Rozprawy doktorskie na temat „Transport Resource Allocation”

Mobile traffic increases rapidly. Based on Ericsson’s forecast [1], mobile traffic is expected to grow at a compound annual growth rate on percentage of 45% as the number of smartphone subscriptions and the consumption per subscriber increase. The monthly data traffic volume is expected to grow 6 times between 2015 and 2021. As demand increases, new technologies are investigated and deployed to cover the user requirements. Intense effort is given by researchers for the arrival of fifth generation (5G) network. High performance and increased capacity requirements drive research to heterogeneous networks. With the term “heterogeneous network”, a network that consists of different technologies and architectures is described. A heterogeneous wireless network involves the combination of macro and micro cells to improve coverage and capacity. All the traffic generated from the mobile network should be transferred from the antenna, through an access network, to the main office and from there to the backbone network. Optical networks are considered as the ideal solution for this purpose and research drives technology towards the usage of optical networks in the Fixed Mobile Convergence (FMC) architectures. The FMC architectures are proposed architectures [2] that focus to converge the fixed, mobile access and aggregation networks into a single transport network. In this study, we analyze the FMC architecture. We particularly analyze the Fronthaul architecture in combination with transport technologies such as Next Generation – Passive Optical Network 2 (NG-PON2) and Wavelength Routed Wavelength Division Multiplexing PON (WR-WDM-PON). We also take under consideration traffic patterns of mobile networks generated in various urban areas in the city of Stockholm, based in different use of land. Based on the traffic pattern, the number of small cells needed per area is calculated. In this thesis project, the traffic patterns from the mobile network and the transport network architectures are studied. The purpose of this thesis is to create an algorithm and study different sharing scenarios of the underlying transport infrastructure. The results of this algorithm will reveal if sharing and reusing resources in the transport infrastructure is beneficial in terms of saving resources.
Mobiltrafik ökar snabbt. Baserat på Ericssons prognos [1], väntas mobiltrafiken få en årlig tillväxttakt på 45% i samband med att antalet smartphone-abonnemang och förbrukning per abonnent ökar. Den månatliga volymen av datatrafik väntas att öka sexfaldigt mellan 2015 och 2021. Allteftersom efterfrågan ökar, undersöks och distribueras ny teknik för att möta användarnas krav. Intensivt forskningsarbetearbete bedrivs inför av femte generationens (5G) nätverk. Högt ställda krav på prestanda och kapacitet är de drivande faktorerna i forskningen av heterogena nätverk. Med heterogena nätverk menas nätverk som består av olika teknologier och arkitekturer. Ett heterogent trådlöst nätverk involverar kombinationen av makrooch mimkroceller för att förbättra täckning och kapacitet. All trafik som genereras i mobila nätverk ska överföras från antennen, genom ett accessnät, till huvudkontoret, och därifrån till backbone-nätverket. Optiska nätverk betraktas som den idealiska lösningen för detta ändamål, och forskare driver teknologin mot användning av optiska nätverk i Fixed Mobile Convergence(FMC) arkitekturer. FMC arktekturer är föreslagna arkitekturerna som fokuserar på att konvergera fasta, mobila och aggregerings-nätverk till ett enda transportnät. I denna studie, analyserar vi FMC-arkitekturen. Vi analyserar särskilt Fronthaul-arkitekturen i kombination med transportteknologier, så som Next Generation Passive Optical Network 2 (NG-PON2) och Wavelength Routed Wavelength Division Multiplexing PON (WR-WDM-PON). Vi tar också hänsyn till trafikmönster i mobila nätverk i olika sorters urbana områden i Stockholm. Baserat på trafikmönstret räknas antalet små celler som behövs per område ut. I detta examensarbete är det trafikmönster från mobila nätverk och transportnätverksarkitekturer som studeras. Syftet med denna avhandling är att skapa en algoritm, och studera olika olika scenarion där den underliggande transportinfrastrukturens resurser delas. Resultatet av denna algoritm avslöjar om delning och återanvändning av resurser i transportnätverket är fördelaktigt när det gäller att spara resurser.

The world's hunger for connectivity appears to be endlessly growing, yet the capacity of the networks that underpin that connectivity is anything but endless. This thesis explores both short and long term solutions for increasing the capacity of the largest and most capacious of these networks, the backbones upon which the Internet is built: optical transport networks. In the short term, Flexi-grid technology has emerged as the evolution of fixed-grid WDM optical networks, providing higher potential throughput but suffering from an aggravated form of the spectrum ragmentation problem that affects fixed-grid networks. A novel path-based metric to better evaluate the fragmentation of spectral resources in flexi-grid networks is presented, which considers both the fact that free spectrum slices may not be available on all the links of a path, and the likelihood that an end-to-end spectral void is usable to route incoming connections, and tested by means of simulations, finding that it outperforms existing ones from literature. For the longer term, Space Division Multiplexing (SDM) is a promising solution to overcome the looming fiber capacity crunch, and, perhaps more importantly, can offer a beneficial ratio between the expected capacity gains and the resulting increase in the cost of the network thanks to Joint and Fractional Joint Switching architectures and integrated transceivers and amplifiers. A model for such network is presented, and multiple heuristics for solving the Routing, Space and Spectrum Allocation problem are described, studied via simulations and iteratively improved, with the objective of quantifying the likely performance of several SDM architectures under multiple traffic scenarios. In addition, possible improvements to joint switching architectures, and an experimental SDN control plane for SDM networks, are presented and characterized, again by means of simulations. SDM is shown to be an attractive technology for increasing future transport networks capacity, at a reasonable cost.

