Archive ouverte UNIGE | last documents for author 'Nicolas Zufferey'https://archive-ouverte.unige.ch/Latest objects deposited in the Archive ouverte UNIGE for author 'Nicolas Zufferey'engSolving the Wire-Harness Design Problem at a European car manufacturerhttps://archive-ouverte.unige.ch/unige:112831https://archive-ouverte.unige.ch/unige:112831abstract not availableFri, 11 Jan 2019 09:41:52 +0100Production and distribution planning for smoothing supply-chain variationshttps://archive-ouverte.unige.ch/unige:104864https://archive-ouverte.unige.ch/unige:104864abstract not availableFri, 01 Jun 2018 13:49:37 +0200Synchronizing heterogeneous vehicles in a routing and scheduling contexthttps://archive-ouverte.unige.ch/unige:104859https://archive-ouverte.unige.ch/unige:104859abstract not availableFri, 01 Jun 2018 13:42:40 +0200Multi-level tabu search for job scheduling in a variable-resource environmenthttps://archive-ouverte.unige.ch/unige:104856https://archive-ouverte.unige.ch/unige:104856abstract not availableFri, 01 Jun 2018 13:39:28 +0200Aircraft landing planning: past, present and futurehttps://archive-ouverte.unige.ch/unige:104854https://archive-ouverte.unige.ch/unige:104854Aircraft Landing Planning is a key issue for airports, as the runway capacity is their main limitation. In this paper, solution methods proposed in the literature are discussed, as wel l as current arrival traffic flow management practices. Based on this, the main missing features are identified and al low us designing the most promising directions for future research.Fri, 01 Jun 2018 13:37:06 +0200Metaheuristics for lexicographic optimization in industryhttps://archive-ouverte.unige.ch/unige:104850https://archive-ouverte.unige.ch/unige:104850In this paper, we highlight the legitimacy of adopting a lexicographic optimization framework for multi-objective problems arising in various industrial contexts such as production, transportation, telecommunication, and supply chain management. We describe the main ingredients of the metaheuristics that are successfully applied to these diverse practical situations.Fri, 01 Jun 2018 13:32:10 +0200Reactive variable neighborhood searchhttps://archive-ouverte.unige.ch/unige:104849https://archive-ouverte.unige.ch/unige:104849Recent works have shown that the variable neighborhood search (VNS) algorithm can be improved by using a reactive component. A component is reactive if its behavior changes dynamically depending on the information obtained from the current run, or depending on the instance to solve. This paper gives a review of these recent works, and proposes a new scheme to select reactively the move used in the shaking step of VNS. This new mechanism is particularly relevant to solve problems involving various types of decisions. For instance, an application of the resulting Reactive-VNS to a short-term production-planning problem is given. Experimental results show that Reactive-VNS outperforms the classical VNS approach on the latter problem.Fri, 01 Jun 2018 13:30:36 +0200Fit between humanitarian professionals and project requirements: hybrid group decision procedure to reduce uncertainty in decision-makinghttps://archive-ouverte.unige.ch/unige:104847https://archive-ouverte.unige.ch/unige:104847Choosing the right professional that has to meet indeterminate requirements is a critical aspect in humanitarian development and implementation projects. This paper proposes a hybrid evaluation methodology for some non-governmental organizations enabling them to select the most competent expert who can properly and adequately develop and implement humanitarian projects. This methodology accommodates various stakeholders’ perspectives in satisfying the unique requirements of humanitarian projects that are capable of handling a range of uncertain issues from both stakeholders and project requirements. The criteria weights are calculated using a two-step multi-criteria decision-making method: (1) fuzzy analytical hierarchy process for the evaluation of the decision maker weights coupled with (2) technique for order preference by similarity to ideal solution to rank the alternatives which provide the ability to take into account both quantitative and qualitative evaluations. Sensitivity analysis have been developed and discussed by means of a real case of expert selection problem for a non-profit organisation. The results show that the approach allows a decrease in the uncertainty associated with decision-making, which proves that the approach provides robust solutions in terms of sensitivity analysis.Fri, 01 Jun 2018 13:18:22 +0200Learning variable neighborhood search for a scheduling problem with time windows and rejectionshttps://archive-ouverte.unige.ch/unige:104846https://archive-ouverte.unige.ch/unige:104846Variable neighborhood search is a local search metaheuristic that uses sequentially different neighborhood structures. This method has been successfully applied to various types of problems. In this work, variable neighborhood search is enhanced with a learning mechanism which helps to drive the search toward promising areas of the search space. The resulting method is applied to a single-machine scheduling problem with rejections, setups, and earliness and tardiness penalties. Experiments are conducted for instances from the literature. They show on the one hand the benefit of the learning mechanism (in terms of solution quality and robustness). On the other hand, the proposed method significantly outperforms state-of-the-art algorithms for the considered problem. Moreover, its flexibility allows its straightforward adaptation to other combinatorial optimization problems.Fri, 01 Jun 2018 13:15:14 +0200Collaborative agent teams (cat): from the paradigm to implementation guidelineshttps://archive-ouverte.unige.ch/unige:104845https://archive-ouverte.unige.ch/unige:104845We propose a general solution method framework based on a Collaborative Agent Teams (CAT) architecture to tackle large-scale mixed-integer optimization problems with complex structures. This frame- work introduces several conceptual improvements over previous agent teams’ approaches. We discuss how to configure the three key compo- nents of a CAT solver for multidimensional optimization problems: the problem representation, the design of agents, and the information sharing mechanisms between agents. Implementation guidelines are also given.Fri, 01 Jun 2018 13:11:31 +0200Collaborative variable neighborhood searchhttps://archive-ouverte.unige.ch/unige:104844https://archive-ouverte.unige.ch/unige:104844Variable neighborhood search (VNS) is a well-known meta- heuristic. Two main ingredients are needed for its design: a collection M = (N1, . . . ,Nr) of neighborhood structures and a local search LS (of- ten using its own single neighborhood L).M has a diversification purpose (search for unexplored zones of the solution space S), whereas LS plays an intensification role (focus on the most promising parts of S). Usually, the used set M of neighborhood structures relies on the same type of modification (e.g., change the value of i components of the decision vari- able vector, where i is a parameter) and they are built in a nested way (i.e., Ni is included in Ni+1). The more difficult it is to escape from the currently explored zone of S, the larger is i, and the more capability has the search process to visit regions of S which are distant (in terms of solution structure) from the incumbent solution. M is usually designed independently from L. In this paper, we depart from this classical VNS framework and discuss an extension, Collaborative Variable Neighbor- hood Search (CVNS), where the design of M and L is performed in a collaborative fashion (in contrast with nested and independent), and can rely on various and complementary types of modifications (in contrast with a common type with different amplitudes).Fri, 01 Jun 2018 13:09:32 +0200Ants cannot color graphshttps://archive-ouverte.unige.ch/unige:104577https://archive-ouverte.unige.ch/unige:104577Ant colony optimizationwas born in 1991. Fromthat date, ant algorithms have been applied to numerous problems. One of its most famous adaptations was for graph coloring in 1997, resulting in an inspiring publication entitled ”Ants can color graphs” and citedmore than 700 times [3]. In this paper, with a certain sense of humor reflected in its title, it is showed that actually, 20 years later, ants cannot color graphs. Tis fact is explained in the light of two weaknesses of the usual ant methodology when making decisions, namely a cumbersome computation and the joint consideration of conflicting ingredients. A reconciliation of these contradictory statements relies in the enlargement of the ant optimization