Proximity Benders: a decomposition heuristic for stochastic programs

In this paper we present a heuristic approach to two-stage mixed-integer linear stochastic programming models with continuous second stage variables. A common solution approach for these models is Benders decomposition, in which a sequence of (possibly infeasible) solutions is generated, until an optimal solution is eventually found and the method terminates. As convergence may require a large amount of computing time for hard instances, the method may be unsatisfactory from a heuristic point of view. Proximity search is a recently-proposed heuristic paradigm in which the problem at hand is modified and iteratively solved with the aim of producing a sequence of improving feasible solutions. As such, proximity search and Benders decomposition naturally complement each other, in particular when the emphasis is on seeking high-quality, but not necessarily optimal, solutions. In this paper, we investigate the use of proximity search as a tactical tool to drive Benders decomposition, and computationally evaluate its performance as a heuristic on instances of different stochastic programming problems.     

A facility neighborhood search heuristic for capacitated facility location with single-source constraints and flexible demand

We consider a generalization of the well-known capacitated facility location problem with single source constraints in which customer demand contains a flexible dimension. This work focuses on providing fast and practically implementable optimization-based heuristic solution methods for very large scale problem instances. We offer a unique approach that utilizes a high-quality efficient heuristic within a neighborhood search to address the combined assignment and fixed-charge structure of the underlying optimization problem. We also study the potential benefits of combining our approach with a so-called very large-scale neighborhood search (VLSN) method. As our computational test results indicate, our work offers an attractive solution approach that can be tailored to successfully solve a broad class of problem instances for facility location and similar fixed-charge problems.     

Heuristics for Multi-Stage Interdiction of Stochastic Networks

We describe and compare heuristic solution methods for a multi-stage stochastic network interdiction problem. The problem is to maximize the probability of sufficient disruption of the flow of information or goods in a network whose characteristics are not certain. In this formulation, interdiction subject to a budget constraint is followed by operation of the network, which is then followed by a second interdiction subject to a second budget constraint. Computational results demonstrate and compare the effectiveness of heuristic algorithms. This problem is interesting in that computing an objective function value requires tremendous effort. We exhibit classes of instances in our computational experiments where local search based on a transformation neighborhood is dominated by a constructive neighborhood.     

A large neighbourhood search heuristic for the aircraft and passenger recovery problem

This paper introduces a large neighbourhood search heuristic for an airline recovery problem combining fleet assignment, aircraft routing and passenger assignment. Given an initial schedule, a list of disruptions, and a recovery period, the problem consists in constructing aircraft routes and passenger itineraries for the recovery period that allow the resumption of regular operations and minimize operating costs and impacts on passengers. The heuristic alternates between construction, repair and improvement phases, which iteratively destroy and repair parts of the solution. The aim of the first two phases is to produce an initial solution that satisfies a set of operational and functional constraints. The third phase then attempts to identify an improved solution by considering large schedule changes while retaining feasibility. The whole process is iterated by including some randomness in the construction phase so as to diversify the search. This work was initiated in the context of the 2009 ROADEF Challenge, a competition organized jointly by the French Operational Research and Decision Analysis Society and the Spanish firm Amadeus S.A.S., in which our team won the first prize.     

A vehicle routing problem with multi-trips and time windows for circular items

This article presents a vehicle routing problem with time windows, multiple trips, a limited number of vehicles and loading constraints for circular objects. This is a real problem experienced by a home delivery service company. A linear model is proposed to handle small problems and a two-step heuristic method to solve real size instances: the first step builds an initial solution through the modification of the Solomon I1 sequential insertion heuristic, and the second step improves the initial solution through the Tabu search algorithm proposed; in both steps, the problems related to circle packing with different sizes and bin packing are solved jointly with the use of heuristics. Finally, the computing results for two different sets of instances are presented.     