The world's hunger for connectivity appears to be endlessly growing, yet the capacity of the networks that underpin that connectivity is anything but endless. This thesis explores both short and long term solutions for increasing the capacity of the largest and most capacious of these networks, the backbones upon which the Internet is built: optical transport networks. In the short term, Flexi-grid technology has emerged as the evolution of fixed-grid WDM optical networks, providing higher potential throughput but suffering from an aggravated form of the spectrum ragmentation problem that affects fixed-grid networks. A novel path-based metric to better evaluate the fragmentation of spectral resources in flexi-grid networks is presented, which considers both the fact that free spectrum slices may not be available on all the links of a path, and the likelihood that an end-to-end spectral void is usable to route incoming connections, and tested by means of simulations, finding that it outperforms existing ones from literature. For the longer term, Space Division Multiplexing (SDM) is a promising solution to overcome the looming fiber capacity crunch, and, perhaps more importantly, can offer a beneficial ratio between the expected capacity gains and the resulting increase in the cost of the network thanks to Joint and Fractional Joint Switching architectures and integrated transceivers and amplifiers. A model for such network is presented, and multiple heuristics for solving the Routing, Space and Spectrum Allocation problem are described, studied via simulations and iteratively improved, with the objective of quantifying the likely performance of several SDM architectures under multiple traffic scenarios. In addition, possible improvements to joint switching architectures, and an experimental SDN control plane for SDM networks, are presented and characterized, again by means of simulations. SDM is shown to be an attractive technology for increasing future transport networks capacity, at a reasonable cost.

Le contrôle de congestion des canaux sans fil pour la communication véhiculaires (V2X) basée sur IEEE 802.11p ont été conçus pour un seul service de sécurité routière notamment ‘Cooperative Awareness’. Les futurs véhicules connectés seront basés sur plusieurs services V2X, avec un besoin de communication hétérogène, auxquels les mécanismes existants ne répondent pas. Dans cette thèse, on analyse plusieurs protocoles de contrôle de congestion et d’allocation de ressources de canaux, normalisés en Europe pour le déploiement V2X initial. Nous présentons des problèmes liés à l’approche existante, notamment l’utilisation inefficace de la capacité des canaux, la coordination dans la pile protocolaire, la gestion des ressources parmi plusieurs services et l’allocation des ressources pour un besoin asymétrique par les véhicules. On propose des améliorations en tenant compte des exigences des futurs scénarios V2X. Nous montrons les limites de la classification statique des services V2X pour la qualité de service et proposons une classification dynamique. Nous concevons un orchestrateur de ressources afin d'améliorer la coordination dans la pile protocolaire et de répartir dynamiquement la ressource de canal parmi plusieurs services d'un véhicule. En outre, on présente un mécanisme permettant d'orchestrer de manière décentralisée les ressources de canal parmi une distribution mixte de véhicules ayant des exigences d'utilisation de canal diverse, dans un canal de communication congestionné. Les résultats analytiques et résultats par simulations montrent la validité de notre approche qui améliore la performance des applications V2X, par rapport aux protocoles standardisés existants
Wireless channel congestion control and decentralized resource allocation for 802.11p based V2X communication have been widely investigated for a single Cooperative Awareness service, considering mostly homogenous communication requirement per vehicle. Future connected vehicles will be based on multiple V2X services, with heterogenous number of services and communication needs, which existing resource allocation mechanisms does not address. In this thesis, we analyze several decentralized congestion control and channel resource allocation protocols standardized in Europe for initial V2X deployment. We present issues with the existing approach, in particular the inefficient channel capacity utilization, problematic cross layer coordination, inability to balance resources among multiple V2X safety services and distributed resource allocation for asymmetric number of services per vehicle. We propose improvements to the shortcomings, considering the requirements of future V2X scenarios. We demonstrate the limitations of classifying V2X services using static priority for Quality of Service and propose dynamic prioritization. We design a resource orchestrator at the Service Layer to improve cross layer coordination and dynamically distribute the limited channel resource among multiple services of a vehicle. Furthermore, we present a mechanism to decentrally orchestrate channel resource among a mixed distribution of vehicles with diverse channel usage requirements under channel congestion. Analytical and simulation-based results show the validity of our proposed approach, and the V2X application performance improvement it renders, compared to existing standardized protocols