paradigm.Tue, 22 May 2018 09:46:33 +0200All-terrain tabu search approaches for production management problemshttps://archive-ouverte.unige.ch/unige:100778https://archive-ouverte.unige.ch/unige:100778A metaheuristic is a refined solution method able to find a satisfying solution to a difficult problem in a reasonable amount of time. A local search metaheuristic works on a single solution and tries to improve it iteratively. Tabu search is one of the most famous local search, where at each iteration, a neighbor solution is generated from the current solution by performing a specific modification (called a move) on the latter. The goal of this chapter is to present tabu search approaches with enhanced exploration and exploitationmechanisms. For this purpose, the following ingredients are discussed: different neighborhood structures (i.e., different types of moves), guided restarts based on a distance function, and deconstruction/reconstruction techniques. The resulting all-terrain tabu search approaches are illustrated for various production problems: car sequencing, job scheduling, resource allocation, and inventory management.Thu, 21 Dec 2017 16:18:22 +0100A simulation-optimization approach for the production of components for a pharmaceutical companyhttps://archive-ouverte.unige.ch/unige:100777https://archive-ouverte.unige.ch/unige:100777abstract not availableThu, 21 Dec 2017 16:15:31 +0100Impact of vehicle tracking on a routing problem with dynamic travel timeshttps://archive-ouverte.unige.ch/unige:100292https://archive-ouverte.unige.ch/unige:100292This paper evaluates the benefits of data obtained via modern information technologies, such as global positioning systems, when solving a vehicle routing problem with dynamic customer requests and dynamic travel times. It is empirically demonstrated that substantial improvements are achieved over a previously reported model which does not assume the availability of such information. We also analyze how the system handles dynamic perturbations to the travel times that lead to earliness or lateness in the planned schedule.Wed, 13 Dec 2017 16:47:35 +0100Inventory deployment with uncertainty on production and lead-times, proceedings of the international conference on industrial engineering and systems managementhttps://archive-ouverte.unige.ch/unige:99575https://archive-ouverte.unige.ch/unige:99575An international Swiss company is facing a complex inventory-deployment problem where expensive items of different models must be dispatched to wholesalers to finally reach the shops. Perturbations are expected at two levels (namely, production and transportation), and efficient reactions must be implemented to face to these uncertainties. Fast and efficient solution methods are proposed to solve realistic instances.Thu, 23 Nov 2017 16:39:25 +0100A push shipping-dispatching approach for high-value items: from modeling to managerial insights, proceedings of the international conference optimization and decision sciencehttps://archive-ouverte.unige.ch/unige:99574https://archive-ouverte.unige.ch/unige:99574real shipping-dispatching problem is considered in a three-level supply chain (plant, wholesalers, shops). Along the way, different perturbations are expected (when manufacturing, when forecasting the demand, and when dispatching the inventory from the wholesalers level), and accurate reactions must be taken. An integer linear program is proposed and some managerial insights are given.Thu, 23 Nov 2017 16:37:23 +0100Multi-modal variations of the vehicle routing problem, proceedings of the 1st informs conference of the transportation science & logistics societyhttps://archive-ouverte.unige.ch/unige:99572https://archive-ouverte.unige.ch/unige:99572In this work, we extend the Vehicle Routing Problem formulation by proposing multi-modal variations of this well-established problem. For these new formulations, we empirically show that a streamline metaheuristic is able to highlight the potential benefit offered by the introduction of multi-modality.Thu, 23 Nov 2017 16:35:59 +0100Performance versus risk in a food supply chain, proceedings of the 3rd international conference project-logistichttps://archive-ouverte.unige.ch/unige:99570https://archive-ouverte.unige.ch/unige:99570In the context of a growing world population using limited resources responsibly, this research is motivated by increasing pressures to the food industry, with issues on food security, safety and waste. Considering a global food supply chain environment, processing time and cost (PTC) are combined with operational risk (ORk) in a novel integrated approach for designing and optimizing monitoring systems. Studying a flagship product widely consumed in the world, the provided quantitative analysis and results are based on real-world data obtained from an international food company. Evidence is given that the designed multiobjective methodology reveals quantitative insights concerning the non-linear relationship between PTC and ORk. Indeed, we numerically show how to significantly decrease PTC with only minor increase of ORk. Moreover, critical monitoring activities were identified, allowing a step-by-step optimization of the entire monitoring system. Generalizable insights are derived for practitioners.Thu, 23 Nov 2017 16:30:09 +0100Energy consumption vs economic profit in supply chain networks, proceedings of the 18th eu/me workshop on metaheuristics for a better worldhttps://archive-ouverte.unige.ch/unige:99569https://archive-ouverte.unige.ch/unige:99569Using a detailed energy consumption infrastructure, this work develops a model to address sustainability in supply chain networks in comparing environmental impact to cost optimization. It takes into account the effect of inventory management policies on energy consumption in transportation and storage. The considered inventory management policy is the replenishment method. A basic supply chain segment is studied in the most general fashion so that the analysis could be applied to any transportation link and storage node of a supply chain. Real data are used, obtained from an Indonesian food factory belonging to a well-known international company. Using a simulation-optimization approach, our results show that the most energy effcient solution is not necessarily detrimental to the economic profits.Thu, 23 Nov 2017 16:22:49 +0100Optimizing the standardization of car cables, proceedings of the 18th annual congress of the french operations research societyhttps://archive-ouverte.unige.ch/unige:99566https://archive-ouverte.unige.ch/unige:99566A set-partitioning problem for car wirings is addressed, in order to minimize the production costs. An approach to divide the problem in smaller sub-problems is presented, and three different constructive methods are proposed and compared according to quality, computing time, and robustness.Thu, 23 Nov 2017 15:53:27 +0100Vehicle routing with multi-modality: a practical application, proceedings of the 18th annual congress of the french operations research societyhttps://archive-ouverte.unige.ch/unige:99564https://archive-ouverte.unige.ch/unige:99564This work proposes multi-modal variants of the well-known Vehicle Routing Problem. It will be empirically shown that even a relatively simple metaheuristic (namely, the Large Neighborhood Search) is able to highlight the potential benefit of multi-modality.Thu, 23 Nov 2017 15:51:21 +0100Model and metaheuristics for a scheduling problem integrating procurement, sale and distribution decisionshttps://archive-ouverte.unige.ch/unige:99560https://archive-ouverte.unige.ch/unige:99560This paper presents an integrated approach for short-term supply chain management (SCM) at a fast moving consumer goods production plant. The problem is to determine the production quantities, to provide a detailed production schedule, to trigger the relevant express deliveries of raw material, and to manage the distribution. We propose a linear integer model, which integrates all of these decisions within scheduling. To find high quality solutions in a reasonable amount of time, various solution methods are proposed, such as a greedy constructive heuristic, two tabu search metaheuristics, a basic variable neighborhood search and an enhanced one, which uses a variable shaking operator. Experiments on realistic instances show that the latter method is efficient and robust. This paper is a contribution to the SCM literature (indeed, only few references address the integration of short term decisions and to the general metaheuristics field (as the variable neighborhood search paradigm is extended).Thu, 23 Nov 2017 15:41:35 +0100Graph multi-coloring for a job scheduling applicationhttps://archive-ouverte.unige.ch/unige:92813https://archive-ouverte.unige.ch/unige:92813In