The thesis focuses on optimization of inbound and outbound truck scheduling with thegoal of minimizing total operation time of cross docking. A model of cross docking isdeveloped; two different methods are applied on the model in order to find an optimaldocking sequence for receiving and shipping trucks and their assignment to receiving andshipping docks, and product routing from receiving to shipping trucks.The two methods used were mathematical modeling and heuristic algorithm. For the firstmethod, a mixed integer programming model was developed to minimize total operationtime; AMPL modeling language is used for the mathematical modeling for small sizedproblems. For the second method, a heuristic algorithm was developed to find nearoptimal solutions fast and was used for problems of larger size. In order to examine theperformance of heuristic algorithm, small problems were solved by both mathematicalmodel and the heuristic algorithm.The results from the mathematical model and the heuristic algorithm are very close withslight differences in receiving and shipping truck docking sequence, and in productrouting between these two methods. In addition, the heuristic algorithm also calculatesnumber of products transferring from receiving trucks to the temporary storage as well asthe number of products transferring from the temporary storage to shipping truck incontrary to the mathematical model. Total number of units of products passing throughthe temporary storage calculated by heuristic algorithm is presented and it can be seenthat the heuristic algorithm transfers to the temporary storage as few products as possible.Furthermore, in cases that receiving and shipping trucks are divided into groups orclusters in the cross docking operation, heuristic algorithm can be used to calculateoptimal number of receiving and shipping docks based on preferences of total operationtime or total number of products passing through the temporary storage.Another issue which is focused on is the problem of dock door assignment. Closeshipping docks to each receiving dock are determined and the percentage of productstransferred from a receiving dock to its close shipping docks is calculated as a method tomeasure the performance of the dock assignment solution.

This thesis presents an approach to modeling Multi Agent Systems (MAS). A framework and its implementation are presented as an extension to .NET. A number of definitions of agents are evaluated for the purpose of a broad understanding of the term software agent. Software agent has been defined in MAS context and its characteristics are identified and implemented. Motivation factors for building framework for MAS have been discussed. A number of existing technologies are discussed and evaluated. A number of agent systems previously developed are also being discussed in the middle part of the thesis. A model software agent has been defined and its characteristics are divided in two basic categories essential and optional. Its implementation has been distributed into different components throughout the MAS framework. Some of these characteristics are jointly implemented by a number of components and others responsibility rest on the individual components. Detail working of the MAS framework (i.e. what to do, when to do) is explained as guide to develop MAS using MAS framework. The protocols followed by the framework components to make communication possible between them are discussed at components level. The required information for developing MAS using MAS framework are also discussed. It answers the why, when and how questions in regards to using MAS framework A case study on Dynamic Truck Scheduling (DTS) system is discussed, designed and implemented using the MAS framework. DTS System has been used as a prototype application to test and evaluate the framework. DTS also represents a model problem that can be answered by using MAS; complete in-depth details about the problem statement are discussed. It also discusses the design and implementation of the solution along with the test results of the framework. Possible future expansion is presented in light of a number of limitations known of the MAS framework. The code working behind the different components of the MAS framework is given in appendices. Some important standards of XML that are used to pass information between agents and MAS framework components are also given in the format of tables.

