Showing papers by "David Pisinger published in 2011"

Fleet deployment, network design and hub location of liner shipping companies

Abstract: A mixed integer linear programming formulation is proposed for the simultaneous design of network and fleet deployment of a deep-sea liner service provider. The underlying network design problem is based on a 4-index (5-index by considering capacity type) formulation of the hub location problem which are known for their tightness. The demand is elastic in the sense that the service provider can accept any fraction of the origin–destination demand. We then propose a primal decomposition method to solve instances of the problem to optimality. Numerical results confirm superiority of our approach in comparison with a general-purpose mixed integer programming solver.

Liner Shipping Cargo Allocation with Repositioning of Empty Containers

Abstract: This paper is concerned with the cargo allocation problem considering empty repositioning of containers for a liner shipping company. The aim is to maximize the profit of transported cargo in a network, subject to the cost and availability of empty containers. The formulation is a multi-commodity flow problem with additional inter-balancing constraints to control repositioning of empty containers. In a study of the cost efficiency of the global container-shipping network, Song et al. (2005) estimate that empty repositioning cost constitutes 27% of the total world fleet running cost. An arc-flow formulation is decomposed using the Dantzig–Wolfe principle to a path-flow formulation. A linear relaxation is solved with a delayed column generation algorithm. A feasible integer solution is found by rounding the fractional solution and adjusting flow balance constraints with leased containers. Computational results are reported for seven instances based on real-life shipping networks. Solving the relaxe...

Train shunting at a workshop area

Abstract: We consider the problem of planning the shunting of train units at a railway workshop area. Before and after the maintenance check, a train unit is parked at a depository track. The problem is to schedule the trains to workshops and depot tracks in order to complete the repairs as soon as possible, while avoiding train blockings at the tracks. We give a formal definition of the problem and present three heuristic approaches based on, respectively, Guided Local Search (GLS), Guided Fast Local Search (GFLS) and Simulated Annealing (SA). Computational experiments are reported for realistic instances. It turns out, that both GLS and SA find within a few minutes solutions that are a few percent from the best MIP solution found.

A Path Based Model for a Green Liner Shipping Network Design Problem

Abstract: —Liner shipping networks are the backbone ofinternational trade providing low transportation cost, whichis a major driver of globalization. These networks are underconstant pressure to deliver capacity, cost effectiveness and envi-ronmentally conscious transport solutions. This article proposesa new path based MIP model for the Liner shipping NetworkDesign Problem minimizing the cost of vessels and their fuelconsumption facilitating a green network. The proposed modelreduces problem size using a novel aggregation of demands.A decomposition method enabling delayed column generationis presented. The subproblems have similar structure to Ve-hicle Routing Problems, which can be solved using dynamicprogramming.Index Terms—liner shipping, network design, mathematicalprogramming, column generation, green logistics I. I NTRODUCTION G LOBAL liner shipping companies provide port to porttransport of containers, on a network which representsa billion dollar investment in assets and operational costs.The liner shipping network can be viewed as a transporta-tion system for general cargo not unlike an urban mass transitsystem for commuters, where each route (service) providestransportation links between ports and the ports allow fortranshipment in between routes (services). The liner shippingindustry is distinct from other maritime transportation modesprimarily due to a fixed public schedule with weekly fre-quency of port calls as an industry standard (Stopford 1997).The network consists of a set of services. A service connectsa sequence of ports in a cycle at a given frequency, usuallyweekly. In Figure 1 a service connecting Montreal-Halifaxand Europe is illustrated. The weekly frequency means thatseveral vessels are committed to the service as illustrated byFigure 1, where four vessels cover a round trip of 28 daysplaced with one week in between vessels. This roundtrip forthe vessel is referred to as a rotation. Note that the Montrealservice carries cargo to the Mediterranean and Asia. Thisillustrates that transhipments to other connecting servicesis at the core of liner shipping. Therefore, the design of aservice is complex, as the set of rotations and their interactionthrough transhipment is a transportation system extending thesupply chains of a multiplum of businesses. Figure 2 illus-trates two services interacting in transporting goods betweenMontreal-Halifax and the Mediterranean, while individually

Branch-and-cut and Branch-and-Cut-and-Price Algorithms for Solving Vehicle Routing Problems

