Die schichtenübergreifende Steuerung von Multi-Layer Netzen ermöglicht die Realisierung fortgeschrittener Netzarchitekturen sowie neuartiger Konzepte zur Steigerung der Ausfallsicherheit. Gegenstand dieser Arbeit ist ein neues ressourcensparendes Konzept zur Kompensation von Core-Router-Ausfallen in IP-Netzen. Core-Router-Ausfälle führen zur Abkopplung der an Ihnen angeschlossenen Zugangsrouter vom Netz. Daher werden die Zugangsrouter üblicherweise mit jeweils zwei oder mehreren verschiedenen Core-Routern verbunden (engl.: dual homing) was jedoch eine Verdoppelung der Anschlusskapazität im IP Netz bedingt. Bei dem neuen Verfahren - Dual Homing mit gemeinsam genutzten Router-Ersatzressourcen (engl.: dual homing with shared backup router resources, DH-SBRR) - erfolgt die Zugangsrouter-Anbindung zum einen zu einem Core-Router des IP-Netzes und zum anderen zu einem Netzelement der darunterliegenden Transportschicht. Damit lassen sich Router-Ersatzressourcen, die im IP-Netz an beliebigen Stellen vorgehalten werden können, uber das Transportnetz an die Stelle eines ausgefallenen Core-Routers schalten. Die Steuerung dieser Ersatzschaltung geschieht über eine schichten übergreifende, d.h. das Transportnetz- und IP-Netz umfassende Control-Plane - beispielsweise auf Basis von GMPLS. Da beim Umschalten der Routerressourcen auch aktuelle Zustände (bspw. Routing-Tabellen) auf die Router-Ersatzressourcen mit übertragen werden müssen, beinhaltet das neue Verfahren auch Konzepte zur Router-Virtualisierung. Zum Vergleich und zur Bewertung der Leistungsfähigkeit des neuen DH-SBRR Verfahrens werden in der Arbeit verschiedene Zugangsrouter-Homing-Varianten hinsichtlich Netz-Kosten, Netz-Verfügbarkeit, Recovery-Zeit und Netz-Energieverbrauch gegenübergestellt. Als Multi-Layer Netzszenarien werden zum einen IP über WDM und zum anderen IP über OTN (ODU) betrachtet. Zur Bestimmung der minimalen Netz-Kosten ist ein generisches Multi-Layer Netzoptimierungsmodell entwickelt worden, welches bei unterschiedlichen Homing-Architekturen angewendet werden kann. Neben dem Optimierungsmodell zur Netzkostenminimierung wird auch eine Modellvariante zur Minimierung des Energieverbrauchs vorgestellt. Um die Rechenzeit für die Lösung der Optimierungsprobleme zu verringern und damit auch größere Netzszenarien untersuchen zu können bedarf es heuristischer Lösungsverfahren. Im Rahmen der Arbeit ist daher eine neue speziell auf die Multilayer-Optimierungsprobleme zugeschnittene Lösungsheuristik entwickelt worden. Aus der Netzkosten-Optimierung ergibt sich, dass durch den Einsatz von DH-SBBR signifikante Kosteneinsparungen im Vergleich zu herkömmlichen Homing-Architekturen realisiert werden können. Änderungen der Verkehrslast, der Kosten der IP-Netzelemente oder der Netztopologie haben keinen signifikanten Einfluss auf dieses Ergebnis. Neben dem Kosten- und Energieeinsparungspotential sind auch die Auswirkungen auf die Netz-Verfügbarkeit und die Recovery-Zeit untersucht worden. Für die Ende-zu-Ende Verfügbarkeit bei Anwendung der verschiedenen Homing-Architekturen Können untere Grenzwerte angegeben werden. Zur Bestimmung der Recovery-Zeit bei Einsatz von DH-SBRR ist ein eigenes analytisches Berechnungsmodell entwickelt und evaluiert worden. Damit kann das DH-SBRR Verfahren zur Einhaltung vorgegebener Recovery-Zeiten (wie sie für bspw. Für bestimmte Dienste gefordert werden) entsprechend parametriert werden. / The emergence of multi-layer networking capabilities opens the path for the development of advanced network architectures and resilience concepts. In this dissertation we propose a novel resource-efficient homing scheme: dual homing with shared backup router resources. The proposed scheme realizes shared router-level redundancy, enabled by the emergence of control plane architectures such as generalized multi-protocol label switching. Additionally, virtualization schemes complement the proposed architecture. Different homing architectures are examined and compared under the prism of cost, availability, recovery time and energy efficiency. Multiple network layers are considered in Internet protocol over wavelength division multiplexing as well as Internet protocol over optical data unit settings - leading to the development of multi-layer optimization techniques. A generic multi-layer network design mathematical model, which can be applied to different homing architecture considerations, is developed. The optimization objective can be adapted to either minimizing the cost for network equipment or the power consumption of the network. In order to address potential issues with regard to computational complexity, we develop a novel heuristic approach specifically targeting the proposed architecture. It is shown that significant cost savings can be achieved - even under extreme changes in the traffic demand volume, in the cost for different types of network equipment, as well as in the network topology characteristics. In order to evaluate occurring tradeoffs in terms of performance, we study the effects on availability and recovery time. We proceed to derive lower bounds on end-to-end availability for the different homing architectures. Additionally, an analytical recovery time model is developed and evaluated. We investigate how service-imposed maximum outage requirements have a direct effect on the setting of the proposed architecture.