this paper, we introduce a graph multi-coloring problem where each vertex must be assigned a given number of different colors, represented as integers, and no two adjacent vertices can share a common color. In the variant considered, the number of available colors is such that not all vertices can be colored. Furthermore, there is a bound on the number of vertices which can be assigned the same color. A gain is associated with each vertex and the first objective is to maximize the total gain over all colored vertices. Secondary objectives consider the sequence of colors assigned to each vertex. More precisely, the range and the number of interruptions must be minimized, where the range corresponds to the difference between the largest and smallest colors assigned to a vertex. This variant of the graph multi-coloring problem is of interest because it can model practical job scheduling applications. An integer linear programming formulation is first proposed to address small-size instances. A construction heuristic, as well as local search methods, are then reported to tackle larger instances. The local search methods are based on several neighborhood structures, each one focusing on a specific property of the problem. Different ways to combine these neighborhood structures are also investigated.Fri, 24 Mar 2017 15:40:50 +0100Makespan minimisation for a parallel machine scheduling problem with preemption and job incompatibilityhttps://archive-ouverte.unige.ch/unige:92812https://archive-ouverte.unige.ch/unige:92812In this paper, an extension of the graph coloring problem is introduced to model a parallel machine scheduling problem with job incompatibility. To get closer to real-world applications, where the number of machines is limited and jobs have different processing times, each vertex of the graph requires multiple colors and the number of vertices with the same color is bounded. In addition, several objectives related to scheduling are considered: makespan, number of preemptions, and summation over the jobs' throughput times. Different solution methods are proposed, namely, two greedy heuristics, two tabu search methods and an adaptive memory algorithm. The latter uses multiple recombination operators, each one being designed for optimizing a subset of objectives. The most appropriate operator is selected dynamically at each iteration, depending on its past performance. Experiments show that the proposed algorithm is effective and robust, while providing high quality solutions on benchmark instances for the graph multi-coloring problem, a simplification of the considered problem.Fri, 24 Mar 2017 15:38:15 +0100Metaheuristics for truck loading in the car production industryhttps://archive-ouverte.unige.ch/unige:92811https://archive-ouverte.unige.ch/unige:92811The delivery of goods to car factories is a challenging problem. The French car manufacturer Renault is facing daily a complex truck loading problem where various goods must be packed into a truck such that they fulfill different constraints. As trucks can deliver goods to different factories on the same tour, classes of items have been defined, where a class is associated with a delivery point. The consideration of these classes in addition to large standard deviations over the sizes of the items are new features in the packing literature. Because of the problem structure and of the computation time limit constraint imposed by practitioners, it will be shown that exact algorithms are not appropriate from a practical standpoint. We propose efficient metaheuristics to tackle this problem. First, in contrast with the classical literature, the proposed tabu search relies on the joint use of different types of moves (an efficient diversification mechanism is also proposed to enhance its performance). Then, the recombination operator used in the developed genetic algorithm takes into account all the problem features and is able to build well-balanced offspring solutions. Finally, within the framework of ant algorithms, the benefit of an unconventional decision selection mechanism is discussed. An extension of the problem is proposed at the end, which consists in tackling all the instances within a common time limit. In this context, it will be showed that a combination of the algorithms is the most powerful strategy.Fri, 24 Mar 2017 15:35:42 +0100Three-level inventory deployment for a luxury watch company facing various perturbationshttps://archive-ouverte.unige.ch/unige:92808https://archive-ouverte.unige.ch/unige:92808A well-known Swiss watch brand, active in the top-end luxury market, is facing a complex inventory deployment problem where watches of different models (more than 100 different models) must be dispatched first to wholesalers to finally reach the shops where consumers come in. Along the way, different perturbations are expected at three levels (production plan, demand, and dispatching), and accurate reactions must be taken to fit to these uncertainties. Solution methods are proposed to solve realistic instances. Results show the relevance of the methods and the robustness of the solutions.Fri, 24 Mar 2017 15:17:33 +0100Competitive and timely food supply combined with operational riskhttps://archive-ouverte.unige.ch/unige:92807https://archive-ouverte.unige.ch/unige:92807The food industry is facing increasing pressures with issues of food safety, waste and sustainable development for a growing population, while using limited resources. We propose a methodology that combines processing time and cost (PTC) with operational risk (ORk) to address these challenges while considering economic interest of each player in the supply chain environment. Quantitative results from a global food company are presented, which show that the joint analysis of PTC and ORk, in contrast with each of the single objective approaches, result in quantifiable, tangible insights for actual practice. The step-by-step optimization enables each player to reconfigure its activities for competitive advantage.Fri, 24 Mar 2017 15:15:58 +0100A global simulation-optimization approach for inventory management in a decentralized supply chainhttps://archive-ouverte.unige.ch/unige:92806https://archive-ouverte.unige.ch/unige:92806The optimal calibration of inventory management policies in a multi-echelon linear supply chain (SC) constitutes a complex problem. The difficulty to provide a clear and unique answer leads to frustration in using serious games for educational purposes. In this work, a simulation-based optimization approach, using a dynamic tabu search metaheuristic, is developed and applied to a specific case of a four-echelon linear SC. A stochastic demand with constant mean is used and different service levels to the market are considered. The results indicate significant differences of the required on-hand inventory, cost contributions and cost element importance between the different echelons. The non-linear influence of the service level on the various parameters and costs is also observed. The obtained results are consistent and provide a clear answer to the stated problem. The methodology developed can be easily generalized to other cases.Fri, 24 Mar 2017 15:12:31 +0100Efficient scheduling policies for dynamic data flow programs executed on multi-corehttps://archive-ouverte.unige.ch/unige:91453https://archive-ouverte.unige.ch/unige:91453An important challenge of dataflow program implementations on multi-core platforms is the partitioning and scheduling providing the best possible throughput when satisfying multiple objective functions. Not only it has been proven that these problems are NP-complete, but also the quality of any heuristic approach can be affected by other factors (e.g., buffer dimensioning, influence of an established partitioning configuration and scheduling strategy on each other, uncertainties of a compiler affecting the profiling information). This paper proposes an evaluation of alternative partitioning and scheduling configurations based on the application profiling results. It investigates the impact of the scheduling on the overall execution time, and verifies which policies could further drive the metaheuristic-based search of a close-to-optimal partitioning configuration.Mon, 30 Jan 2017 09:14:31 +0100Food supply tomorrow: combining performance & operational risk?https://archive-ouverte.unige.ch/unige:91445https://archive-ouverte.unige.ch/unige:91445The food industry is facing increasing pressures with issues of food safety, waste and sustainable development for a growing population, while using limited resources. We propose a methodology that combines processing time and cost(PTC) with operational risk (ORk) to address these challenges while considering economic interest of each player in the supply chain. Quantified evidence from a global food company is presented, which shows that the joint analysis of PTC and ORk, in contrast to each single objective approach, results in tangible insights for actual practice. The step-by-step optimization enables each player to reconfigure its activities for competitive advantage.Fri, 27 Jan 2017 16:47:41 +0100Local inventory management policies for optimal global performances in a decentralized supply chainhttps://archive-ouverte.unige.ch/unige:91444https://archive-ouverte.unige.ch/unige:91444The optimal calibration of inventory management policies in a multi-echelon linear supply chain (SC) constitutes a complex problem. The difficulty to provide a clear and unique answer leads to frustration in using serious games for educational purposes. In this work, a simulation-based optimization approach, using a dynamic tabu search metaheuristic, is developed and applied to a specific case of a four-echelon linear SC. A stochastic demand with constant mean is used and different service levels to the market are considered. The results indicate significant differences of the required on-hand inventory, cost contributions and cost element importance between the different echelons. The non-linear influence of the service level on the various parameters and costs is also observed. The obtained results are consistent and provide a clear answer to the stated problem. The methodology developed can be easily generalized to other cases.Fri, 27 Jan 2017 15:56:50 +0100Collaborative agent teams for a supply chain network design problemhttps://archive-ouverte.unige.ch/unige:91443https://archive-ouverte.unige.ch/unige:91443When facing complex optimization models, it is relevant to use the best tools available to tackle each model or sub-model. A multi-agent system allows for that much flexibility. Extending an earlier study, this paper proposes a Collaborative Agent Teams (CAT) approach to tackle multi-period multi-product supply chain network design problems, for which new best results are obtained.Fri, 27 Jan 2017 15:48:57 +0100Dynamic multi-trip vehicle routing with unusual time-windows for the pick-up of blood samples and delivery of medical materialhttps://archive-ouverte.unige.ch/unige:91202https://archive-ouverte.unige.ch/unige:91202Given a fleet of identical vehicles and a set of n clients to be served from a single depot, the well-known vehicle routing problem (VRP) consists in serving each client (with a deterministic demand) once with a unique vehicle, with the aim of minimizing the total traveled distance. In this work, the basic VRP is extended within a medical environment, leading to MVRP (for medical VRP). Indeed, the depot is typically a laboratory for blood analysis, and a client is assumed to be a medical location at which blood samples should be picked up by a vehicle. In order to have efficient tests at the laboratory, at most 90 minutes should elapse between the release time of the blood sample and the delivery time at the laboratory. In addition, only a proportion of the demand is known in advance and the travel times depend on the traffic conditions. A fleet of non-identical vehicle is considered (with different speeds and capacities), and each location has to be visited anytime a blood sample is available. Finally, medical items should be daily delivered from the laboratory to some medical locations. A transportation cost function with three components has to be minimized. Solution methods are proposed, which are able to account for all the specific features of the problem. The experiments highlight the benefit of considering diversion opportunities (which consists in diverting a vehicle away from its planned destinations).Fri, 20 Jan 2017 16:21:12 +010018th international conference on computational management sciencehttps://archive-ouverte.unige.ch/unige:91199https://archive-ouverte.unige.ch/unige:91199Ant algorithms are well-known metaheuristics which have been widely used since two decades. In most of the literature, an ant is a constructive heuristic able to build a solution from scratch. However, other types of ant algorithms have recently emerged: the discussion is thus not limited by the common framework of the constructive ant algorithms. Generally, at each generation of an ant algorithm, each ant builds a solution step by step by adding an element to it. Each choice is based on the greedy force (also called the visibility, the short term profit or the heuristic information) and the trail system (central memory which collects historical information of the search process). Usually, all the ants of the population have the same characteristics and behaviors. In contrast in this paper, a new type of ant metaheuristic is proposed, namely SMART (for Solution Methods with Ants Running by Types). It relies on the use of different population of ants, where each population has its own personality.Fri, 20 Jan 2017 15:24:30 +0100Performance estimation based multi-criteria partitioning approach for dynamic dataflow programshttps://archive-ouverte.unige.ch/unige:91075https://archive-ouverte.unige.ch/unige:91075The problem of partitioning a dataflow program onto a target architecture is a difficult challenge for any application design. In general, since the problem is NP-complete, it consists of looking for high quality solutions in terms of maximizing the achievable data throughput. The difficulty is given by the exploration of the design space which results in being extremely large for parallel platforms. The paper describes a heuristic partitioning methodology applicable to dynamic dataflow programs. The methodology is based on two elements: an execution model of the dynamic dataflow programwhich is used as estimation of the performance for the exploration of the large design space and several partitioning algorithms competing to lead to specific high quality solutions. Experimental results are validated with executions on a virtual platform.Fri, 13 Jan 2017 16:32:48 +0100Tabu search approaches for two car sequencing problems with smoothing constraintshttps://archive-ouverte.unige.ch/unige:91065https://archive-ouverte.unige.ch/unige:91065Nowadays, new constraints, known as smoothing constraints, are attracting a growing attention in the area of job scheduling, and in particular for car sequencing problems, where cars must be scheduled before production in an order respecting various constraints (colors, optional equipments, due dates, etc.), while avoiding overloading some important resources. The first objective of the car industry is to assign a production day to each customer-ordered car and the second one consists of scheduling the order of cars to be put on the line for each production day, while satisfying as many requirements as possible of the plant shops (e.g., paint shop, assembly line). The goal of this chapter is to propose tabu search approaches for two car sequencing problems involving smoothing constraints. The first one is denoted (P1) and was the subject of the ROADEF 2005 international Challenge proposed by the automobile manufacturer Renault, whereas the second one is denoted(P2) and extends some important features of (P1).Fri, 13 Jan 2017 15:34:06 +0100Metaheuristics for a job scheduling problem with smoothing costs relevant for the car industryhttps://archive-ouverte.unige.ch/unige:91056https://archive-ouverte.unige.ch/unige:91056on nonidentical machines with applications in the car industry, inspired by the problem proposed by the car manufacturer Renault in the ROADEF 2005 Challenge. Makespan, smoothing costs and setup costs are minimized following a lexicographic order, where smoothing costs are used to balance resource utilization. We first describe a mixed integer linear programming (MILP) formulation and a network interpretation as a variant of the well-known vehicle routing problem. We then propose and compare several solution methods, ranging from greedy procedures to a tabu search and an adaptive memory algorithm. For small instances (with up to 40 jobs) whose MILP formulation can be solved to optimality, tabu search provides remarkably good solutions. The adaptive memory algorithm, using tabu search as an intensification procedure, turns out to yield the best results for large instances.Fri, 13 Jan 2017 15:25:21 +0100Tabu search for partitioning dynamic dataflow programshttps://archive-ouverte.unige.ch/unige:91053https://archive-ouverte.unige.ch/unige:91053An important challenge of dataflow programming is the problem of partitioning dataflow components onto a target architecture. A common objective function associated to this problem is to find the