In the last years the steady development of autonomous driving technology has enabled the deployment of more mature autonomous vehicles. These vehicles have been applied in several pilot projects worldwide, most commonly in the form of small buses. At the same time, the amount of people traveling in especially urban areas is continuously growing, resulting in more trips in the transportation system. An efficient transportation system is therefore required to serve the growing passenger demand. Autonomous buses (AB) are assumed to have lower operational costs and with that public transport (PT) systems can potentially be designed more efficiently to facilitate the increased demand better. In this study, an AB specific simulation-based optimization framework is proposed which allows analyzing the impacts AB have on line-based PT systems. The thesis focuses on the transition from existing PT systems towards line-based PT systems operated partially or exclusively by AB. Existing work on PT service design is extended so that realistic AB systems can be investigated. This is achieved by (i) using AB specific operator cost formulations, (ii) integrating infrastructure costs required for AB operations, (iii) utilizing a dynamic, stochastic and schedule-based passenger assignment model for the simulation of PT networks and by (iv) formulating a multi-objective optimization problem allowing to investigate the stakeholder-specific impacts of AB. In Paper I the effects of AB, concerning service frequency and vehicle capacity, on fixed-line PT networks are investigated. Among other metrics, the changes are evaluated based on differences in level of service and passenger flow. Additionally, the sequential introduction of AB in existing PT systems is studied. The framework addresses a case study in Kista, Sweden. The study confirmed the initial hypothesis that the deployment of AB leads to an increase in service frequency and a marginal reduction in vehicle capacity. Furthermore, it could be seen that the deployment of AB increases the passenger load on AB lines and that passengers can shift from other PT modes towards the AB services. Paper II incorporates a multi-objective heuristic optimization algorithm in the simulation framework. The study investigates changes in transport network design based on the deployment of AB. The differences in user-focused and operator-focused network design are analyzed and the impact of AB on these is quantified. This study is applied to a case study in Barkarby, Sweden where a full-sized, line-based PT network is designed to exclusively operate AB. Among other findings, we show that the autonomous technology reduces the number of served bus stops and reduces the total PT network size. Additionally, average passenger waiting time can be reduced when deploying AB on user-focused PT networks, which in turn leads to a further reduction of user cost.
De senaste årens framsteg inom autonom körteknik har lett till mer mogna autonoma fordon. Dessa fordon har setts tillämpas i flera pilotprojekt över hela världen, oftast i form av små bussar. Samtidigt växer mängden människor som reser, särskilt i stadsområden, kontinuerligt vilket resulterar i fler resor i transportsystemet. Därför krävs ett effektivt transportsystem för att tillgodose det växande antalet passagerare. Autonoma bussar (AB) antas ha lägre driftskostnader och därmed kan system för kollektivtrafik (public transport, PT) potentiellt utformas mer effektivt för att underlätta den ökade efterfrågan bättre. I denna studie föreslås ett AB-specifikt simuleringsbaserat optimeringsramverk som gör det möjligt att analysera effekterna AB har på linjebaserade PT-system. Avhandlingen fokuserar på övergången från befintliga PT-system till linjebaserade PT-system som delvis eller uteslutande drivs av AB. Befintligt arbete med PT-tjänstdesign utvidgas så att realistiska AB-system kan undersökas. Detta uppnås genom att (i) använda AB-specifika operatörskostnadsformuleringar, (ii) integrera infrastrukturkostnader som krävs för AB-verksamhet, (iii) använda en dynamisk, stokastisk och schemabaserad modell för att tilldela passagerare vid simulering av PT-nät samt genom att (iv) formulera ett multifunktionellt optimeringsproblem som gör det möjligt att undersöka AB: s intressespecifika effekter. I artikel I undersöks effekterna av AB, med avseende på servicefrekvens och fordonskapacitet, på fasta linjer i PT-nät. Förändringar utvärderas bland annat utifrån skillnader i servicenivå och passagerarflöde. Dessutom studeras den sekventiella introduktionen av AB i befintliga PT-system. Det föreslagna ramverket tillämpas på en fallstudie i Kista, Sverige. Studien bekräftade den initiala hypotesen att utplaceringen av AB leder till en ökning av servicefrekvensen och en marginell minskning av fordonens kapacitet. Vidare kunde man se att utplaceringen av AB ökar passagerarbelastningen på AB-linjer och att passagerare kan skifta från andra PT-former mot AB-tjänsterna. Artikel II integrerar en multifunktionell heuristisk optimeringsalgoritm i ramverket för simuleringen. Studien undersöker förändringar i transportnätverkets design baserat på implementeringen av AB. Skillnaderna i användarfokuserad och operatörsfokuserad nätverksdesign analyseras och AB: s inverkan på dessa kvantifieras. Denna studie tillämpas på en fallstudie i Barkarby, Sverige, där ett fullstort linjebaserat PT-nät är utformat för att exklusivt driva AB. Vi visar bland annat att den autonoma tekniken reducerar antalet använda busshållplatser och reducerar den totala PT-nätstorleken. Dessutom kan implementeringen av AB på användarfokuserade PT-nät ytterligare förbättra servicenivån främst genom att minska den genomsnittliga väntetiden per passagerare.