Orientador: Rubén Augusto Romero Lázaro / Resumo: Neste trabalho, consideram-se a análise teórica e a implementação computacional dos algoritmos Primal Simplex Canalizado (PSC) e Dual Simplex Canalizado (DSC) especializados. Esses algoritmos foram incorporados em um algoritmo Branch and Bound (B&B) de modo a resolver o problema de Planejamento da Expansão de Sistemas de Transmissão (PEST). Neste caso, o problema PEST foi modelado usando os chamados modelo de Transportes e modelo Linear Disjuntivo (LD), o que produz um problema de Programação Linear Inteiro Misto (PLIM). O algoritmo PSC é utilizado na resolução do problema de Programação Linear (PL) inicial após desconsiderar a restrição de integralidade do problema PLIM original. Juntamente com o algoritmo PSC, foi implementada uma estratégia para reduzir o número de variáveis artificiais adicionadas ao PL, consequentemente reduzindo o número de iterações do algoritmo PSC. O algoritmo DSC é utilizado na reotimização eficiente dos subproblemas gerados pelo algoritmo B&B, através do quadro ótimo do PL inicial, excluindo, assim, a necessidade da resolução completa de cada subproblema e, consequentemente, reduzindo o consumo de processamento e memória. Nesta pesquisa, é apresentada uma nova proposta de otimização, e, consequentemente, a implementação computacional usando a linguagem de programação FORTRAN que opera independentemente de qualquer solver. / Doutor

Le problème d’approvisionnement conjoint (JDP) proposé est un problème de planification des tournées de livraisons sur un horizon de temps décomposé en périodes élémentaires, l’horizon de temps étant la période commune de livraison de tous les produits,. La donnée de ces paramètres permet d’obtenir une formulation linéaire du problème, avec des variables de décision binaires. Le modèle intègre aussi des contraintes de satisfaction de la demande à partir des stocks et des quantités livrées, des contraintes sur les capacités de stockage et de transport.Afin de résoudre aussi le problème de choix des tournées de livraison, il est nécessaire d'introduire dans le modèle des contraintes et des variables liées aux sites visités au cours de chaque tour. Il est proposé de résoudre le problème en deux étapes. La première étape est le calcul hors ligne du coût minimal de la tournée associé à chaque sous-ensemble de sites. On peut observer que pour tout sous-ensemble donné de sites, le cycle hamiltonien optimal reliant ces sites à l'entrepôt peut être calculé à l'avance par un algorithme du problème du voyageur de commerce (TSP). Le but ici n'est pas d'analyser pleinement le TSP, mais plutôt d'intégrer sa solution dans la formulation de JRP. .Dans la deuxième étape, des variables binaires sont associées à chaque tour et à chaque période pour déterminer le sous-ensemble de sites choisi à chaque période et son coût fixe associé. / The proposed joint delivery problem (JDP) is a delivery tour planning problem on a time horizon decomposed into elementary periods or rounds, the time horizon being the common delivery period for all products. The data of these parameters provides a linear formulation of the problem, with binary decision variables. The model also incorporates the constraints of meeting demand from stock and the quantities supplied, storage and transport capacity constraints.In order to also solve the problem of choice of delivery rounds, it is necessary to introduce in the model several constraints and variables related to the sites visited during each round. It is proposed to solve the problem in two steps. The first step is the calculation of the minimum off-line cost of the tour associated with each subset of sites. One can observe that for any given subset of sites, the optimal Hamiltonian cycle linking those sites to the warehouse can be calculated in advance by a traveling salesman problem algorithm (TSP). The goal here is not to fully analyze the TSP, but rather to integrate its solution in the formulation of the JRP. In the second stage, binary variables are associated with each subset and each period to determine the selected subset of sites in each period and its associated fixed cost.