Abstract: This paper presents a branch-and-cut-and-price algorithm for the vehicle routing problem with time windows. The standard Dantzig-Wolfe decomposition of the arc flow formulation leads to a set partitioning problem as the master problem and an elementary shortest path problem with resource constraints as the pricing problem. We introduce the subset-row inequalities, which are Chvatal-Gomory rank-1 cuts based on a subset of the constraints in the master problem. Applying a subset-row inequality in the master problem increases the complexity of the label-setting algorithm used to solve the pricing problem since an additional resource is added for each inequality. We propose a modified dominance criterion that makes it possible to dominate more labels by exploiting the step-like structure of the objective function of the pricing problem. Computational experiments have been performed on the Solomon benchmarks where we were able to close several instances. The results show that applying subset-row inequalities in the master problem significantly improves the lower bound, and in many cases makes it possible to prove optimality in the root node.

Separation and extension of cover inequalities for second-order conic knapsack constraints with GUBs

Abstract: We consider the second-order conic equivalent of the classic knapsack polytope where the variables are subject to generalized upper bound constraints. We describe and compare a number of separation and extension algorithms which make use of the extra structure implied by the generalized upper bound constraints in order to strengthen the second-order conic equivalent of the classic cover cuts. We show that determining whether a cover can be extended with a variable is NP-hard. Computational experiments are performed comparing the proposed separation and extension algorithms. These experiments show that applying these extended cover cuts can greatly improve solution time of second-order cone programs.

Optimal wafer cutting in shuttle layout problems

Abstract: A major cost in semiconductor manufacturing is the generation of photo masks which are used to produce the dies. When producing smaller series of chips it can be advantageous to build a shuttle mask (or multi-project wafer) to share the startup costs by placing different dies on the same mask. The shuttle layout problem is frequently solved in two phases: first, a floorplan of the shuttle is generated. Then, a cutting plan is found which minimizes the overall number of wafers needed to satisfy the demand of each die type. Since some die types require special production technologies, only compatible dies can be cut from a given wafer, and each cutting plan must respect various constraints on where the cuts may be placed. We present an exact algorithm for solving the minimum cutting plan problem, given a floorplan of the dies. The algorithm is based on delayed column generation, where the pricing problem becomes a maximum vertex-weighted clique problem in which each clique consists of cutting compatible dies. The resulting branch-and-price algorithm is able to solve realistic cutting problems to optimality in a couple of seconds.

Pricing and Capacity Planning Problems in Energy Transmission Networks

Abstract: (29/12/2018) Pricing and Capacity Planning Problems in Energy Transmission Networks Efficient use of energy is an increasingly important topic. Environmental and climate concerns as well as concerns for security of supply has made renewable energy sources a viable alternative to traditional energy sources. However, the intermittent nature of for instance wind and solar energy necessitates a radical change in the way we plan and operate energy systems. Another paradigm change which began in the 1990’s for electricity systems is that of deregulation. This has led to a variety of different market structures implemented across the world. In this thesis we discuss capacity planning and transmission pricing problems in energy transmission networks. Although the modelling framework applies to energy networks in general, most of the applications discussed concern the transmission of electricity. A number of the problems presented involves transmission switching, which allows the operator of an electricity transmission network to switch lines in and out in an operational context in order to optimise the network flow. We show that transmission switching in systems with large-scale wind power may alleviate network congestions and reduce curtailment of wind power leading to higher utilisation of installed wind power capacity. We present formulations of — and efficient solution methods for— the transmission line capacity expansion problem and the unit commitment problem with transmission switching. We also show that transmission switching may radically change the optimal line capacity expansion strategy. In the Nordic electricity system a market with zonal prices is adopted. We consider the problem of designing zones in an optimal way explicitly considering uncertainty. Finally, we formulate the integrated problem of pipeline capacity expansion planning and transmission pricing in natural gas transmission networks.

Robust long-term production planning

Abstract: We present an application of an Adaptive Large Neighborhood Search (ALNS) algorithm to the Resource-Constrained Project Scheduling Problem (RCPSP). The ALNS framework was first proposed by Pisinger and Røpke (2007) and can be described as a large neighborhood search algorithm with an adaptive layer, where a set of destroy/repair neighborhoods compete to modify the current solution in each iteration of the algorithm. Experiments are performed on the well-known J30, J60 and J120 benchmark instances, which show that the proposed algorithm is competitive and confirms the strength of the ALNS framework previously reported for different variants of the Vehicle Routing Problem.