Pour accompagner le développement des fermes urbaines professionnelles et des circuits courts et de proximité dans les grandes métropoles, nous proposons des outils à base de programmes mathématiques d'aide à la conception de fermes maraîchères diversifiées ciblant les demandes de différents types de clients en fruits et légumes frais locaux. Les solutions techniques de culture en ville se développent rapidement, mais compte-tenu des contraintes fortes de la production agricole urbaine (foncier limité, coûts opérationnels élevés), la définition de modèles économiques viables pour des fermes pérennes est un vrai défi pour les maraîchers urbains. Trois modèles en programmation linéaire mixte sont présentés et confrontés aux résultats du terrain : un modèle de dimensionnement stratégique annuel, un modèle de sélection de la meilleure combinaison de clients et un modèle de dimensionnement stratégique pluri-annuel des fermes maraîchères. Des résultats numériques et les performances des modèles sont présentés à partir de cas concrets multi-produits, multi-techniques et multi-périodes. Nos contributions résident dans la prise en compte de la périssabilité des produits frais dans le dimensionnement stratégique des systèmes de production, incluant le dimensionnement de la main d’œuvre agricole
To support the urban farm emergence trend in large metropolises, we propose decision support tools based on mathematical programs to design market gardening farms targetting the demands of different categories of clients in local fresh fruits and vegetables. Technical solutions develop rapdily, but the strong constraints linked to urban farming, such as limited surface and high operating costs, make difficult to define viable and sustainable business models for urban market gardeners. Three mixed integer linear programming models are presented: An annual strategic sizing model, a client combination selection model and a plurennial strategic sizing model for diversified fruit and vegetable farms. Numerical results and model performances are presented, based on multi-products, multi-techniques and multi-periods real cases. Our main contributions are the consideration of the perishable nature of fruits and vegetables in strategic production systems sizing models, including notably the investments and workforce sizing

Biomass production is spatially distributed resulting in high transportation costs when moving dedicated biomass crops and crop residues. A multifaceted approach was taken to address this issue as the low bulk and energy density of biomass limits transportation efficiency. Two systems were analyzed for the conversion of biomass into a denser feedstock applicable to on-farm use. Pelletization was able to densify the material into a solid fuel. Using a pilot scale flat ring pellet mill, the density of the material was able to be increased to at least 4.4 times that of uncompressed material. Pellet durability was found to be strongly related to the moisture content of the material entering the mill. Unlike with ring roller pellet mills, a higher durability was typically seen forbiomass materials with a preconditioned moisture content of 20% (w.b.). From a liquid fuel standpoint, the conversion of lignocellulosic material into biobutanol on-farm was the second method investigated. For the pretreatment of biomass, alkaline hydrogen peroxide spray was demonstrated to be an effective enhancer of saccharification. The viability of on-farm biobutanol preprocessing bunker facilities within Kentucky was analyzed using Geographic Information systems (GIS) to specifically address transportation related factors. The spatial variability of corn field production, size, and location were resolved by utilizing ModelBuilder to combine the various forms of data and their attributes. Centralized and Distributed preprocessing with Centralized refining (DC) transportation systems were compared. Centralized was defined as transport of corn stover directly from the field to a refinery. Distributed-Centralized was specified as going from the field to the biobutanol bunker with corn stover and from the bunker to the refinery with a dewatered crude biobutanol solution. For the DC design, the location of the field and refinery were fixed with the biobutanol bunker location being variable and dependent upon differing maximum transportation (8-80 km) cutoffs for biomass transport from the field to biobutanol bunkers. The DC designs demonstrated a lower (38 - 59%) total transportation cost with a reduced fuel use and CO2 emissions compared to the centralized system.

This thesis aims to investigate how the objectives of neoliberal economic thought and total defence structure are relatable in defence planning. With a theoretical base in research on outsourcing of military logistics, the thesis sets out to contribute to the field of research by conceptualising the objectives of a total defence structure as a way of understanding the defence posture in Sweden since 2015. This is analysed through an in-depth study of Swedish defence planning reports and interviews with relevant actors using a thematic approach of data analysis.  The research identifies that the objectives of neoliberal economic thought and total defence structure correlate to a certain extent but there are also some discrepancies that affect the defence planning and ability. The result indicates that there is a potential to combine the objectives more efficiently if the conceptualisations were developed to fit the current societal structure where outsourcing is a common practice.