[pt] Portfólios de fontes renováveis de energia elétrica são mecanismos de gerenciamento de risco interessantes para comercialização de energia em mercados de negociação bilateral. Quando formados por agentes que pertencem a diferentes companhias sua estabilidade depende da maneira com que os benefícios de mitigação de risco gerados pelo portfólio são alocados individualmente entre os participantes. O problema de se encontrar uma solução estável pode ser matematicamente formulado através da busca de um vetor de alocação de quotas que pertença ao núcleo do jogo cooperativo, que por sua vez pode ser formulado como um conjunto de restrições lineares que aumenta exponencialmente com o número de participantes. Adicionalmente, o lado direito de cada restrição que define o núcleo do jogo cooperativo define o valor de uma determinada coalisão que, no presente trabalho, é obtido através de um modelo de otimização estocástica de dois estágios. Este trabalho compara diferentes metodologias computacionalmente eficientes baseadas em programação linear inteira mista e na técnica de decomposição de Benders para encontrar vetores de alocação de quotas que pertençam ao núcleo de portfólios de larga escala de geradores de energia renovável. São apresentados estudos de casos que utilizam dados reais do sistema elétrico brasileiro. / [en] Portfolios of renewable electricity sources are interesting risk-management mechanisms for trading in electricity contract markets. When they are formed by players belonging to different companies, their stability relies on the way the riskmitigation benefit generated by the optimal portfolio is allocated through individual participants. The problem of reaching a stable allocation can be mathematically formulated in terms of finding a quota-sharing vector belonging to the Core of a cooperative game, which can be formulated as a set of linear constraints that exponentially grows with the number of participants. Moreover, the right-hand-side of each constraint defining the Core relies on a given coalition value which, in the present work, is obtained by a two-stage stochastic optimization model. This work presents and compares efficient methodologies mainly based on mixed integer linear programming and Benders decomposition to find quota allocation vectors that belongs to the Core of large-scale renewable energy portfolios. Case studies are presented with realistic data from the Brazilian power system.

Submitted by JEFERSON BACK VANDERLINDE null (jefersonbv@yahoo.com.br) on 2017-11-01T16:38:25Z No. of bitstreams: 1 jeferson_tese_final_20171101.pdf: 4860852 bytes, checksum: 2f99c37969be3815f82b1b4455a40230 (MD5) / Approved for entry into archive by LUIZA DE MENEZES ROMANETTO (luizamenezes@reitoria.unesp.br) on 2017-11-13T15:38:34Z (GMT) No. of bitstreams: 1 vanderlinde_jb_dr_ilha.pdf: 4860852 bytes, checksum: 2f99c37969be3815f82b1b4455a40230 (MD5) / Made available in DSpace on 2017-11-13T15:38:34Z (GMT). No. of bitstreams: 1 vanderlinde_jb_dr_ilha.pdf: 4860852 bytes, checksum: 2f99c37969be3815f82b1b4455a40230 (MD5) Previous issue date: 2017-09-06 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Neste trabalho, consideram-se a análise teórica e a implementação computacional dos algoritmos Primal Simplex Canalizado (PSC) e Dual Simplex Canalizado (DSC) especializados. Esses algoritmos foram incorporados em um algoritmo Branch and Bound (B&B) de modo a resolver o problema de Planejamento da Expansão de Sistemas de Transmissão (PEST). Neste caso, o problema PEST foi modelado usando os chamados modelo de Transportes e modelo Linear Disjuntivo (LD), o que produz um problema de Programação Linear Inteiro Misto (PLIM). O algoritmo PSC é utilizado na resolução do problema de Programação Linear (PL) inicial após desconsiderar a restrição de integralidade do problema PLIM original. Juntamente com o algoritmo PSC, foi implementada uma estratégia para reduzir o número de variáveis artificiais adicionadas ao PL, consequentemente reduzindo o número de iterações do algoritmo PSC. O algoritmo DSC é utilizado na reotimização eficiente dos subproblemas gerados pelo algoritmo B&B, através do quadro ótimo do PL inicial, excluindo, assim, a necessidade da resolução completa de cada subproblema e, consequentemente, reduzindo o consumo de processamento e memória. Nesta pesquisa, é apresentada uma nova proposta de otimização, e, consequentemente, a implementação computacional usando a linguagem de programação FORTRAN que opera independentemente de qualquer solver. / In this research, the theoretical analysis and computational implementation of the specialized dual simplex algorithm (DSA) and primal simplex algorithm (PSA) for bounded variables is considered. These algorithms have been incorporated in a Branch and Bound (B&B) algorithm to solve the Transmission Network Expansion Planning (TNEP) problem. In this case, the TNEP problem is modeled using transportation model and linear disjunctive model (DM), which produces a mixed-integer linear programming (MILP) problem. After relaxing the integrality of investment variables of the original MILP problem, the PSA is used to solve the initial linear programming (LP) problem. Also, it has been implemented a strategy in PSA to reduce the number of artificial variables which are added into the LP problem, and consequently reduces the number of iterations of PSA. Through optimal solution of the initial LP, the DSA is used in efficient reoptimization of subproblems, resulting from the B&B algorithm, thus excludes the need for complete resolution of each subproblems, which results reducing the CPU time and memory consumption. This research presents the implementation of the proposed approach using the FORTRAN programming language which operates independently and does not use any commercial solver.