maximum data processing throughput. This NP-complete problem is very difficult to solve with high quality close-to-optimal solutions for the very large size of the design space and the possibly large variability of input data. This paper introduces four variants of the tabu search metaheuristic expressly developed for partitioning components of a dataflow program. The approach relies on the use of a simulation tool, capable of estimating the performance for any partitioning configuration exploiting a model of the target architecture and the profiling results. The partitioning solutions generated with tabu search are validated for consistency and high accuracy with experimental platform executions.Fri, 13 Jan 2017 15:22:47 +0100Order acceptance and scheduling with earliness and tardiness penaltieshttps://archive-ouverte.unige.ch/unige:91048https://archive-ouverte.unige.ch/unige:91048This paper addresses a production scheduling problem in a single-machine environment, where a job can be either early, on time, late, or rejected. In order acceptance and scheduling (OAS) contexts, it is assumed that the production capacity of a company is overloaded. The problem is therefore to decide which orders to accept and how to sequence their production. In contrast with the existing literature, the considered problem jointly takes into account the following features: earliness and tardiness penalties(which can be linear or quadratic), sequence-dependent setup times and costs, rejection penalties, and the possibility of having idle times. The practical relevance of this new NP-hard problem is discussed and various solution methods are proposed, ranging from a basic greedy algorithm to refined metaheuristics (e.g., tabu search, the adaptive memory algorithm, population-based approaches loosely inspired on ant algorithms). The methods are compared for instances with various structures containing up to 200 jobs. For small linear instances, the metaheuristics are favorably compared with an exact formulation using CPLEX 12.2. Managerial insights and recommendations are finally given.Fri, 13 Jan 2017 15:09:48 +0100[Review of:] Jean-François Billeter, contre François Jullienhttps://archive-ouverte.unige.ch/unige:83674https://archive-ouverte.unige.ch/unige:83674abstract not availableWed, 11 May 2016 15:12:23 +0200Ant Metaheuristics with Adapted Personalities for the Vehicle Routing Problemhttps://archive-ouverte.unige.ch/unige:79982https://archive-ouverte.unige.ch/unige:79982abstract not availableFri, 22 Jan 2016 14:12:54 +0100Multiple Neighborhood in Tabu Search: Successful Applications in Operations Managementhttps://archive-ouverte.unige.ch/unige:79974https://archive-ouverte.unige.ch/unige:79974abstract not availableFri, 22 Jan 2016 13:09:26 +0100Graph coloring models and metaheuristics for packing applicationshttps://archive-ouverte.unige.ch/unige:79971https://archive-ouverte.unige.ch/unige:79971abstract not availableFri, 22 Jan 2016 13:07:11 +0100Metaheuristics for constrained production scheduling problemshttps://archive-ouverte.unige.ch/unige:73544https://archive-ouverte.unige.ch/unige:73544In this thesis, high level metaheuristics are proposed to solve production scheduling problems. Production scheduling is a crucial function for companies, which helps to manage the operations at the production shop level. It aims to reduce the costs, to increase the service level to customers, and to achieve better production efficiency. The design of powerful optimization methods for such problems is then decisive for the competitiveness of the companies. This thesis is divided into five chapters which focus on different production scheduling problems. For each problem, an integer linear programming formulation is given, and various heuristics and metaheuristics are proposed and compared. The first two chapters concern an extension of the graph coloring problem, which is extended to correspond to realistic job scheduling problems. Chapters three and four are related to a single machine scheduling problem. The last chapter studies the integration of short term supply chain management decisions.Wed, 24 Jun 2015 12:04:01 +0200From packing to dispatching: supply chain optimization techniqueshttps://archive-ouverte.unige.ch/unige:73467https://archive-ouverte.unige.ch/unige:73467This thesis aims at proposing relevant optimization techniques, such as metaheuristics, for various logistics problems faced by international companies. Four different projects are studied, each one having its own specificity. They all appear at different levels of the supply chain.Wed, 24 Jun 2015 09:14:58 +0200Tabu search for a single machine scheduling problem with rejected jobs, setups and deadlineshttps://archive-ouverte.unige.ch/unige:73316https://archive-ouverte.unige.ch/unige:73316This paper addresses a single machine scheduling problem with release dates, deadlines, setup costs and times, and the possibility to reject some jobs while encountering an abandon cost. The objective function to minimize is a sum of regular functions depending on the completion time of the jobs. The problem is inspired by a manufacturing scheduling problem. We design a greedy algorithm and a tabu search approach for the problem. We studied several restriction procedures. Realistic instances with up to 500 jobs are tackled.Tue, 23 Jun 2015 08:48:00 +0200Graph Coloring Tabu Search for Project Schedulinghttps://archive-ouverte.unige.ch/unige:73315https://archive-ouverte.unige.ch/unige:73315Consider a project consisting of a set of n operations to be performed. Some pairs {j,j′} of operations are incompatible, which can have two different meanings. On the one hand, it can be allowed to perform j and j′ at common time periods. In such a case, incompatibility costs are encountered and penalized in the objective function. On the other hand, it can be strictly forbidden to perform j and j′ concurrently. In such a case, the overall project duration has to be minimized. In this paper, three project scheduling problems (P 1), (P 2) and (P 3) are considered. It will be showed that tabu search relying on graph coloring models is a very competitive method for such problems. The overall approach is called graph coloring tabu search and denoted GCTS.Tue, 23 Jun 2015 08:47:36 +0200Adaptive Memory Algorithm with the Covering Recombination Operatorhttps://archive-ouverte.unige.ch/unige:73314https://archive-ouverte.unige.ch/unige:73314The adaptive memory algorithm (AMA) is a population-based metaheuristics initially developed in 1995 by Rochat and Taillard. AMA relies on a central memory M and consists in three steps: generate a new solution s from M with a recombination operator, improve s with a local search operator, and use s to update M with a memory update operator. In 1999, Galinier and Hao proposed the GPX recombination operator for the graph coloring problem. In this paper, AMC, a general type of evolutionary algorithm, is formalized and called Adaptive Memory Algorithm with the Covering Recombination Operator. It relies on a specific formulation of the considered problem and on a generalization of the GPX recombination operator. It will be showed that AMC has obtained promising results in various domains, such as graph coloring, satellite range scheduling and project management.Tue, 23 Jun 2015 08:47:12 +0200Learning Tabu Search for Combinatorial Optimizationhttps://archive-ouverte.unige.ch/unige:73313https://archive-ouverte.unige.ch/unige:73313In this paper, a new type of local search algorithm is proposed, called Learning Tabu Search and denoted LTS. It is assumed that any solution of the considered problem can be represented with a list of characteristics. LTS involves a learning process relying on a trail system. The trail system is based on the idea that if some combinations of characteristics often belong to good solutions during the search process, such combinations of characteristics should be favored when generating new solutions. It will be showed that LTS obtained promising results on a refueling problem in a railway network.Tue, 23 Jun 2015 08:46:44 +0200A learning tabu search for a truck allocation problem with linear and nonlinear cost componentshttps://archive-ouverte.unige.ch/unige:73268https://archive-ouverte.unige.ch/unige:73268The two-level problem studied in this article consists of optimizing the refueling costs of a fleet of locomotives over a railway network. The goal consists of determining: (1) the number of refueling trucks contracted for each yard (truck assignment problem denoted TAP) and (2) the refueling plan of each locomotive (fuel distribution problem denoted FDP). As the FDP can be solved efficiently with existing methods, the focus is put on the TAP only. In a first