Modeling formulation and a new heuristic for the railroad blocking problem

The railroad blocking problem is an important issue at the tactical level of railroad freight transportation. This problem consists of determining paths between the origins and destinations of each shipment to minimize the operating and user costs while satisfying the railroad supply and demand restrictions. A mixed-integer program (MIP) is developed to find the optimal paths, and a new heuristic is developed to solve the proposed model. This heuristic decomposes the model into two sub-problems of manageable size and then provides feasible solutions. We discuss the performance of the proposed heuristic for a set of instances with up to 90 stations. A comparison with the CPLEX MIP solver shows that the heuristic gives the exact solution for 10 out of 15 instances. For the remaining instances, the heuristic obtained solutions within a tolerance of 0.03–0.84%. Furthermore, compared with the CPLEX MIP solver, the heuristic reduced the run time by an average of 85% for all 15 instances. Finally, we present the computational results of the heuristic applied to Iranian railroads.     

An Improved Petal Heuristic for the Vehicle Routeing Problem

Solutions produced by the first generation of heuristics for the vehicle routeing problem are often far from optimal. Recent adaptations of local search improvement heuristics, like tabu search, produce much better solutions but require increased computing time. However there are situations where good solutions must be obtained quickly. The algorithm proposed in this paper yields solutions almost as good as those produced by tabu search adaptations, but at only a small fraction of their computing time. This heuristic can be seen as an improved version of the original petal heuristic. On 14 benchmark test problems, the proposed heuristic yields solutions whose values lie on average within 2.38% of that of the best known solutions.     

Open vehicle routing problem with driver nodes and time deadlines

In this paper, we consider a variant of the open vehicle routing problem in which vehicles depart from the depot, visit a set of customers, and end their routes at special nodes called driver nodes. A driver node can be the home of the driver or a parking lot where the vehicle will stay overnight. The resulting problem is referred to as the open vehicle routing problem with driver nodes (OVRP-d). We consider three classes of OVRP-d: with no time constraints, with a maximum route duration, and with both a maximum route duration as well as time deadlines for visiting customers. For the solution of these problems, which are not addressed previously in the literature, we develop a new tabu search heuristic. Computational results on randomly generated instances indicate that the new heuristic exhibits a good performance both in terms of the solution quality and computation time.     

Lagrangian relaxation guided problem space search heuristics for generalized assignment problems

We develop and test a heuristic based on Lagrangian relaxation and problem space search to solve the generalized assignment problem (GAP). The heuristic combines the iterative search capability of subgradient optimization used to solve the Lagrangian relaxation of the GAP formulation and the perturbation scheme of problem space search to obtain high-quality solutions to the GAP. We test the heuristic using different upper bound generation routines developed within the overall mechanism. Using the existing problem data sets of various levels of difficulty and sizes, including the challenging largest instances, we observe that the heuristic with a specific version of the upper bound routine works well on most of the benchmark instances known and provides high-quality solutions quickly. An advantage of the approach is its generic nature, simplicity, and implementation flexibility.     

A fast heuristic for solving a large-scale static dial-a-ride problem under complex constraints

This paper presents a heuristic, which concentrates on solving a large-scale static dial-a-ride problem bearing complex constraints. In this heuristic, a properly organized local search strategy and a diversification strategy are used to improve initial solutions. Then the improved solutions can be refined further by an intensification strategy. The performance of this heuristic was evaluated by intensive computational tests on some randomly generated instances. Small gaps to the lower bounds from the column generation method were obtained in very short time for instances with no more than 200 requests. Although the result is not sensitive to the initial solution, the computational time can be greatly reduced if some effort is spent to construct a good initial solution. With this good initial solution, larger instances up to 2000 requests were solved in less than 10 hours on a popular personal computer.     

