It's commonly assumed that experience leads to efficiency, yet this is largely unaccounted for in resource-constrained project scheduling. This thesis considers the idea that learning effects could allow selected activities to be completed within reduced time, if they're scheduled after activities where workers learn relevant skills. This paper computationally explores the effect of this autonomous, intra-project learning on optimal makespan and problem difficulty. A learning extension is proposed to the standard RCPSP scheduling problem. Multiple parameters are considered, including project size, learning frequency, and learning intensity. A test instance generator is developed to adapt the popular PSPLIB library of scheduling problems to this model. Four different Constraint Programming model formulations are developed to efficiently solve the model. Bounding techniques are proposed for tightening optimality gaps, including four lower bounding model relaxations, an upper bounding model relaxation, and a Destructive Lower Bounding method. Hundreds of thousands of scenarios are tested to empirically determine the most efficient solution approaches and the impact of learning on project schedules. Potential makespan reduction as high as 50% is discovered, with the learning effects resembling a learning curve with a point of diminishing returns. A combination of bounding techniques is proven to produce significantly tighter optimality gaps.

University course planning and scheduling is the process of determining what courses to offer, how many sections are needed, determining the best term to offer each section, assigning a faculty member to instruct each section, and scheduling each section to a timeslot to avoid conflicts. The result of this task has an impact on every student and faculty member in the department. The process is typically broken down into three major phases: course offering planning, faculty assignment to planned course sections, and course scheduling into timeslots. This thesis looks at each of these phases for the Industrial and Manufacturing department and brings them together into a decision support and scheduling system. A decision support tool is created to facilitate planning of course offerings. Operations research is applied to assign sections to faculty members using a faculty preference driven integer linear programming model in order to minimize dissatisfaction in the department. Next, the faculty-section pairs are scheduled into university timeslots using a complex integer linear programming model. This scheduling model takes into consideration the faculty member time availability and preferences and general student time slot preferences as it minimizes dissatisfaction while avoiding conflicts among labs, faculty members and courses offered for each class level.

In a dynamically monitored environment the analyst team need timely and accurate information to conduct proactive action over complex situations. Typically, there are thousands of reported activities in a real time operation, therefore steps are taken to direct the analyst’s attention to the most important activity. The data fusion community have introduced the information fusion model, with multiple situational assessments. Each process lends itself to ranking the most important activities into a predetermined order. Unfortunately, the capability of a real time system can be hindered by the knowledge limitation problem, particularly when the underlying system is processing multiple sensor information. Consequently, the situational awareness domains may not rank the identified situation as perfect, as desired by the decision-making resources. This thesis presents advanced research carried out to evaluate the ranking capability of information from the situational awareness domains: perception, comprehension and projection. The Ranking Capability Score (RCS) has been designed for evaluating the prioritisation process. The enhanced (RCS) has been designed for addressing the knowledge representation problem in the user system relation under a situational assessment where the proposed number of tracking activities are dynamically shifted. Finally, the Scheduling Capability Score was designed for evaluating the scheduling capability of the situational awareness system. The proposed performance metrics have been successful in fulfilling their objectives. Furthermore, they have been validated and evaluated using an analytical approach, through conducting a rigorous analysis of the prioritisation and scheduling processes, despite any constraints related to a domain-specific configuration.

We developed a deep fusion methodology of non-destructive (NDT) in-situ infrared and ex- situ ultrasonic images for localization of porosity detection without compromising the integrity of printed components that aims to improve the Laser-based additive manufacturing (LBAM) process. A core challenge with LBAM is that lack of fusion between successive layers of printed metal can lead to porosity and abnormalities in the printed component. We developed a sensor fusion U-Net methodology that fills the gap in fusing in-situ thermal images with ex-situ ultrasonic images by employing a U-Net Convolutional Neural Network (CNN) for feature extraction and two-dimensional object localization. We modify the U-Net framework with the inception and LSTM block layers. We validate the models by comparing our single modality models and fusion models with ground truth X-ray computed tomography images. The inception U-Net fusion model localized porosity with the highest mean intersection over union score of 0.557.