version of the problem (denoted (P1)), various linear costs (e.g., fuel, fixed cost associated with each refueling, weekly operating costs of trucks) have to be minimized while satisfying a set of constraints (e.g., limited capacities of the locomotives and the trucks). In contrast with the existing literature on this problem, two types of nonlinear cost components will also be considered, based on the following ideas: (1) if several trucks from the same fuel supplier are contracted for the same yard, the supplier is likely to propose discounted prices for that yard (Problem (P2)); (2) if a train stops too often on its route, a penalty is incurred, which represents the dissatisfaction of the clients (Problem (P3)). Even if exact methods based on a mixed integer linear program formulation are available for (P1), they are not appropriate anymore to tackle (P2) and (P3). Various methods are proposed for the TAP: a descent local search, a tabu search, and a learning tabu search (LTS). The latter is a new type of local search algorithm. It involves a learning process relying on a trail system, and it can be applied to any combinatorial optimization problem. Results are reported and discussed for a large set of instances (for (P1), (P2), and (P3)), and show the good performance of LTS.Mon, 22 Jun 2015 11:06:18 +0200Metaheuristics for a scheduling problem with rejection and tardiness penaltieshttps://archive-ouverte.unige.ch/unige:73267https://archive-ouverte.unige.ch/unige:73267In this paper, we consider a single-machine scheduling problem (P) inspired from manufacturing instances. A release date, a deadline, and a regular (i.e., non-decreasing) cost function are associated with each job. The problem takes into account sequence-dependent setup times and setup costs between jobs of different families. Moreover, the company has the possibility to reject some jobs/orders, in which case a penalty (abandon cost) is incurred. Therefore, the problem at hand can be viewed as an order acceptance and scheduling problem. Order acceptance problems have gained interest among the research community over the last decades, particularly in a make-to-order environment. We propose and compare a constructive heuristic, local search methods, and population-based algorithms. Tests are performed on realistic instances and show that the developed metaheuristics significantly outperform the currently available resolution methods for the same problem.Mon, 22 Jun 2015 11:05:48 +0200A Generalized Consistent Neighborhood Search for Satellite Range Scheduling Problemshttps://archive-ouverte.unige.ch/unige:73266https://archive-ouverte.unige.ch/unige:73266Many optimization problems require the use of a local search to find a satisfying solution in a reasonable amount of time, even if the optimality is not guaranteed. Usually, local search algorithms operate in a search space which contains complete solutions (feasible or not) to the problem. In contrast, in Consistent Neighborhood Search (CNS), after each variable assignment, the conflicting variables are deleted to keep the partial solution feasible, and the search can stop when all the variables have a value. In this paper, we propose a generalized version of CNS, discuss its performance according to various criteria, and present successful adaptations of CNS to three types of satellite range scheduling problems. Such problems are motivated by applications encountered by the French National Space and Aeronautic Agencies and the US Air Force Satellite Control Network. The described numerical experiments will demonstrate that CNS is a powerful and flexible method, which can be easily combined with efficient ingredients.Mon, 22 Jun 2015 11:04:48 +0200Solution Methods for Fuel Supply of Trainshttps://archive-ouverte.unige.ch/unige:47595https://archive-ouverte.unige.ch/unige:47595The considered problem consists in optimizing the refueling costs of a fleet of locomotives over a railway network. The goal consists in determining the number of trucks contracted for each yard (truck assignment problem) and to determine the refueling plan of each locomotive (fuel distribution problem), while minimizing the costs and satisfying constraints. A two-levels approach is proposed to tackle this NP-hard problem. Three metaheuristics (namely a descent procedure, a tabu search, and an ant local search algorithm) are proposed for the truck assignment level, and a flow model is designed for the fuel distribution level. A post-optimization procedure can be combined with the latter flow model. Six algorithms are proposed for the whole problem, and were tested on a realistic instance proposed by the Railway Applications Section of INFORMS. Competitive results were obtained. A strength of the proposed approach is its flexibility, as it can be easily adapted to non linear cases.Thu, 05 Mar 2015 15:19:20 +0100Ant Local Search for Combinatorial Optimizationhttps://archive-ouverte.unige.ch/unige:47395https://archive-ouverte.unige.ch/unige:47395In ant algorithms, each individual ant makes decisions according to the greedy force (short term profit) and the trail system based on the history of the search (information provided by other ants). Usually, each ant is a constructive process, which starts from scratch and builds step by step a complete solution of the considered problem. In contrast, in Ant Local Search (ALS), each ant is a local search, which starts from an initial solution and tries to improve it iteratively. In this paper are presented and discussed successful adaptations of ALS to different combinatorial optimization problems: graph coloring, a refueling problem in a railway network, and a job scheduling problem.Tue, 03 Mar 2015 16:21:57 +0100La coloration des sommets d'un graphe par colonies de fourmishttps://archive-ouverte.unige.ch/unige:46139https://archive-ouverte.unige.ch/unige:46139abstract not availableMon, 02 Feb 2015 09:36:04 +0100Tabu Search for an Intermodal Transportation Problemhttps://archive-ouverte.unige.ch/unige:46138https://archive-ouverte.unige.ch/unige:46138abstract not availableMon, 02 Feb 2015 09:35:32 +0100Tabu search with guided restarts for a car production problem with a 2/3 balancing penalty : Proceedings of 5th International Conference on Metaheuristics and Nature Inspired Computinghttps://archive-ouverte.unige.ch/unige:45332https://archive-ouverte.unige.ch/unige:45332abstract not availableThu, 15 Jan 2015 14:31:15 +0100Ant Algorithms for a Truck Loading Problem with Multiple Destinations : Proceedings of the 14th International Workshop on Project Management and Schedulinghttps://archive-ouverte.unige.ch/unige:45328https://archive-ouverte.unige.ch/unige:45328abstract not availableThu, 15 Jan 2015 14:29:05 +0100Online Vehicle Routing and Scheduling with Continuous Vehicle Tracking : Proceedings of the 15th Annual Congress of the French Operations Research Societyhttps://archive-ouverte.unige.ch/unige:45326https://archive-ouverte.unige.ch/unige:45326abstract not availableThu, 15 Jan 2015 14:28:36 +0100A deconstruction-reconstruction metaheuristic for a job scheduling problem : Proceedings of 5th International Conference on Metaheuristics and Nature Inspired Computinghttps://archive-ouverte.unige.ch/unige:45324https://archive-ouverte.unige.ch/unige:45324abstract not availableThu, 15 Jan 2015 14:27:51 +0100Variable Neighborhood Search for a Scheduling Problem with Time Window Penalties : Proceedings of the 14th International Workshop on Project Management and Schedulinghttps://archive-ouverte.unige.ch/unige:45322https://archive-ouverte.unige.ch/unige:45322abstract not availableThu, 15 Jan 2015 14:27:13 +0100Multi-objectives parallel machines scheduling with incompatible jobs : Proceedings of the 15th Annual Congress of the French Operations Research Societyhttps://archive-ouverte.unige.ch/unige:45321https://archive-ouverte.unige.ch/unige:45321abstract not availableThu, 15 Jan 2015 14:26:33 +0100Tabu Search with Diversity Control and Simulation for an Inventory Management Problem : Proceedings of 5th International Conference on Metaheuristics and Nature Inspired Computinghttps://archive-ouverte.unige.ch/unige:45320https://archive-ouverte.unige.ch/unige:45320abstract not availableThu, 15 Jan 2015 14:25:30 +0100Design and Classification of Ant Metaheuristics : Proceedings of the 22nd Euromicro International Conference on Parallel, Distributed and Network-Based Processinghttps://archive-ouverte.unige.ch/unige:45319https://archive-ouverte.unige.ch/unige:45319abstract not availableThu, 15 Jan 2015 14:25:00 +0100Graph Coloring and Job Scheduling: from Models to Powerful Tabu Search Solution Methods : Proceedings of the 14th International Workshop on Project Management and