Les collisions et les blocages sont des situations que l'on souhaite éviter dans beaucoup d'application de transport. Les méthodes de synthèse de contrôleurs peuvent garantir qu'un système en boucle fermé ne puisse pas atteindre de telles situations. Les méthodes issues de l'approche historique de Ramadge et Wonham permettent de traiter cette problématique mais font face à l'explosion combinatoire, qui empêche leurs applications sur les systèmes de grandes tailles. Dans le cadre de l'évitement des blocages, la grande majorité des approches font l'hypothèse que le système est sous contrôlabilité et observabilité totales. Cependant, cette hypothèse n'est pas vérifiée pour tous les systèmes. Cette thèse propose des méthodes de synthèses de contrôleurs applicables sur des systèmes sous observabilité et contrôlabilité partielles sans souffrir de l'explosion combinatoire. Ces méthodes se basent sur le formalisme des Réseaux de Petri, et concernent uniquement des classes pour la modélisation de systèmes d'allocation de ressources. Pour ne pas avoir recours à l'exploration des états, ces méthodes identifient les situations de blocages grâce à la résolution d'un problème d'optimisation mixte en nombres entiers. Ces méthodes sont par la suite utilisées pour le routage des trains dans un nœud ferroviaire. Une méthode systématique de modélisation de la circulation des trains dans un nœud ferroviaire est proposée. Le modèle ainsi obtenu est un Réseau de Petri pour la modélisation des systèmes d'allocation de ressources. Pour éviter les collisions et les blocages au seins de ce modèle, une méthode de synthèse reposant sur les apports théoriques de cette thèse est appliquée
Collisions and deadlocks are unwanted situations in many transport applications such as railway systems. Control synthesis methods can assure that a closed-loop system cannot lead to deadlock situations. Methods issued from the historical approach of Ramadge and Wonham allow dealing with this issue but encounter the problem of combinatorial explosion, that prevents its use on large systems. In the case of deadlock avoidance, the majority of approaches are based on the assumption of total controllability and observability. However, this assumption is not verified in all kinds of systems. This thesis proposes control synthesis methods that do not suffer from the combinatorial explosion. These methods are based on the formalism of Petri nets and uniquely concern classes of resource allocation systems. In a way to not resort to state exploration, these methods identify deadlock situations by resolving a mixed integer programming problem. These methods are then used for routing trains in a railway node. A systematic modeling approach of the trains' circulations in a railway node is developed. The model obtained is a Petri net for resources allocation system. To avoid collisions and deadlocks in the model, a control synthesis method, based on the theoretical background developed, is proposed

Les collisions et les blocages sont des situations que l'on souhaite éviter dans beaucoup d'application de transport. Les méthodes de synthèse de contrôleurs peuvent garantir qu'un système en boucle fermé ne puisse pas atteindre de telles situations. Les méthodes issues de l'approche historique de Ramadge et Wonham permettent de traiter cette problématique mais font face à l'explosion combinatoire, qui empêche leurs applications sur les systèmes de grandes tailles. Dans le cadre de l'évitement des blocages, la grande majorité des approches font l'hypothèse que le système est sous contrôlabilité et observabilité totales. Cependant, cette hypothèse n'est pas vérifiée pour tous les systèmes. Cette thèse propose des méthodes de synthèses de contrôleurs applicables sur des systèmes sous observabilité et contrôlabilité partielles sans souffrir de l'explosion combinatoire. Ces méthodes se basent sur le formalisme des Réseaux de Petri, et concernent uniquement des classes pour la modélisation de systèmes d'allocation de ressources. Pour ne pas avoir recours à l'exploration des états, ces méthodes identifient les situations de blocages grâce à la résolution d'un problème d'optimisation mixte en nombres entiers. Ces méthodes sont par la suite utilisées pour le routage des trains dans un nœud ferroviaire. Une méthode systématique de modélisation de la circulation des trains dans un nœud ferroviaire est proposée. Le modèle ainsi obtenu est un Réseau de Petri pour la modélisation des systèmes d'allocation de ressources. Pour éviter les collisions et les blocages au seins de ce modèle, une méthode de synthèse reposant sur les apports théoriques de cette thèse est appliquée
Collisions and deadlocks are unwanted situations in many transport applications such as railway systems. Control synthesis methods can assure that a closed-loop system cannot lead to deadlock situations. Methods issued from the historical approach of Ramadge and Wonham allow dealing with this issue but encounter the problem of combinatorial explosion, that prevents its use on large systems. In the case of deadlock avoidance, the majority of approaches are based on the assumption of total controllability and observability. However, this assumption is not verified in all kinds of systems. This thesis proposes control synthesis methods that do not suffer from the combinatorial explosion. These methods are based on the formalism of Petri nets and uniquely concern classes of resource allocation systems. In a way to not resort to state exploration, these methods identify deadlock situations by resolving a mixed integer programming problem. These methods are then used for routing trains in a railway node. A systematic modeling approach of the trains' circulations in a railway node is developed. The model obtained is a Petri net for resources allocation system. To avoid collisions and deadlocks in the model, a control synthesis method, based on the theoretical background developed, is proposed