O objetivo de estudo desta dissertação é o desenvolvimento de um modelo de planejamento agregado da produção que apóie as decisões de nível gerencial e de diretoria das usinas de açúcar e álcool no que tange às variedades de cana colhidas em cada semana, às compras de cana-de-açúcar de terceiros, ao tipo de transporte (próprio ou terceirizado) a se utilizar em cada semana, ao total de cana moída por semana para atendimento da demanda e aos processos (industrial e comercial) que se devem escolher para produzir e comercializar açúcar e álcool. As decisões devem ocorrer em função de preços nos mercados interno, externo e mercado futuro, do fluxo de caixa da empresa, da capacidade da usina para armazenar açúcar e álcool e da possibilidade de uso de estoque de terceiros. As decisões por compra de cana, escolha de processos e venda de produtos são tomadas semanalmente num horizonte móvel de planejamento de 52 semanas, que inclui o tempo de safra no centro-sul do Brasil (meados de março a meados de dezembro, aproximadamente 36 semanas) mais o período de entressafra (aproximadamente 16 semanas, de meados de dezembro a meados de março). A procura por melhores estratégias de comercialização de tal forma a auxiliar a tomada de decisões é uma necessidade constante dos empresários do setor, que muitas vezes são surpreendidos pelas variações de preços de açúcar e álcool no mercado interno, externo e mercado futuro. Na parte comercial, este trabalho utiliza o método Delphi de previsão de preços de açúcar e álcool que balizam as tomadas de decisão no planejamento e controle da produção das usinas de açúcar e álcool. Define-se Hedge como a operação financeira de proteger determinado ativo de uma empresa contra variações inesperadas de preços. Neste trabalho, utiliza-se um modelo de escolha de mix de produto para Hedge vinculado à lucratividade e minimização de risco denominado Modelo de Semi- Variância com análise de cenários de Markowitz. Nas decisões relacionadas com as partes agrícola, industrial e comercial, faz-se uso de um modelo de programação linear inteira mista e para resolvê-lo utiliza-se o software de programação matemática LINGO e suas interfaces com a planilha eletrônica Excel. Nas decisões vinculadas ao mix ótimo para o Hedge em cada semana, faz-se uso de um modelo de programação quadrática resolvido pelo LINGO e suas interfaces com a planilha eletrônica Excel. Um estudo de caso foi realizado numa usina de açúcar e álcool no município de Junqueirópolis (SP) para validar o modelo proposto. / The objective of study this dissertation is to develop a model of aggregate production planning to support the decisions of management and board level of sugar and alcohol plants in regard to varieties of cane harvested each week, purchasing cane of nonsugar, the type of transport (own or outsourced) to use each week, the total cane processed per week for taking care of the demand and processes (industrial and commercial) and that must be chosen to produce and sell sugar and alcohol. Decisions must occur in terms of domestic, foreign and future market prices, the company\'s cash flow and the capacity to store sugar and alcohol and the possibility of using stock to third parties. Decisions about buying cane, choice of processes and products for sale are made in a weekly mobile planning horizon of 52 weeks, which includes the time of harvest in central-southern Brazil (mid-March to mid-December, approximately 36 weeks) plus the off-season (approximately 16 weeks, from mid-December to mid March). The demand for better marketing strategies to help such decision making is a constant need for entrepreneurs in the sector, which are often surprised by the changes in prices of sugar and alcohol in the internal, external and future market. In the commercial part, this study uses the Delphi method of forecasting the price of sugar and alcohol that guides the decision-making in planning and controlling the production of sugar and alcohol plants. Hedging is defined as a financial transaction to protect certain assets of a business against unexpected changes in prices. In this work, it is used a model of choice of product mix for Hedge linked to profitability and minimizing risk named Model of Semi-Variance analysis with scenarios of Markowitz. In decisions related to the agricultural, industrial and commercial parts it is used a type of mixed integer linear programming and to solve it is used the mathematical programming software LINGO and its interface with Excel spreadsheets. In decisions related to the optimal mix for Hedge in each week, is used a quadratic programming model solved by LINGO and its interface with Excel spreadsheets. A case study was conducted in a sugar mill and alcohol in the city of Junqueirópolis (SP) to validate the proposed model.