Schedulinghttps://archive-ouverte.unige.ch/unige:45318https://archive-ouverte.unige.ch/unige:45318abstract not availableThu, 15 Jan 2015 14:24:30 +0100Ant Decision Systems for Combinatorial Optimization with Binary Constraintshttps://archive-ouverte.unige.ch/unige:40152https://archive-ouverte.unige.ch/unige:40152abstract not availableWed, 10 Sep 2014 14:36:18 +0200Tabu Search for a Single Machine Scheduling Problem with Discretely Controllable Release Dateshttps://archive-ouverte.unige.ch/unige:40151https://archive-ouverte.unige.ch/unige:40151abstract not availableWed, 10 Sep 2014 14:35:46 +0200Tabu Search for a Preemptive Scheduling Problem with Job Incompatibilitieshttps://archive-ouverte.unige.ch/unige:40150https://archive-ouverte.unige.ch/unige:40150abstract not availableWed, 10 Sep 2014 14:35:08 +0200Multi-coloring and job-scheduling with assignment and incompatibility costshttps://archive-ouverte.unige.ch/unige:37815https://archive-ouverte.unige.ch/unige:37815abstract not availableSun, 15 Jun 2014 19:03:23 +0200Tabu search for a car sequencing problemhttps://archive-ouverte.unige.ch/unige:26492https://archive-ouverte.unige.ch/unige:26492abstract not availableMon, 25 Feb 2013 16:02:47 +0100Nonlinear global optimization relevant to discrete choice models estimationhttps://archive-ouverte.unige.ch/unige:26491https://archive-ouverte.unige.ch/unige:26491abstract not availableMon, 25 Feb 2013 16:02:17 +0100A heuristic for nonlinear global optimization relevant to discrete choice models estimationhttps://archive-ouverte.unige.ch/unige:26489https://archive-ouverte.unige.ch/unige:26489abstract not availableMon, 25 Feb 2013 16:01:45 +0100Tabu Search for a Project Scheduling Problem with Incompatibility and Assignment Costshttps://archive-ouverte.unige.ch/unige:26488https://archive-ouverte.unige.ch/unige:26488abstract not availableMon, 25 Feb 2013 16:01:00 +0100A Local Search for Refueling Locomotiveshttps://archive-ouverte.unige.ch/unige:26487https://archive-ouverte.unige.ch/unige:26487abstract not availableMon, 25 Feb 2013 16:00:10 +0100A Meta-Heuristic Approach to Planning Intermodal Transportation of Hazardous Materialshttps://archive-ouverte.unige.ch/unige:26486https://archive-ouverte.unige.ch/unige:26486abstract not availableMon, 25 Feb 2013 15:59:44 +0100A Population Exposure Approach to Planning Intermodal Transportation of Hazardous Materialshttps://archive-ouverte.unige.ch/unige:26485https://archive-ouverte.unige.ch/unige:26485abstract not availableMon, 25 Feb 2013 15:59:15 +0100A Reconstructive Evolutionary Metaheuristic for the Vertex Coloring Problemhttps://archive-ouverte.unige.ch/unige:26484https://archive-ouverte.unige.ch/unige:26484abstract not availableMon, 25 Feb 2013 15:58:50 +0100Local Search Techniques for a Job Scheduling Problem with Overlapping Costs when Preemptions are Allowedhttps://archive-ouverte.unige.ch/unige:26483https://archive-ouverte.unige.ch/unige:26483abstract not availableMon, 25 Feb 2013 15:58:20 +0100Heuristics for a Multi-Machine Multi-Objective Job Scheduling Problem with Smoothing Costshttps://archive-ouverte.unige.ch/unige:26482https://archive-ouverte.unige.ch/unige:26482abstract not availableMon, 25 Feb 2013 15:57:43 +0100Ant Local Search for Fuel Supply of Trains in Americahttps://archive-ouverte.unige.ch/unige:26481https://archive-ouverte.unige.ch/unige:26481abstract not availableMon, 25 Feb 2013 15:57:17 +0100Tabu Search to Minimize Regular Objective Functions for a Single Machine Scheduling Problem with Rejected Jobs, Setups and Time Windowshttps://archive-ouverte.unige.ch/unige:26480https://archive-ouverte.unige.ch/unige:26480abstract not availableMon, 25 Feb 2013 15:56:52 +0100Consistent Neighborhood Search for Constrained Assignment Problemshttps://archive-ouverte.unige.ch/unige:26479https://archive-ouverte.unige.ch/unige:26479abstract not availableMon, 25 Feb 2013 15:56:25 +0100Tabu Search using Variable Amplitudes for Dimensioning an Assembly/Disassembly Production Systemhttps://archive-ouverte.unige.ch/unige:26477https://archive-ouverte.unige.ch/unige:26477abstract not availableMon, 25 Feb 2013 15:55:14 +0100Dynamic Tabu Search with Simulation for a Resource Allocation Problem within a Production Environmenthttps://archive-ouverte.unige.ch/unige:26476https://archive-ouverte.unige.ch/unige:26476abstract not availableMon, 25 Feb 2013 15:54:43 +0100A Renault Truck Loading Problem: from Benchmarking to Improvementshttps://archive-ouverte.unige.ch/unige:26475https://archive-ouverte.unige.ch/unige:26475abstract not availableMon, 25 Feb 2013 15:54:13 +0100A Meta-Heuristic Approach to Rail-Truck Intermodal Transportation of Hazardous Materialshttps://archive-ouverte.unige.ch/unige:26474https://archive-ouverte.unige.ch/unige:26474abstract not availableMon, 25 Feb 2013 15:53:26 +0100Successful Elements for Developing Job-Shop Scheduling Metaheuristicshttps://archive-ouverte.unige.ch/unige:26473https://archive-ouverte.unige.ch/unige:26473abstract not availableMon, 25 Feb 2013 15:52:58 +0100How to Build and Evaluate a Solution Method: an Illustration for the Vehicle Routing Problemhttps://archive-ouverte.unige.ch/unige:26472https://archive-ouverte.unige.ch/unige:26472abstract not availableMon, 25 Feb 2013 15:52:24 +0100A New Ant Algorithm for Graph Coloringhttps://archive-ouverte.unige.ch/unige:26180https://archive-ouverte.unige.ch/unige:26180Let G = (V;E) be a graph with vertex set V and edge set E. The k-coloring problem is to assign a color (a number chosen in {1, ..., k}) to each vertex of V so that no edge has both endpoints with the same color. We describe in this paper a new ant algorithm for the k-coloring problem. Computational experiments give evidence that our algorithm is competitive with the existing ant algorithms for this problem, while giving a minor role to each ant. Our algorithm is however not competitive with the best known coloring algorithms.Mon, 04 Feb 2013 15:49:35 +0100Adaptive Memory Algorithms for Graph Coloringhttps://archive-ouverte.unige.ch/unige:26179https://archive-ouverte.unige.ch/unige:26179Let G=(V,E) be a graph with vertex set V and edge set E. The graph coloring problem consists in assigning an integer (called color) to each vertex so that adjacent vertices get different colors, and the total number of different colors is minimized. The adaptive memory algorithm is an hybrid evolutionary heuristic based on a central memory. At each generation, the idea is to use the information of a central memory for producing an offspring which is then improved by using a local search algorithm. The so obtained solution is finally used to update the information contained in the memory. In this paper, we propose an adaptive memory algorithm to the graph coloring problem. The results obtained so far give evidence that our method is competitive with the best known coloring algorithms.Mon, 04 Feb 2013 15:49:02 +0100A variable neighborhood search for graph coloringhttps://archive-ouverte.unige.ch/unige:26178https://archive-ouverte.unige.ch/unige:26178Descent methods for combinatorial optimization proceed by performing a sequence of local changes on an initial solution which improve each time the value of an objective function until a local optimum is found. Several meta-heuristics have been proposed which extend in various ways this scheme and avoid being trapped in local optima. For example, Hansen and Mladenovic have recently proposed the variable neighborhood search method which has not yet been applied to many combinatorial optimization problems. The aim of this paper is to propose an adaptation of this new method to the graph coloring problem.Mon, 04 Feb 2013 15:47:53 +0100Lower bounding and tabu search procedures for the frequency assignment problem with polarization constraintshttps://archive-ouverte.unige.ch/unige:26177https://archive-ouverte.unige.ch/unige:26177The problem retained for the ROADEF'2001 international challenge was a Frequency Assignment Problem with polarization constraints (FAPP). This NP-hard problem was proposed by the CELAR of the French Department of Defense, within the context of the CALMA project. Twenty seven competitors took part to this contest, and we present in this paper the contribution of our team that allowed us to be selected as one of the six finalists qualified for the final round of the competition. There is typically no solution satisfying all constraints of the FAPP. For this reason, some electromagnetic compatibility constraints can be progressively relaxed, and the objective is to find a feasible solution with the lowest possible level of relaxation. We have developed a procedure that computes a lower bound on the best possible level of relaxation, as well as two tabu search algorithms for the FAPP, one for the frequency assignment, and one for