Este trabalho teve como objetivo estudar métodos para a alocação dos custos de transporte na coleta de leite buscando maior aprofundamento e debate sobre as diversas formas geralmente empregadas e novas fórmulas para esta alocação, como a metodologia de custeio ABC. As metodologias tradicionalmente utilizadas baseiam-se no volume ou na distância para a alocação de custos aos produtores, gerando muitas vezes distorções que penalizam alguns produtores em detrimento de outros. Isto ocorre pois tais metodologias não requerem o conhecimento do processo logístico de coleta de leite para serem aplicadas. Por outro lado, a metodologia de custeio ABC traz contribuições significativas ao processo de alocação ao levar em consideração tanto as variáveis de volume e distância quanto tempo. Adiciona-se a isso a sua função gerencial ao exigir maior conhecimento e detalhamento do processo logístico, permitindo, desta forma, a identificação das atividades que resultam em maiores custos. A comparação das metodologias tradicionais e da metodologia de custeio ABC ocorreu com a realização de um estudo de caso aplicado à Cooperativa dos Produtores Agropecuários de São Pedro – COOPAMSP, a qual contava, no período de tempo considerado, com 46 produtores distribuídos em 3 linhas de coleta (L1, L2 e L3). Destacam-se os resultados obtidos pela comparação da metodologia empregada pela cooperativa, que aloca os custos em proporção direta ao volume, e a metodologia de custeio ABC. É evidente o sobrecusteio dos produtores situados à linha de coleta mais eficiente (L1) em benefício dos produtores das demais linhas (L2 e L3), menos eficientes, ao se utilizar a metodologia empregada pela cooperativa. Tal distorção é eliminada quando se utiliza a metodologia de custeio ABC.
This dissertation had the objective to study methods of transport cost allocation on milk collection in order to deepening and debating about several ways usually used and new ways to proceed the allocation, such as Activity Based Costing methodology. Methodologies traditionally used consider volume or distance to allocate costs to farmers generating distortions that penalize some farmers and benefit others in many cases. This has happened because traditional methodologies do not require knowledge about logistic process of milk collection to be applied. However, the Activity Based Costing methodology contributes to cost allocation process by taking into account variables as volume, distance and time. In addiction, the ABC methodology has managerial function by demanding a well-known logistic process allowing, in this way, to identify the most costly activities. The comparison between traditional methodologies and ABC methodology took place through a case study applied to Cooperativa dos Produtores Agropecuários de São Pedro – COOPAMSP, that had 46 farmers producing milk in three routes (L1, L2 and L3), in the period of time that was considered. The results obtained by comparison between cooperative's methodology, that allocates costs in direct proportion to volume, and ABC methodology are emphasized. The methodology used by the cooperative allocates the highest costs to the more efficient route (L1) in aid of less efficient farmers (L2 and L3). That distortion does not happen when ABC methodology is used.

O estudo tem como finalidade tratar a alocação de recursos no nível operacional com a presença de incertezas. Para isso, foi proposta uma abordagem de otimização usando métodos heurísticos. As soluções de problemas de produção e logística, comumente abordadas em pesquisa operacional, exploram diversos parâmetros dentre os quais o presente estudo considera três como de incerteza: demanda, tempo de execução e indisponibilidade de recursos. Para tal finalidade foi escolhido como estudo de caso a resolução de um problema de logística. O problema consiste na minimização dos custos de operação, na seleção de veículos em uma frota heterogênea, consolidação das cargas para cada cliente e na seleção do tipo de frete utilizado. Quanto ao tipo de frete, são considerados dois, os quais se diferenciam quanto aos ativos envolvidos na produção e ao tipo de prestação de serviço, sendo eles: i) frota da empresa com serviço terceirizado; ii) frota e serviços totalmente terceirizados. O problema original foi decomposto em duas etapas: i) Compartimentalizador e ii) Alocador. As duas etapas são solucionadas via a abordagem de Busca Tabu, sendo que a primeira etapa (Compartimentalizador) gera uma lista dos carregamentos factíveis que atenda pedidos de até três clientes distintos. O Alocador se utiliza da lista dos carregamentos factíveis para definir como e quando cada pedido será atendido. Os resultados indicam a viabilidade da adoção desta abordagem para a solução de problemas reais.
The study aims to address the allocation of resources at the operational level under uncertainties. For this reason, it was proposed an optimization approach based on heuristic methods. The resolutions of production and logistics problems, commonly addressed in operational research, explore various parameters among which the present study considers three variables of uncertainty: demand, operation time and resources availability. For this purpose a logistics problem was chosen as study of case. The problem consists in minimizing cost operation, selection of vehicles in a heterogeneous fleet, consolidation of loads for each client and selecting the type of freight payables. Regarding of freight payables types, there are centered in two different tariffs, mainly due to assets and service negotiation, such as: i) fleet controlled by company and service outsource; ii) fleet and service completely outsource. The resolution of the original problem was broke down in two steps: i) Compartmentalizer and ii) Allocator. Both steps are solved through Tabu Search approach; the first step (Compartmentalizer) generates a list of feasible shipments to fulfill orders up to three different customers. The second step, the allocator uses the list of feasible shipments to define how and when each request will be supplied. The results aim the feasibility of assumes this approach in order to solve real problems.