Public sector services often face challenges in allocating limited resources effectively. Despite their fundamental importance to societal welfare, these systems often operate without sufficient analytical support, and their decision-making processes remain understudied in academic literature. While data-driven analytical approaches offer promising solutions for addressing complex tradeoffs and resource constraints, the unique characteristics of public systems create significant challenges for modeling and developing efficient solutions. This dissertation addresses these challenges by applying stochastic models to enhance decision-making in two critical areas: emergency medical services in healthcare and jail diversion in the criminal justice system.The first part focuses on integrating drones into emergency medical services to shorten response times and improve patient outcomes. We develop a Markov Decision Process (MDP) model to address the coordination between aerial and ground vehicles, accounting for uncertain travel times and bystander availability. To solve this complex problem, we develop a tractable approximate policy iteration algorithm that approximates value function through neural networks, with basis functions tailored to the spatial and temporal characteristics of the EMS system. Case studies using historical data from Indiana provide valuable insights for managing real-time EMS logistics. Our results show that drone augmentation can reduce response times by over 30% compared to traditional ambulances. This research provides practical guidelines for implementing drone-assisted emergency medical services while contributing to the literature on hybrid delivery systems.The second part develops data-driven analytical tools to improve placement decisions in jail diversion programs, balancing public safety and individual rehabilitation. Community corrections programs offer promising alternatives to incarceration but face their own resource constraints. We develop an MDP model that captures the complex tradeoffs between individual recidivism risks and the impacts of overcrowding. Our model extends beyond traditional queueing problems by incorporating criminal justice-specific features, including deterministic service times and convex occupancy-dependent costs. To overcome the theoretical challenges, we develop a novel unified approach that combines system coupling with policy deviation bounds to analyze value functions, ultimately establishing the superconvexity. This theoretical foundation enables us to develop an efficient algorithm based on time-scale separation, providing practical tools for optimizing diversion decisions. Case study based on real data from our community partner shows our approach can reduce recidivism rates by 28% compared to current practices. Beyond academic impact, this research has been used by community partners to secure program funding for future staffing.<p></p>