the polarization assignment.Mon, 04 Feb 2013 15:47:26 +0100Inventory control of raw materials under stochastic and seasonal lead timeshttps://archive-ouverte.unige.ch/unige:26175https://archive-ouverte.unige.ch/unige:26175Most inventory modelling has assumed stochastic demands and constant lead times. However, here we consider a problem for which the opposite situation holds; namely, there is a known constant demand rate, but lead times are random variables. Moreover, the probability distributions of the lead times change in a seasonal fashion. Also, shortages of raw materials result in lost sales. The goal of this paper is to propose heuristic methods for minimizing the expected costs in such a situation. This study was motivated by a problem of management of raw material at a sawmill.Mon, 04 Feb 2013 15:46:40 +0100An Ant Algorithm for the Steiner Tree Problem in Graphshttps://archive-ouverte.unige.ch/unige:26174https://archive-ouverte.unige.ch/unige:26174The Steiner Tree Problem (STP) in graphs is a well-known NP-hard problem. It has regained attention due to the introduction of new telecommunication technologies, since it is the mathematical structure behind multi-cast communications. The goal of this paper is to design an ant algorithm (called ANT-STP) for the STP in graphs which is better than TM, which is a greedy constructive method for the STP proposed in [34]. We derive ANT-STP from TM as follows: each ant is a constructive heuristic close to TM, but the population of ants can collaborate by exchanging information by the use of the trail systems. Inm addition, the decision rule used by each individual ant is different from the decision rule used in TM. We compare TM and ANT-STP on a set of benchmark problems of the OR-Library.Mon, 04 Feb 2013 15:46:04 +0100A graph coloring heuristic using partial solutions and a reactive tabu schemehttps://archive-ouverte.unige.ch/unige:26173https://archive-ouverte.unige.ch/unige:26173Most of the recent heuristics for the graph coloring problem start from an infeasible k-coloring (adjacent vertices may have the same color) and try to make the solution feasible through a sequence of color exchanges. In contrast, our approach (called FOOPARTIALCOL), which is based on tabu search, considers feasible but partial solutions and tries to increase the size of the current partial solution. A solution consists of k disjoint stable sets (and, therefore, is a feasible, partial k-coloring) and a set of uncolored vertices. We introduce a reactive tabu tenure which substantially enhances the performance of both our heuristic as well as the classical tabu algorithm (called TABUCOL) proposed by Hertz and de Werra [Using tabu search techniques for graph coloring, Computing 1987;39:345–51].We will report numerical results on different benchmark graphs and we will observe that FOO-PARTIALCOL, though very simple, outperforms TABUCOL on some instances, provides very competitive results on a set of benchmark graphs which are known to be difficult, and outperforms the best-known methods on the graph flat300_28_0. For this graph, FOO-PARTIALCOL finds an optimal coloring with 28 colors. The best coloring achieved to date uses 31 colors. Algorithms very close to TABUCOL are still used as intensification procedures in the best coloring methods, which are evolutionary heuristics. FOO-PARTIALCOL could then be a powerful alternative. In conclusion FOO-PARTIALCOL is one of the most efficient simple local search coloring methods yet available.Mon, 04 Feb 2013 15:45:34 +0100An adaptive memory algorithm for the k-coloring problemhttps://archive-ouverte.unige.ch/unige:26172https://archive-ouverte.unige.ch/unige:26172Let G = (V ,E) be a graph with vertex set V and edge set E. The k-coloring problem is to assign a color (a number chosen in {1, ..., k}) to each vertex of G so that no edge has both endpoints with the same color. The adaptive memory algorithm is a hybrid evolutionary heuristic that uses a central memory. At each iteration, the information contained in the central memory is used for producing an offspring solution which is then possibly improved using a local search algorithm. The so obtained solution is finally used to update the central memory. We describe in this paper an adaptive memory algorithm for the k-coloring problem. Computational experiments give evidence that this new algorithm is competitive with, and simpler and more flexible than, the best known graph coloring algorithms.Mon, 04 Feb 2013 15:45:02 +0100Variable space search for graph coloringhttps://archive-ouverte.unige.ch/unige:26171https://archive-ouverte.unige.ch/unige:26171Let G = (V, E) be a graph with vertex set V and edge set E. The k-coloring problem is to assign a color (a number chosen in {1, ..., k}) to each vertex of G so that no edge has both endpoints with the same color. We propose a new local search methodology, called Variable Space Search, which we apply to the k-coloring problem. The main idea is to consider several search spaces, with various neighborhoods and objective functions, and to move from one to another when the search is blocked at a local optimum in a given search space. The k-coloring problem is thus solved by combining different formulations of the problem which are not equivalent, in the sense that some constraints are possibly relaxed in one search space and always satisfied in another. We show that the proposed algorithm improves on every local search used independently (i.e., with a unique search space), and is competitive with the currently best coloring methods, which are complex hybrid evolutionary algorithms.Mon, 04 Feb 2013 15:44:34 +0100Three tabu search methods for the MI-FAP applied to 802.11 networkshttps://archive-ouverte.unige.ch/unige:26170https://archive-ouverte.unige.ch/unige:26170Wireless LAN using IEEE 802.11 networks are now widely deployed at home by residential users or in hot spots by telecommunication operators. A hot spot is a place where a set of access points (APs) are located nearby each other and can serve many users. Since perturbations can degrade the quality of the signal, a careful channel assignment to each AP has to be done. Channel assignment of APs at hot spots, and more generally setup configuration and management, is still often done manually. In this paper, we consider a modeling that enables optimization of channel assignment with respect to the dynamic behavior of end-users. We prove our problem's formulation to correspond to the Minimum Interference Frequency Assignment Problem, and hence the problem to be NP-hard. We propose and compare three different tabu search methods to solve the problem of channel assignment in 802.11 WLAN networks. The first one, called TabuObj, tackles the problem using directly the objective function associated with the model. The second one, called TabuApproxObj, uses a simplified and approximate objective function in order to visit more solutions during the same amount of time, i.e. to be quicker than TabuObj. The third one, called TabuLevel, is even more quicker and is based on the following philosophy: under time constraints, it could be judicious to explore very quickly lots of solutions, rather than spending much computation time for the evaluation of each solution, and hence only considering a few solutions. Those three methods are then compared based on time constraints and on the quality of their solutions.Mon, 04 Feb 2013 15:43:57 +0100A solution method for a car fleet management problem with maintenance constraintshttps://archive-ouverte.unige.ch/unige:26168https://archive-ouverte.unige.ch/unige:26168The problem retained for the ROADEF’99 international challenge was an inventory management problem for a car rental company. It consists in managing a given fleet of cars in order to satisfy requests from customers asking for some type of cars for a given time period. When requests exceed the stock of available cars, the company can either offer better cars than those requested, subcontract some requests to other providers, or buy new cars to enlarge the available stock. Moreover, the cars have to go through a maintenance process at a regular basis, and there is a limited number of workers that are available to perform these maintenances. The problem of satisfying all customer requests at minimum cost is known to be NP-hard. We propose a solution technique that combines two tabu search procedures with algorithms for the shortest path, the graph coloring and the maximum weighted independent set problems. Tests on benchmark instances used for the ROADEF’99 challenge give evidence that the proposed algorithm outperforms all other existing methods (thirteen competitors took part to this contest).Mon, 04 Feb 2013 15:42:32 +0100