Les réseaux véhiculaires font l'objet de recherches actives aussi bien dans le domaine des réseaux que dans celui des transports. Le potentiel des réseaux véhiculaires à fournir des services comme l'information sur le trafic en temps réel ou sur les accidents font de cette technologie un domaine de recherche très important. Ces réseaux peuvent comporter des communications véhicule-à-véhicule (V2V), véhicule-à-infrastructure (V2I), ou une combinaison des deux. La norme IEEE 1609.4 est la spécification multicanal pour l'IEEE802.11p/WAVE des réseaux véhiculaires (VANETs). Elle utilise sept canaux, l'un étant un canal de contrôle (CCH) qui est écouté par les équipements de façon périodique, et les six autres canaux sont utilisés comme canaux de service (SCH). Elle définit également une division du temps en alternance entre les intervalles CCH et les intervalles SCH. L'objet de cette thèse de doctorat est d'évaluer les performances des réseaux VANETs dans le cas des communications véhiculaires sans infrastructure, et au niveau des couches inférieures du standard 802.11p. Dans la première partie, nous proposons une approche MAC d'allocation multicanal opportuniste dans un contexte sans infrastructure. Cette approche est conforme à la norme IEEE1609.4 -2010 de l'architecture WAVE pour un fonctionnement multicanal, et elle est conçue pour des applications de services de données (non urgentes), tout en assurant la transmission des messages de sécurité routière et des paquets de contrôle. Pour maintenir la qualité de service des deux types de messages (urgents et non-urgents) en exploitant la capacité du canal, deux solutions sont proposées. Dans la deuxième partie, lorsque le véhicule sélectionne son canal et contrôle son alternance temporelle entre CCH et SCH, il commence à transmettre ses paquets, en particulier sur le canal CCH, lesquels ont une durée de péremption. Nous présentons une approche visant à minimiser les collisions des émetteurs tout en évitant la contention de début d'intervalle, en particulier dans un contexte de densité élevée de véhicules. Même si les mécanismes proposés ci-dessus diminuent le taux de collision, il n'est pas possible de les supprimer complètement. Dans la troisième partie, nous traitons le problème des collisions entre les paquets diffusés sur le CCH, en particulier quand la charge des messages transmis dépasse la capacité du canal. Pour cela, nous proposons un nouveau mécanisme de codage réseau analogique adapté à la modulation QPSK pour les messages diffusés sur le CCH. Dans cette approche des symboles connus sont envoyés avant d'envoyer les paquets pour estimer les paramètres du canal et une solution explicite est utilisée pour inverser le système de la superposition de deux paquets

博士
國立交通大學
交通運輸研究所
97
Different transport stakeholders have different needs for transport infrastructure and services. Meeting the needs of stakeholders implies a trade-off of benefits and costs between supply and demand and creates issues of transport diversity. However, the literature has largely ignored these issues. This study aims to provide a framework evaluating transport diversity to promote quality of life. Transport diversity is defined as the satisfied level of stakeholder needs in this study and measured as the gap between expected goal and present values of stakeholder needs in the form of the Shannon-Weaver Index, or namely the Entropy. Transport diversity can assess whether the level to which important needs are satisfied equitably, and monitor whether the transportation system is moving towards sustainability via confirming the targets and the basic level of quality of life. The proposed framework suggests that the evaluation of transport diversity should consider distinct transportation needs, such as mobility, economic health, reliability, safety, accessibility, affordability, level of universal design, externality and resource over-utilization, of different stakeholder including various modal users, specific users and non-users simultaneously. Accordingly, transport diversity is thus the necessary condition for quality of life and sustainability. Improving the sustainability and quality of life with regard to transportation requires the support of transport diversity. This study hopes that the conceptual framework developed can assist decision-makers in understanding the relationship between transport diversity and sustainability, and provide a new assessing method for improvements in quality of life. After identifying the appropriate indicators referring to stakeholder needs, a preliminary spatiotemporal analysis is illustrated through an empirical study to discuss the managerial implications in the Taipei metropolitan area. The results reveal that transport diversity is improved from 2000 to 2005. In fact, the transportation system has progressed in the satisfactions of safety and reliability. However, the achievements of emission, accessibility and level of universal design perform poorly. Decision-makers could understand better resource allocation policies according to the analytical results. In particular, deficient quality of life for disable users should be improved effectively and efficiently. Additionally, resource allocation policies help planners decide when and how to invest transportation infrastructure and services. However, policies for improving transport diversity are difficult to design, implement, and quantify because of the uncertainty, feedback interaction, and complexity of system relationships. This study proposes a hybrid model integrating system dynamics, cognitive maps, and sensitivity model to tackle the problems. Notably, the result of sensitivity model reveals that the increment of private vehicle trips reduces transport diversity due to the increase of energy consumption, emission and accident rate. However, tuning policy delays does not significantly impact system performance through managerial choices of resource allocation. Moreover, the simulation results indicate that the gaps in stakeholder needs are generally opposite to transport diversity and positive proportion to private vehicle trips. This verifies that increasing public transit trips helps the system bridge the gap between user satisfactions of stakeholder needs. According to the system relationships constructed by the hybrid systematic simulation tool, fuzzy multi-objectives programming, a Pareto based mathematical approach is employed to solve the non-linear multi-objectives problems focusing on urban public transit systems for determine the impact of resource allocation on need satisfactions related to stakeholder behaviors. The proposed approach evades inefficient and inequitable resource allocation. Furthermore, analytical outcomes show that recent investments allocated to public transit system considered equitable stakeholder satisfactions both of MRT and bus, as well as promoted transport diversity in the Taipei metropolitan area.