Lors du transport du bois de la forêt vers les usines, de nombreux événements imprévus peuvent se produire, événements qui perturbent les trajets prévus (par exemple, en raison des conditions météo, des feux de forêt, de la présence de nouveaux chargements, etc.). Lorsque de tels événements ne sont connus que durant un trajet, le camion qui accomplit ce trajet doit être détourné vers un chemin alternatif. En l’absence d’informations sur un tel chemin, le chauffeur du camion est susceptible de choisir un chemin alternatif inutilement long ou pire, qui est lui-même "fermé" suite à un événement imprévu. Il est donc essentiel de fournir aux chauffeurs des informations en temps réel, en particulier des suggestions de chemins alternatifs lorsqu’une route prévue s’avère impraticable. Les possibilités de recours en cas d’imprévus dépendent des caractéristiques de la chaîne logistique étudiée comme la présence de camions auto-chargeurs et la politique de gestion du transport. Nous présentons trois articles traitant de contextes d’application différents ainsi que des modèles et des méthodes de résolution adaptés à chacun des contextes. Dans le premier article, les chauffeurs de camion disposent de l’ensemble du plan hebdomadaire de la semaine en cours. Dans ce contexte, tous les efforts doivent être faits pour minimiser les changements apportés au plan initial. Bien que la flotte de camions soit homogène, il y a un ordre de priorité des chauffeurs. Les plus prioritaires obtiennent les volumes de travail les plus importants. Minimiser les changements dans leurs plans est également une priorité. Étant donné que les conséquences des événements imprévus sur le plan de transport sont essentiellement des annulations et/ou des retards de certains voyages, l’approche proposée traite d’abord l’annulation et le retard d’un seul voyage, puis elle est généralisée pour traiter des événements plus complexes. Dans cette ap- proche, nous essayons de re-planifier les voyages impactés durant la même semaine de telle sorte qu’une chargeuse soit libre au moment de l’arrivée du camion à la fois au site forestier et à l’usine. De cette façon, les voyages des autres camions ne seront pas mo- difiés. Cette approche fournit aux répartiteurs des plans alternatifs en quelques secondes. De meilleures solutions pourraient être obtenues si le répartiteur était autorisé à apporter plus de modifications au plan initial. Dans le second article, nous considérons un contexte où un seul voyage à la fois est communiqué aux chauffeurs. Le répartiteur attend jusqu’à ce que le chauffeur termine son voyage avant de lui révéler le prochain voyage. Ce contexte est plus souple et offre plus de possibilités de recours en cas d’imprévus. En plus, le problème hebdomadaire peut être divisé en des problèmes quotidiens, puisque la demande est quotidienne et les usines sont ouvertes pendant des périodes limitées durant la journée. Nous utilisons un modèle de programmation mathématique basé sur un réseau espace-temps pour réagir aux perturbations. Bien que ces dernières puissent avoir des effets différents sur le plan de transport initial, une caractéristique clé du modèle proposé est qu’il reste valable pour traiter tous les imprévus, quelle que soit leur nature. En effet, l’impact de ces événements est capturé dans le réseau espace-temps et dans les paramètres d’entrée plutôt que dans le modèle lui-même. Le modèle est résolu pour la journée en cours chaque fois qu’un événement imprévu est révélé. Dans le dernier article, la flotte de camions est hétérogène, comprenant des camions avec des chargeuses à bord. La configuration des routes de ces camions est différente de celle des camions réguliers, car ils ne doivent pas être synchronisés avec les chargeuses. Nous utilisons un modèle mathématique où les colonnes peuvent être facilement et naturellement interprétées comme des itinéraires de camions. Nous résolvons ce modèle en utilisant la génération de colonnes. Dans un premier temps, nous relaxons l’intégralité des variables de décision et nous considérons seulement un sous-ensemble des itinéraires réalisables. Les itinéraires avec un potentiel d’amélioration de la solution courante sont ajoutés au modèle de manière itérative. Un réseau espace-temps est utilisé à la fois pour représenter les impacts des événements imprévus et pour générer ces itinéraires. La solution obtenue est généralement fractionnaire et un algorithme de branch-and-price est utilisé pour trouver des solutions entières. Plusieurs scénarios de perturbation ont été développés pour tester l’approche proposée sur des études de cas provenant de l’industrie forestière canadienne et les résultats numériques sont présentés pour les trois contextes. / When wood is transported from forest sites to mills, several unforeseen events may occur, events which perturb planned trips (e.g., because of weather conditions, forest fires, or the occurrence of new loads). When such events take place while the trip is under way, the truck involved must be rerouted to an alternative itinerary. Without relevant information on such alternative itineraries, the truck driver may choose a needlessly long one or, even worse, an itinerary that may itself be "closed" by an unforeseen event (the same event as for the original itinerary or another one). It is thus critical to provide drivers with real-time information, in particular, suggestions of alternative itineraries, when the planned one cannot be performed. Recourse strategies to deal with unforeseen events depend on the characteristics of the studied supply chain, such as the presence of auto-loaders and the management policy of forestry transportation companies. We present three papers dealing with three differ- ent application contexts, as well as models and solution methods adapted to each context. In the first paper, we assume a context where truck drivers are provided a priori with the whole weekly plan. In this context, every effort must be made to minimize the changes in the initial plan. Although the fleet of trucks is homogeneous, there is a priority ranking of the truck drivers. The priority drivers are ensured the highest work- loads. Minimizing the changes in their plans is also a priority. Since the consequences of unforeseen events on transportation are cancellations and/or delaying of some trips, the proposed approach deals first with single cancellations and single delayed trips and builds on these simple events to deal with more complex ones. In this approach, we try to reschedule the impacted trips within the same week in such a way that a loader is free at the truck arrival time both at the forest site and at the mill. In this way, none of the other trips will be impacted or changed. This approach provides the dispatchers with alternative plans in a few seconds. Better solutions could be found if the dispatcher is allowed to make more changes to the original plan. In the second paper, we assume a context where only one trip at a time is communicated to the drivers. The dispatcher waits until the truck finishes its trip before revealing the next trip. This context is more flexible and provides more recourse possibilities. Also, the weekly problem can be divided into daily problems since the demand is daily and the mills are open only for limited periods in the day. We use a mathematical programming model based on a time-space network representation to react to disruptions. Although the latter can have different impacts on the initial transportation plan, one key characteristic of the proposed model is that it remains valid for dealing with all the unforeseen events, regardless of their nature. Indeed, the impacts of such events are reflected in the time-space network and in the input parameters rather than in the model itself. The model is solved for the current day each time an unforeseen event is revealed. In the last paper, the fleet of trucks is heterogeneous, including trucks with onboard loaders. The route configuration of the latter is different than the regular truck routes, since they do not have to be synchronized with the loaders. We use a mathematical model where the columns can be easily and naturally interpreted as truck routes. We solve this model using column generation. As a first step, we relax the integrality of the decision variables and consider only a subset of feasible routes. The feasible routes with a potential to improve the solution are added iteratively to the model. A time-space network is used both to represent the impacts of unforeseen events and to generate these routes. The solution obtained is generally fractional and a heuristic branch-and-price algorithm is used to find integer solutions. Several disruption scenarios were developed to test the proposed approach on case studies from the Canadian forest industry and numerical results are presented for the three contexts.