GRASP with path relinking for the orienteering problem

In this paper, we address an optimization problem resulting from the combination of the well-known travelling salesman and knapsack problems. In particular, we target the orienteering problem, originated in the context of sport, which consists of maximizing the total score associated with the vertices visited in a path within the available time. The problem, also known as the selective travelling salesman problem, is NP-hard and can be formulated as an integer linear program. Since the 1980s, several solution methods for this problem have been developed and applied to a variety of fields, particularly in routing and tourism. We propose a heuristic method—based on the Greedy Randomized Adaptive Search Procedure (GRASP) and the Path Relinking methodologies—for finding approximate solutions to this optimization problem. We explore different constructive methods and combine two neighbourhoods in the local search of GRASP. Our experimentation with 196 previously reported instances shows that the proposed procedure obtains high-quality solutions employing short computing times.     

A deterministic tabu search algorithm for the fleet size and mix vehicle routing problem

The fleet size and mix vehicle routing problem consists of defining the type, the number of vehicles of each type, as well as the order in which to serve the customers with each vehicle when a company has to distribute goods to a set of customers geographically spread, with the objective of minimizing the total costs. In this paper, a heuristic algorithm based on tabu search is proposed and tested on several benchmark instances. The computational results show that the proposed algorithm produces high quality results within a reasonable computing time. Some new best solutions are reported for a set of test problems used in the literature.     

Efficient solution of large scale,single-source,capacitated plant location problems

The single-source, capacitated plant location problem is considered. This problem differs from the capacitated plant location problem by the additional requirement that each customer must be supplied with all its demand from a single plant. An efficient heuristic solution, capable of solving large problem instances, is presented. The heuristic combines Lagrangian relaxation with restricted neighbourhood search. Computational experiments on two sets of problem instances are presented.     

The prize-collecting generalized minimum spanning tree problem

We introduce the prize-collecting generalized minimum spanning tree problem. In this problem a network of node clusters needs to be connected via a tree architecture using exactly one node per cluster. Nodes in each cluster compete by offering a payment for selection. This problem is NP-hard, and we describe several heuristic strategies, including local search and a genetic algorithm. Further, we present a simple and computationally efficient branch-and-cut algorithm. Our computational study indicates that our branch-and-cut algorithm finds optimal solutions for networks with up to 200 nodes within two hours of CPU time, while the heuristic search procedures rapidly find near-optimal solutions for all of the test instances.     

Minimizing the total completion time in a two-machine flowshop with sequence-independent setup times

We consider the problem of minimizing the sum of completion times in a two-machine permutation flowshop subject to setup times. We propose a new priority rule, several constructive heuristics, local search procedures, as well as an effective multiple crossover genetic algorithm. Computational experiments carried out on a large set of randomly generated instances provide evidence that a constructive heuristic based on newly derived priority rule dominates all the proposed constructive heuristics. More specifically, we show that one of our proposed constructive heuristics outperforms the best constructive heuristic in the literature in terms of both error and computational time. Furthermore, we show that one of our proposed local search-based heuristics outperforms the best local search heuristic in the literature in terms of again both error and computational time. We also show that, in terms of quality-to-CPU time ratio, the multiple crossover genetic algorithm performs consistently well.     

A covering tour approach to the location of satellite distribution centers to supply humanitarian aid

This article concerns the location of satellite distribution centers (SDCs) to supply humanitarian aid to the affected people throughout a disaster area. In such situations, it is not possible for the relief teams to visit every single home. Instead, the people are required to go to a satellite distribution center in order to obtain survival goods, provided that these centers are not too far from their homes. The SDCs are usually within walking distance. However, these SDCs need to be supplied from a central depot, using a heterogeneous and capacitated fleet of vehicles. We model this situation as a generalization of the covering tour problem, introduce the idea of split delivery, and propose an efficient heuristic approach to solve it. Numerical experiments on randomly-generated data show that, first, only very small instances can be solved efficiently using the mathematical model and, second, our heuristic produces high-quality solutions and solves real-size instances in a reasonable computing time.     