O ambiente de produção abordado neste trabalho é o Job Shop Flexível (JSF), uma generalização do Job Shop (JS). O problema de programação de tarefas, ou jobs, no ambiente JS é classificado por Garey; Johnson e Sethi (1976) como NP-Difícil e o JSF é, no mínimo, tão difícil quanto o JS. O JSF é composto por um conjunto de jobs, cada qual constituído por operações. Cada operação deve ser processada individualmente, sem interrupção, em uma única máquina de um subconjunto de máquinas habilitadas. O principal critério de desempenho considerado é a minimização dos atrasos dos jobs. São apresentados modelos de Programação Linear Inteira Mista (PLIM) para minimizar o atraso total e o instante de término da última operação, o makespan. São propostas novas regras de prioridade dos jobs, além de adaptações de regras da literatura. Tais regras são utilizadas por heurísticas construtivas e são aliadas a estratégias cujo objetivo é explorar características específicas do JSF. Visando aprimorar as soluções inicialmente obtidas, são propostas buscas locais e outros mecanismos de melhoria utilizados no desenvolvimento de três meta-heurísticas de diferentes categorias. Essas meta-heurísticas são: Iterated Local Search (ILS), classificada como meta-heurística de trajetória; Greedy Randomized Adaptive Search (GRASP), meta-heurística construtiva; e Artificial Bee Colony (ABC), meta-heurística populacional recentemente proposta. Esses métodos foram selecionados por alcançarem bons resultados para diversos problemas de otimização da literatura. São realizados experimentos computacionais com 600 instâncias do JSF, permitindo comparações entre os métodos de resolução. Os resultados mostram que explorar as características do problema permite que uma das regras de prioridade propostas supere a melhor regra da literatura em 81% das instâncias. As meta-heurísticas ILS, GRASP e ABC chegam a conseguir mais de 31% de melhoria sobre as soluções iniciais e a obter atrasos, em média, somente 2,24% superiores aos das soluções ótimas. Também são propostas modificações nas meta-heurísticas que permitem obter melhorias ainda mais expressivas sem aumento do tempo de execução. Adicionalmente é estudada uma versão do JSF com operações de Montagem e Desmontagem (JSFMD) e os experimentos realizados com um conjunto de 150 instâncias também indicam o bom desempenho dos métodos desenvolvidos. / The production environment addressed herein is the Flexible Job Shop (FJS), a generalization of the Job Shop (JS). In the JS environment, the jobs scheduling problem is classified by Garey; Johnson and Sethi (1976) as NP-Hard and the FJS is at least as difficult as the JS. FJS is composed of a set of jobs, each consisting of operations. Each operation must be processed individually, without interruption, in a single machine of a subset of enabled machines. The main performance criterion is minimizing the jobs tardiness. Mixed Integer Linear Programming (MILP) models are presented. These models minimize the total tardiness and the completion time of the last operation, makespan. New priority rules of jobs are proposed, as well as adaptations of rules from the literature. These rules are used by constructive heuristics and are combined with strategies aimed at exploiting specific characteristics of FSJ. In order to improve the solutions initially obtained, local searches and other improvement mechanisms are proposed and used in the development of metaheuristics of three different categories. These metaheuristics are: Iterated Local Search (ILS), classified as trajectory metaheuristic; Greedy Randomized Adaptive Search (GRASP), constructive metaheuristic, and Artificial Bee Colony (ABC), recently proposed population metaheuristic. These methods were selected owing to their good results for various optimization problems in the literature. Computational experiments using 600 FJS instances are carried out to allow comparisons between the resolution methods. The results show that exploiting the characteristics of the problem allows one of the proposed priority rules to exceed the best literature rule in about 81% of instances. Metaheuristics ILS, GRASP and ABC achieve more than 31% improvement over the initial solutions and obtain an average tardiness only 2.24% higher than the optimal solutions. Modifications in metaheuristics are proposed to obtain even more significant improvements without increased execution time. Additionally, a version called Disassembly and Assembly FSJ (DAFJS) is studied and the experiments performed with a set of 150 instances also indicate good performance of the methods developed.

