract

“…We derive the instances used in our computational study from the C and C+ instances described in detail in Crainic et al (2001). These instances have been used to benchmark the performance of many algorithms for the capacitated fixed charge network design problem (Ghamlouch et al 2003, Ghamlouch et al 2004, Katayama et al 2009, Hewitt et al 2010, Yaghini et al 2012. The instances vary with respect to the number of nodes (20, 30), arcs (230,300,520,700), commodities (40,100,200, and 400), whether the variable costs, c ij , outweigh the fixed costs, f ij , and whether the arcs are loosely or tightly capacitated.…”

Section: Instancesmentioning

confidence: 99%

The Continuous-Time Service Network Design Problem

Boland

Hewitt

Marshall

et al. 2017

Operations Research

137

View full text Add to dashboard Cite

Consolidation carriers transport shipments that are small relative to trailer capacity. To be cost-effective, the carrier must consolidate shipments, which requires coordinating their paths in both space and time, i.e., the carrier must solve a Service Network Design problem. Most service network design models rely on discretization of time, i.e., instead of determining the exact time at which a dispatch should occur, the model determines a time interval during which a dispatch should occur. While the use of time discretization is widespread in service network design models, a fundamental question related to its use has never been answered: "Is it possible to produce an optimal continuous time solution without explicitly modeling each point in time?" We answer this question in the affirmative. We develop an iterative refinement algorithm using partially time-expanded networks that solves continuous time service network design problems. An extensive computational study demonstrates that the algorithm is not only of theoretical interest, but also performs well in practice. have long been supported by solving the so-called Service Network Design problem (Crainic 2000, Wieberneit 2008), which decides the paths for the shipments and the services (or resources) necessary to execute them. Service network design decisions for a consolidation carrier have both a geographic and temporal component, e.g., "dispatch a truck from Chicago, IL to Atlanta, GA at 9.05 pm." A common technique for modeling the temporal component is discretization; instead of deciding the exact time at which a dispatch should occur (e.g., 7.38 pm), the model decides a time interval during which the dispatch should occur (e.g., between 6pm and 8 pm).When discretizing time, service network design problems can be formulated on a time-expanded network Fulkerson 1958, 1962), in which a node encodes both a location and a time interval, and solutions prescribe dispatch time intervals for resources (trucks, drivers, etc.) and shipments. Service network design models calculate the costs for a set of dispatch decisions by estimating consolidation opportunities, i.e., by recognizing that prescribed dispatch time intervals for shipments allow travel together using the same resource. For example, shipments that should dispatch from the same origin node to the same destination node in the same dispatch time interval (say from Louisville, KY to Jackson, MI between 6 and 11 pm) are candidates for consolidation.Clearly, the granularity of the time discretization has an impact on the candidate consolidation opportunities identified. At the same time, the granularity of the time discretization also impacts the computational tractability. With an hourly discetization of a week-long planning horizon, 168 timed copies of a node representing a location will be created. With a 5-minute discretization of a week-long planning horizon, 2,016 timed copies of a node representing a location will be created.The latter discretization will likely yield a service network design prob...

show abstract

“…Construction cost of the network is traded-off against delay, or against the total sum of travel distances of the commodities. Frequently, there are additional requirements or constraints; for example, the problem may include limited capacity arcs, stochastic demand, survivability requirements, and so on [18][19][20][21]. The NDP has also been proven to be NP-complete even for simple cases [22].…”

Section: Introductionmentioning

confidence: 97%