Lower and upper bounding procedures for the bin packing problem with concave loading cost

We address the one-dimensional bin packing problem with concave loading cost (BPPC), which commonly arises in less-than-truckload shipping services. Our contribution is twofold. First, we propose three lower bounds for this problem. The first one is the optimal solution of the continuous relaxation of the problem for which a closed form is proposed. The second one allows the splitting of items but not the fractioning of bins. The third one is based on a large-scale set partitioning formulation of the problem. In order to circumvent the challenges posed by the non-linearity of the objective function coefficients, we considered the inner-approximation of the concave load cost and derived a relaxed formulation that is solved by column generation. In addition, we propose two subset-sum-based heuristics. The first one is a constructive heuristic while the second one is a local search heuristic that iteratively attempts to improve the current solution by selecting pairs of bins and solving the corresponding subset sum-problem. We show that the worst-case performance of any BPPC heuristic and any concave loading cost function is bounded by 2. We present the results of an extensive computational study that was carried out on large set of benchmark instances. This study provides empirical evidence that the column generation-based lower bound and the local search heuristic consistently exhibit remarkable performance.     

Finding near-optimal independent sets at scale

The maximum independent set problem is NP-hard and particularly difficult to solve in sparse graphs, which typically take exponential time to solve exactly using the best-known exact algorithms. In this paper, we present two new novel heuristic algorithms for computing large independent sets on huge sparse graphs, which are intractable in practice. First, we develop an advanced evolutionary algorithm that uses fast graph partitioning with local search algorithms to implement efficient combine operations that exchange whole blocks of given independent sets. Though the evolutionary algorithm itself is highly competitive with existing heuristic algorithms on large social networks, we further show that it can be effectively used as an oracle to guess vertices that are likely to be in large independent sets. We then show how to combine these guesses with kernelization techniques in a branch-and-reduce-like algorithm to compute high-quality independent sets quickly in huge complex networks. Our experiments against a recent (and fast) exact algorithm for large sparse graphs show that our technique always computes an optimal solution when the exact solution is known, and it further computes consistent results on even larger instances where the solution is unknown. Ultimately, we show that identifying and removing vertices likely to be in large independent sets opens up the reduction space—which not only speeds up the computation of large independent sets drastically, but also enables us to compute high-quality independent sets on much larger instances than previously reported in the literature.     

Simulation-based Selectee Lane queueing design for passenger checkpoint screening

There are two kinds of passenger checkpoint screening lanes in a typical US airport: a Normal Lane and a Selectee Lane that has enhanced scrutiny. The Selectee Lane is not effectively utilized in some airports due to the small amount of passengers selected to go through it. In this paper, we propose a simulation-based Selectee Lane queueing design framework to study how to effectively utilize the Selectee Lane resource. We assume that passengers are classified into several risk classes via some passenger prescreening system. We consider how to assign passengers from different risk classes to the Selectee Lane based on how many passengers are already in the Selectee Lane. The main objective is to maximize the screening system’s probability of true alarm. We first discuss a steady-state model, formulate it as a nonlinear binary integer program, and propose a rule-based heuristic. Then, a simulation framework is constructed and a neighborhood search procedure is proposed to generate possible solutions based on the heuristic solution of the steady-state model. Using the passenger arrival patterns from a medium-size airport, we conduct a detailed case study. We observe that the heuristic solution from the steady-state model results in more than 4% relative increase in probability of true alarm with respect to the current practice. Moreover, starting from the heuristic solution, we obtain even better solutions in terms of both probability of true alarm and expected time in system via a neighborhood search procedure.     

The time-dependent capacitated profitable tour problem with time windows and precedence constraints

We introduce the time-dependent capacitated profitable tour problem with time windows and precedence constraints. This problem concerns determining a tour and its departure time at the depot that maximizes the collected profit minus the total travel cost (measured by total travel time). To deal with road congestion, travel times are considered to be time-dependent. We develop a tailored labeling algorithm to find the optimal tour. Furthermore, we introduce dominance criteria to discard unpromising labels. Our computational results demonstrate that the algorithm is capable of solving instances with up to 150 locations (75 pickup and delivery requests) to optimality. Additionally, we present a restricted dynamic programing heuristic to improve the computation time. This heuristic does not guarantee optimality, but is able to find the optimal solution for 32 instances out of the 34 instances.