Ce manuscrit présente des travaux visant à intégrer un module d'ordonnancement (ProSched) à l'environnement de modélisation et simulation dynamique hybride PrODHyS dans le but d'automatiser la génération de scénarii de simulation de procédés discontinus sur la base d'une recette et d'une liste d'ordres de fabrication (OF). La méthodologie développée repose sur une approche mixte optimisation/simulation. Dans ce cadre, trois points essentiels ont été développés dans ces travaux : - tout d'abord, concevoir et développer des composants réutilisables (classes de recette) permettant de modéliser de manière hiérarchisée et systématique le déroulement des opérations unitaires. Pour cela, les notions de jeton Task et de macro-place paramétrable ont été introduites dans les RdPDO et permettent de décrire les recettes à réaliser par assemblage de ces composants prédéfinis. - ensuite, définir un modèle mathématique générique d'ordonnancement basé sur un formalisme de représentation bien établi (le R.T.N.) qui permet de modéliser les principales caractéristiques d'un procédé discontinu et de fournir l'ensemble des données d'entrée nécessaires au modèle de simulation. Pour cela, un modèle PLNE basé sur la formulation Unit Specific Event a été mis en œuvre. - enfin, définir l'interface existant entre le modèle d'optimisation et le modèle de simulation, à travers la notion de place de pilotage et de centre de décision au niveau du simulateur. Dans ce cadre, différentes stratégies de couplage sont proposées. Les potentialités de cette approche sont illustrées par la simulation d'un procédé complet. / This thesis presents works which aim to incorporate a scheduling module (ProSched) to an environment for modeling and dynamic hybrid simulation PrODHyS in order to automate the generation of scenarios for simulation of batch processes based on a recipe and a list of production orders (OF). The methodology developed is based on a mixed optimization / simulation approach. In this context, three key points have been developed in this work: - First, design and develop reusable components (recipe classes) for the hierarchical and systematic modeling of the sequencing of unit operations. For this, the notions of Task token and macro-place have been introduced in the RdPDO formalism and allow the modeling of recipes by assembling these predefined components. - Secondly, define a generic mathematical model of scheduling based on a well defined graphical formalism (RTN) that models the main characteristics of batch processes and provide all input data necessary to the simulation model. For this, a MILP model based on the Unit Specific Event formulation has been implemented. - Finally, define the interface between the optimization model and the simulation model through the concept of control place and decision-making center at the simulator level. In this context, various strategies of mixing optimization and simulation are proposed. The potential of this approach is illustrated by the simulation of a complete manufacturing process

[EN] In this Thesis, we analyze an industry type that has two basic features: Its products are distributed in a geographic level and it reuses material after that is consumed to made new products. Because of these features, it represents a logistics challenge. This kind of supply chains is characterized also by having various stages that permit to carry the products from facility plants to the final customer. First, it is analyzed the supply chain structure, from production, in facility plants to the final customer by retailers, downstream; and from the customer, when returnable materials are liberated, classification and transport upstream to facility plants , where are prepared and used many times, until the end of their usable life. The attention focus is the integration of two fundamental functions of the supply chain management, in this case, production planning and distribution planning including the collection of reusable materials. It is precise to plan the production to satisfy retailers demand and to the procurement of returnable materials which after being consumed return to the process. It is made a bibliographical review about production planning, reverse logistics and vehicle routing and, it is built mathematical programming models to make decisions in production planning (in facility plants), inventories planning (in all the supply chain), and the deliveries that should be done in both senses in their respective routes. As a result of this research, there are three models of Mixed Integer Linear Programming: -Production planning model: Mathematical model which represents all the facilities in each one of the levels of the supply chain and their geographical localization. It lets to determine production and stock levels and delivery quantities in order to satisfy the demands in a planning horizon of several periods. This model has capacity constraints for both production and storage, constraints of demand, compatibility between products and facilities because not all the products can be in all facilities, and constraints of direct deliveries among determinate nodes in the supply chain (particularly between plants and retailers), especially. -Vehicle routing model: mathematical model that represents products distribution from different source nodes to destine nodes, it use a heterogeneous fleet, in a time period. One of the model assumptions is that a node acts as product emitting or recipient (facility plants or retailers). However, an intermediate node can take one of these roles depending if is the origin or destine of goods. This model has fleet use constraints, volume, and mass capacity vehicle constraints, fluxes in the supply chain. A vehicle can visit several destine nodes after leave from a source node, and a node can be visited by one or more vehicles (because of compatibility conditions between products and nodes or by economical aspects) and there are soft and strong time windows. -Integrated logistics and production management model: It is a model that join the two previous models, because of it, it is possible to plan production, inventories, and vehicle routing for a horizon time of several periods. Three proposed models have the objective of minimizing the total cost and are evaluated with testing instances. The Models were built using Gurobi 6.5.1 optimization model and Python2.7 pro- gramming language. Moreover, it