Metropolitan area consists of huge population density, one or more urban areas, satellite cities, rural areas and towns. It highly concentrates economic activities to attract people from rural areas and is a complex spatial structures and reached a milestone in the development of road infrastructure to support transportation systems. Due to increase in the number of vehicles, many traffic related problems have arised, such as increase in delays, wastage of resources, high traffic accidents, economic losses, environmental pollution, security, privacy, etc. These problems are solved by using Intelligent Public Transportation System (IPTS), which consists of advanced technologies for transportation engineering including information technology (IT), computational intelligence, system control technology and system engineering. These advanced technologies play significant role in reducing traffic congestion, traffic accidents, traffic pollution, energy consumption and improving road traffic safety. In this thesis, we have designed IPTS for public transport in a metropolitan area. It consists of 5 major functional units, such as commuters’ arrival prediction, transport resource allocation, privacy preservation, context-based service management and transport management. We have used Emergent Intelligence (EI) technique with agents, as it is efficiently used, in several applications for collecting, analyzing, monitoring, and sharing information. Also, it takes dynamic decisions to give independence to each agent to take a decision. Some of these features are essential for public transport system. First, we have designed Commuters’ Arrival Prediction (CAP) scheme, which analyzes and predicts commuters’ patterns in the bus station using commuters’ arrival density, resource availability, history, and dynamic arrivals of commuters and vehicles. The proposed scheme provides an accurate analysis and prediction of commuters’ arrival density pattern at each bus station in the vicinity of transport depot. The predicted commuters’ density will be used to choose an optimal route to reach the desired destination in a metropolitan area. We have designed a Transport Resource Allocation (TRA) scheme using EI technique. The proposed scheme efficiently manages and provides resource over space and time based on predicted commuters’ arrival pattern, availability of resources, history, deficit resources and surplus resources of neighborhood transport depots. The EI technique is used to collect, analyze and share resources, reliability of resources, reliability of neighborhood depots, resource gathering time delay reliability and travel time reliability. We have built an analytical model for TRA scheme using discrete time Markov chain to analyse the system at steady state. The proposed scheme optimally allocates reliable resources and provides reliable public transport services to the commuters. In the metropolitan area, there will be privacy breach during communication among transport depot’s staff and agents, which results in anomalies like impersonation, malicious activities and greedy behavior and they lead to issues, such as traffic accidents, wastage of resources, delays, etc. For this, we have designed Public Transport Depot Privacy Preservation Scheme (PTDPPS) using EI technique in a metropolitan area. The proposed scheme consists of two phases: First, policy based privacy preservation for depot staff and Second, pseudonymous authentication based privacy preservation for communicating agents. The proposed scheme preserves privacy of resources, vehicle dispatch, allocation, private and public information of transport depot staff during communication inside and outside the transport depot. In a metropolitan area, managing and providing suitable services are difficult or impossible due to the non-linear and complex nature of transportation system. Thus, we have designed Context based Public Transportation Service Management (CPTSM). The EI technique is used for collecting, analyzing and sharing context information of vehicles, staff, commuters, routes, bus station units and environmental. The collected and analyzed context information are used to develop the policies to provide relevant services to commuters. The proposed system makes public transportation system most reliable and efficient for the commuters. We have designed an efficient public transportation management system in a metropolitan area by combining CAP, TRA, PTDPPS, and CPTSM, and made as a single entity at a transport depot in a metropolitan area. We have presented management of public transportation during normal conditions and disaster situations in one of the zone in the vicinity of transport depot using EI technique in a metropolitan area. During normal conditions, we have estimated and predicted the commuters’ arrival pattern and their density, and also estimated and allocated the resources at every bus stations. Using these information public transportation management is done at every bus stations in the vicinity of transport depot. During the disaster situations, proposed evacuation management system estimates boundary of disaster, evacuation exit points, victims density, severity of victims and resources required. The evacuation management system assigns fraction of victims to the desired destination through multiple evacuation exit points, and avoids the occurrence of secondary disaster in the same disaster zone in the metropolitan area. In summary, we have used EI technique with agents to design: 1) commuters’ arrival prediction, which efficiently analyzes and predicts the commuters arrival density; 2) transport resource allocation scheme, which dynamically allocates optimal resources; 3) privacy preservation schemes, which preserves the privacy of transport depot staff and agents; 4) context based service management, which efficiently makes use of context information of various transportation entities to provide accurate services; and 5) transport management for evacuation of victims, which effectively manages the disaster victims. We have simulated above mentioned schemes in different scenarios with various performance parameters. The results obtained in simulation and analysis show the importance of the approaches and effectiveness of the proposed system