O problema de programaÃÃo de caminhÃes à um tema de grande relevÃncia na gestÃo de frota das empresas. Estas enfrentam dificuldades em gerenciar seus veÃculos devido Ãs diversas variÃveis inerentes ao processo, tais como o tamanho ideal da frota, os diversos tipos de caminhÃes disponÃveis, a capacidade de carga do caminhÃo, as informaÃÃes tÃcnicas do cliente e o agendamento das viagens. No que se refere ao transporte de derivados de petrÃleo, existem diversas caracterÃsticas operacionais que dificultam a programaÃÃo de caminhÃes. Embora a literatura sobre a programaÃÃo de veÃculos seja vasta, as abordagens para a programaÃÃo de caminhÃes para este tipo de transporte ainda à limitada. O presente trabalho tem como objetivo desenvolver um modelo de programaÃÃo inteira para a otimizaÃÃo da programaÃÃo de veÃculos de mÃltiplos tipos para o transporte de derivados de petrÃleo para obras de construÃÃo de rodovias. Dados reais sobre uma empresa de transporte desse setor foram coletados. Foi desenvolvido um modelo que buscasse a minimizaÃÃo da frota de caminhÃes disponÃveis. Como conclusÃes, pode-se ressaltar que a metodologia empregada serviu para minimizar a frota necessÃria no perÃodo analisado. O desenvolvimento de indicadores de desempenho permitiu avaliar a qualidade das soluÃÃes geradas. / The truck scheduling problem is an important topic in the companiesâ fleet management. Many companies face difficulties to manage their vehicles due to several variables inherent to the management process, such as the optimal fleet size, multiple types of trucks available, trucks capacity, the technical information from the clientâs construction site and trips scheduling. In terms of the transportation of oil products, there are many operational characteristics that make the truck scheduling difficult. Despite the extensive literature about vehicle scheduling, truck scheduling for this type of transport is limited. The purpose of this research study was to develop an Integer Programming model to optimize the multi-type truck scheduling for the transportation of oil products for road construction sites. Data from a real company were gathered. A model that aimed to minimize the truck fleet available was developed. The method used has minimized the available fleet in the period under review. The development of key performance indicators allows to evaluate the quality of the solutions created.

<p>The transportation cost of logs from forest to customers is a large part of the overall cost for the Swedish forestry industry. Finding good routes from harvesting points to saw and pulp mills is a complex task, where the total number of feasible routes is extremely high. In this thesis we present two methods for log truck scheduling. </p><p>The first is to, from a given set of routes, find the most valuable subset that fulfils the customers demand. We use a model that is similar to the set partitioning problem and a method that is referred to as a composite pricing coupled with Branch and Bound. The composite pricing based method prices the routes (columns) and chooses the most valuable ones that are then added to the LP relaxation. Once an LP optimum is found, the Branch and Bound method is used to find an integer optimum solution. We have tested this on a case of realistic size. </p><p>The second method is a tabu search heuristic. Here, the purpose is to create efficient and qualitative routes from a given number of trips (referred to as predefined trips). From a start solution tabu search systematically generates new solutions. This method was tested on a small problem and on a five times larger problem to study how the size of the problem affected the result. It was also tested and compared on two cases in which the backhauling possibilities (i.e. instead of traveling empty the truck picks up another load on the return trip) had and had not been studied. The composite pricing based method and the tabu search method proved to be very useful for this kind of scheduling.</p>

The transportation cost of logs from forest to customers is a large part of the overall cost for the Swedish forestry industry. Finding good routes from harvesting points to saw and pulp mills is a complex task, where the total number of feasible routes is extremely high. In this thesis we present two methods for log truck scheduling. The first is to, from a given set of routes, find the most valuable subset that fulfils the customers demand. We use a model that is similar to the set partitioning problem and a method that is referred to as a composite pricing coupled with Branch and Bound. The composite pricing based method prices the routes (columns) and chooses the most valuable ones that are then added to the LP relaxation. Once an LP optimum is found, the Branch and Bound method is used to find an integer optimum solution. We have tested this on a case of realistic size. The second method is a tabu search heuristic. Here, the purpose is to create efficient and qualitative routes from a given number of trips (referred to as predefined trips). From a start solution tabu search systematically generates new solutions. This method was tested on a small problem and on a five times larger problem to study how the size of the problem affected the result. It was also tested and compared on two cases in which the backhauling possibilities (i.e. instead of traveling empty the truck picks up another load on the return trip) had and had not been studied. The composite pricing based method and the tabu search method proved to be very useful for this kind of scheduling.