was built a constructive heuristics and a genetic algorithm using Python programming language and Pyevolve software, as an alternative to solving routing vehicles, to reduce execution time. / [ES] En esta tesis se analiza un tipo de industria en el que: sus productos son distribuidos ampliamente a nivel geográfico y emplean elementos que una vez usados regresan al proceso de fabricación para hacer parte de nuevos productos. Este tipo de cadena se caracteriza por la presencia de varios eslabones que permiten acercar el producto al consumidor final. Se hace un análisis de la estructura de la cadena de suministros, desde la producción, en las plantas hasta su llegada al cliente final por intermedio de los detallistas, en el sentido "aguas abajo", y desde el consumo del producto y la consecuente liberación de los materiales reutilizables que deben conducirse "aguas arriba" hasta las plantas de producción, donde son preparados y utilizados una y otra vez hasta el final de su vida útil. El foco de atención consiste en la integración de dos funciones fundamentales de la cadena de suministros: la planeación de la producción y la distribución del producto con la consecuente recolección de los materiales de retorno. Se precisa realizar la planeación de la producción de manera que dé satisfacción a la demanda que tiene lugar en los detallistas, y que se abastezca de los materiales de retorno que una vez consumido el producto se reincorporan al proceso. Se hace una revisión bibliográfica sobre planeación de la producción, logística inversa y ruteo de vehículos y; se construyen modelos de programación matemática que soportan la toma de decisiones en la planeación de la producción, de los inventarios (en toda la cadena de suministros), y de los envíos que deben realizarse en uno y otro sentido en las correspondientes rutas. Como resultado de la investigación se presentan tres modelos de programación lineal entera mixta (MILP), así: -Modelo de planeación de la producción: representa la cadena de suministros a partir de las instalaciones de cada uno de los niveles y su ubicación geográfica, y determina los volúmenes a producir, a almacenar y las cantidades que se deben enviar desde cada nodo de la cadena de suministros a los demás, de modo que se satisfaga la demanda en un horizonte de planeación compuesto por varios periodos. Este modelo presenta restricciones de capacidad tanto de producción como de almacenamiento, de demanda, de compatibilidad de productos con las instalaciones, y de envíos directos entre determinados nodos de la cadena de suministros. -Modelo de ruteo de vehículos: representa la forma en que se deben distribuir los productos desde los diferentes nodos fuente a los diferentes nodos destino, haciendo uso de una flota heterogénea de vehículos, en un periodo de tiempo. El modelo se hace bajo el supuesto de que los nodos son o emisores o receptores del producto, un nodo intermedio asume uno de estos roles según si es origen o destino de las mercancías. Presenta restricciones asociadas al uso de la flota, a su capacidad tanto en masa como en volumen, los flujos a lo largo de la cadena bajo la premisa de que un vehículo parte de un nodo fuente y puede visitar varios nodos destino antes de regresar, y de que un nodo puede ser visitado por varios vehículos (por condiciones de compatibilidad entre nodos y productos o porque resulta más económico), ante la presencia de ventanas de tiempo tanto rígidas como suaves. -Modelo integrado de gestión de producción y logística: reúne los dos modelos anteriores, es decir, permite la planeación de la producción, la gestión de los inventarios y el ruteo de vehículos para un horizonte de planeación determinado. Los tres modelos se proponen con el objetivo de minimizar los costes totales y son evaluados con instancias de prueba. Los modelos fueron implementados haciendo uso del software de optimización Gurobi y del lenguaje de programación Python. Además, se implementó una heurística constructiva y un algortimo genético utilizando Python y el software Pyevolve, como alternativa a la solución del VRP, para reducir el tiem / [CAT] Aquesta tesi s'analitza un tipus d'indústria que: els seus productes són distribuïts àmpliament a nivell geogràfic i fan ús d'elements que una vegada usats tornen al procés de fabricació per a fer part dels nous productes. Aquest tipus de cadenes es caracteritza per la presència de diversos eslavons que permeten acostar el producte des de les plantes de producció fins a arribar al consumidor final. Es fa una anàlisi de l'estructura de la cadena de subministraments, des de la producció, en les plantes fins a la seua arribada al client final per mitjà dels detallistes, en el sentit "aigües a baix", i des del consum del producte i el conseqüent alliberament dels materials reutilitzables que han de conduir-se "aigües a dalt" fins a les plantes de producció, on són preparats i utilitzats una vegada i una altra, fins al final de la seua vida útil. El focus d'atenció consisteix en la integració de dues funcions fonamentals de la cadena de subministraments: la planificació de la producció i la distribució del producte amb la conseqüent recol·lecció dels materials de tornada. Sobre aquest tema, es precisa realitzar la planificació de la producció de manera que done satisfacció a la demanda que té lloc en els detallistes, i que es proveïsca dels materials de tornada que, una vegada consumit el producte, es reincorporen al procés. Es fa una revisió bibliogràfica del relacionat amb planificació de la producció, logística inversa i enrutament de vehicles i es construeixen models de programació matemàtica que suporten la presa de decisions en la planificació de la producció, dels inventaris (tota la cadena de subministraments), i dels enviaments que han de realitzar-se en un i un altre sentit en les corresponents rutes. Com a resultat de la recerca es presenten tres models de Programació Lineal Sencera Mixta (MILP), així: -Model de planificació de la producció: representa la cadena de subministraments a partir de les instal·lacions de cadascun dels nivells i la seua ubicació geogràfica, i es determinen els volums a produir, a emmagatzemar i les quantitats que s'han d'enviar des de cada node de la cadena de subministraments als altres, de manera que se satisfaça la demanda en un horitzó de planificació compost per diversos períodes. Aquest model presenta restriccions de capacitat tant de producció com