Dans une plateforme de cross-docking, les produits sont déchargés descamions entrants, triés puis directement rechargés dans les camions sortants– chaque produit passe moins de 24 heures sur la plateforme.L’analyse des écarts entre la littérature et les observations réalisées sur leterrain permet de dégager deux axes de recherche : la prise en compte desincertitudes opérationnelles d’une part, et de la capacité des ressourceshumaines de la plateforme d’autre part.Le problème de planification des camions entrants et sortants avec fenêtrede temps est modélisé par un programme linéaire et résolu par troisheuristiques différentes. La robustesse des plannings obtenus est ensuitetestée à l’aide d’un modèle de simulation à événements discrets, qui permetd’évaluer plusieurs reformulations robustes du modèle initial.Le problème de planification des employés sur la plateforme est traité àl’aide de trois programmes linéaires mixtes, résolus de façon séquentielle.La combinaison des deux modèles permet d’obtenir un modèle d’aide àla décision pour une plateforme de cross-docking. / In a cross-docking platform, goods are unloaded from inbound trucks,sorted and directly reloaded in outbound trucks – each product typicallystays less than 24 hours in the platform.By analyzing the gaps between the literature and on-field observations,we highlight two research directions: accounting for operational uncertainties,and for the human resource capacity in the platform.A truck scheduling problem with time windows for the inbound andoutbound trucks is modeled with an integer program and solved withthree different heuristics. The robustness of the schedules obtained is thentested with a discrete-event simulation model, which enables to evaluateseveral robust reformulations of the initial model.The employee timetabling and rostering problem in the platform is addressedwith three mixed integer linear problems solved sequentially. Thetwo models can be combined to serve as a decision-support tool for across-docking platform.

Afin d’accélérer les flux de produits, de réduire les niveaux de stocks et de faire des économies de transport, les entreprises de presque toutes les industries ont mis en place des centres de crossdock. Ces centres sont un point intermédiaire de consolidation dans une chaîne logistique. Les constructeurs automobiles Renault et Nissan s’appuient sur un réseau international de plateformes crossdock pour lier des fournisseurs de pièces de première monte avec des usines de production lointaines, généralement en outre-mer. Dans un cadre d’un partenariat académique-industriel entre le laboratoire LIMOS et Renault, cette thèse est focalisée sur la planification et l’ordonnancement des activités dans ces centres de crossdock. Des études de terrain menées chez Renault et Nissan nous ont permis d’identifier les caractéristiques, les contraintes et les inducteurs de coûts des plateformes de crossdock, ainsi que de cibler notre revue de la littérature. Sur ces bases, nous proposons une approche d’optimisation séquentielle, comprenant deux modèles en programmation linéaire en nombres entiers, implémentés dans CPLEX et testés sur des données industrielles de deux plateformes Renault. Les résultats des expérimentations obtenus sur le premier modèle (planification) ont montré une nette amélioration en termes de coûts, par rapport à la méthode Renault. Fort de ce constat, une implémentation industrielle a été faite, avec des résultats aussi probants. Le deuxième modèle (ordonnancement) s’avère pertinent pour des instances de moyenne taille. L’approche proposée permet de répondre à la configuration actuelle des AILN Renault et nous considérons qu’elle est adaptable à d’autres industries. / In order to accelerate product flow, reduce inventory levels and make economies in transportation, companies in almost all industries have set up cross-dock centres. These centres are an intermediate point of consolidation in a supply chain. Car manufacturers Renault and Nissan rely on an international network of crossdock platforms to link suppliers of OEM parts with overseas production plants. In a framework of an academic-industrial partnership between the LIMOS laboratory and Renault, this thesis focuses on the activity planning and scheduling at these crossdock centres.Field studies conducted at Renault and Nissan allowed us to identify the characteristics, constraints and cost drivers of crossdock platforms, as well as to target our review of the literature. Based on this, we propose a sequential optimization approach, comprising two integer linear programming models, implemented in CPLEX and tested on industrial data of two Renault platforms. Numerical experiments’ results obtained on the first model (planning) showed a significant improvement in cost, compared to the Renault method. In light of this results, an industrial implementation was made, with such convincing results. The second model (scheduling) proved to be relevant for medium-sized instances. The proposed approach fits to the current configuration of AILN Renault and we consider that it is adaptable to other industries.