d'emmagatzematge, de demanda, de compatibilitat de productes amb les instal·lacions, i d'enviaments directes entre determinats nodes de la cadena de subministraments, entre altres aspectes. -Model de enrutament de vehicles: representa la forma en què s'han de distribuir els productes des dels diferents nodes font als diferents nodes destine, fent ús d'una flota heterogènia de vehicles, en un període de temps. Els nodes són o emissors o receptors del producte, encara que en la seua utilització un node intermedi pot assumir un d'aquests rols segons si és origen o destinació de les mercaderies. Per a aquest model es tenen en compte les restriccions associades a l'ús de la flota, a la seua capacitat tant en massa com en volum, els fluxos al llarg de la cadena sota la premissa que un vehicle parteix d'un node font i pot visitar diversos nodes destine abans de tornar, i que un node pot ser visitat per diversos vehicles, davant la presència de finestres de temps tant rígides com suaus. -Model integrat de gestió de producció i logística: reuneix els dos models anteriors, és a dir, permet la planificació de la producció, la gestió dels inventaris i el enrutament de vehicles per a un horitzó de planificació determinat. Els tres models es proposen amb l'objectiu de minimitzar els costos totals i són avaluats amb instàncies de prova, apropiades. Els models van ser implementats fent ús del programari d'optimització Gurobi i del llenguatge de programació Python. A més, es va implementar una heurística constructiva i un algortime genètic utilitzant Python i el programari Pye / Parra Peña, J. (2016). Modelizado para la planeación de la producción y la logística directa e inversa de una cadena de suministro cuyo abastecimiento depende parcialmente de los materiales de retorno [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/73726 / TESIS

Le transport fluvial de conteneurs constitue une activité économique importante qui suscite un intérêt grandissant de la part de scientifiques. Considéré comme durable et économique, le transport par barge a été identifié comme étant une alternative compétitive pour le transport de marchandises, en complément des modes traditionnels de transport, routier et ferroviaire. Néanmoins, les travaux de recherche en rapport avec la planification et le management du transport par barge, en particulier dans le contexte du transport intermodal, sont encore peu abondants. Le but de cette thèse est d’apporter une contribution dans ce domaine, par la proposition de modèles et de méthodes de planification et gestion avancées, dans le cadre d’un système d’aide à la décision pour le transport de conteneurs par barge développé pour accompagner les opérateurs de transport. La méthodologie proposée fait appel à des concepts et principes de gestion du revenu, des ressources et des services de transport pour la conception de plans de services réguliers avec horaires, au niveau tactique. Les opérateurs de transport peuvent ainsi offrir des plans de transport avec des services plus flexibles pour leurs clients, tout en assurant un meilleur niveau de fiabilité. Plus de demandes de transport pourront ainsi être satisfaites, avec globalement une plus grande satisfaction des chargeurs. Une originalité importante proposée par notre approche est l’utilisation de principes et techniques de gestion du revenu (segmentation du marché, classes tarifaires...) aussi bien au niveau opérationnel de la modélisation qu’au niveau tactique. Les problèmes d’optimisation sont formalisés sous forme de modèles de programmation linéaire mixte en nombres entiers (PLNE), implémentés et testés sous différentes configurations de réseaux de transport et différents scénarios de demandes, et ce pour chaque niveau de décision. Au niveau tactique, une nouvelle approche de résolution, combinant la recherche adaptative à voisinage large (ALNS) et la recherche taboue, est proposée pour résoudre des problèmes PLNE de grande taille. Une plateforme de simulation, qui intègre les niveaux tactique et opérationnel de prise de décision, est proposée pour la validation du système d’aide à la décision sous différentes configurations : différentes topologies du réseau physique, différents paramètres pour la gestion du revenu, différents degrés de précision caractérisant les prévisions de demande. Pour l’analyse des résultats numériques ainsi obtenus, plusieurs types d’indicateurs de performance sont proposés et utilisés. / Barge transportation is an important research topic that started to draw increasing scientific attention in the recent decade. Considered as sustainable, environment-friendly and economical, barge transportation has been identified as a competitive alternative for freight transportation, complementing the traditional road and rail modes. However, contributions related to barge transportation, especially in the context of intermodal transportation, are still scarce. The objective of this thesis is to contribute to fill this gap by proposing a reactive decision support system for freight intermodal barge transportation from the perspective of the carriers. The proposed system incorporates resource and revenue management concepts and principles to build the optimal set of scheduled services plans at the tactical level. Carriers may thus benefit from transportation plans offering increased flexibility and reliability. They could thus serve more demands and better satisfy customers. One novelty of the approach is the application of revenue management considerations (e.g., market segmentation and price differentiation) at both operational and tactical planning levels. The optimization problems are mathematically formalized and mixed integer linear programming (MILP) models are proposed, implemented and tested against various network settings and demand scenarios, for each decision level. At the tactical level, a new solution approach, combining adaptive large neighborhood search (ALNS) and Tabu search is designed to solve large scale MILP problems. An integrated simulation framework, including the tactical and the operational levels jointly, is proposed to validate the decision support system in different settings, in terms of physical network topology, revenue management parameters and accuracy degree of demand forecasts. To analyze the numerical results corresponding to the solutions of the optimization problems, several categories of performance indicators are proposed and used.