La question de l'accès au soin est cruciale dans notre société moderne. Un effet évident de la demande accrue de services de santé est l'augmentation du taux d'occupation dans les hôpitaux. La principale différence entre la dispensation de soins à l'hôpital et au domicile est la suivante: le patient doit se déplacer et toutes les ressources nécessaires à son traitement se trouvent dans le même endroit, tandis que dans les soins délivrés au domicile, les ressources doivent être déplacées au chevet du patient. Il existe plusieurs défis afin de pouvoir réaliser ce changement. Dans cette thèse nous traitons trois problèmes importants dans la conception des structures de soins à domicile. D’abord, la localisation des structures en minimisant les coûts logistiques, où nous développons trois modèles incluant différentes caractéristiques comme du système de santé comme les coûts liés aux déplacements des ressources, la variation de la demande dans le temps et l’existence et évolution des ressource libérales. Ces modèles nous permettent de proposer des localisations robustes dans le temps tout en assurant une couverture maximale et en minimisant les coûts. La deuxième problématique consiste au choix des activités et couverture épidémiologique et spatiale en tenant compte différentes activités et types de ressources, les autorisations pour réaliser les pathologies et la couverture. Deux modèles développés nous ont permis montrer les effets sur l’affectation de la demande et le dimensionnement de ressources induits par changements dans les coûts des libéraux, salaires et d’autorisation de servir la demande. Le troisième problème et celui du dimensionnement de ressources avec incertitudes de demande (volume, épidémiologique et géographique) et le modèle proposé tient compte du problème sous-jacent de déplacement des ressources à l’aide d’une estimation de la tournée réalisée. / The issue of access to heamthcare is crucial in our modern society. One obvious effect of the augmentation of healthcare services demand is the increasing occupancy rates in hospitals. The main difference between the provision of care at the hospital and at home is as follows: the patient is at hospital and all the resources necessary for its treatment are in the same place, while in the care delivered at home, resources must be moved to the bedside. There are several challenges in order to achieve this change. In this thesis we address three important issues in the design of structures of home care. First, the location of structures minimizing logistics costs, where we develop three models with different features such as traveling costs of resources, changes in demand over time and evolution of freelance resources. These models allow us to provide robust location over time while ensuring maximum coverage and minimizing costs. The second issue is the choice of activities, epidemiological and spatial coverage, taking into account different types of activities and resources, permissions to serve some pathologies and coverage. Two models developed allow us to show the effects on the demand allocation and resources planning induced by changes in the costs of freelance professionals and authorization to serve some pathologies. The third problem is the dimensioning of resources with demand uncertainty (volume, epidemiological and geographical) and the proposed model takes into account the underlying problem of moving resources using an estimate of the routes performed.

Il y a soixante ans, le 4 octobre 1957, Spoutnik, le premier satellite artificiel conçu par l’homme, est envoyé dans l’espace. Sa seule fonction est d’émettre un bip radio à des fréquences de 20 et 40 MHz pour démontrer la puissance spatiale de l’URSS. Depuis cette époque les satellites se sont multipliés et leurs missions se sont diversifiées. Aujourd’hui, les missions des satellites sont si variées que certains quittent l’orbite terrestre. On parle dans ce cas de sondes, même si, dans le reste de cette thèse, ils seront inclus dans le terme "satellite". La mission des satellites la plus connue du grand public est la découverte de l’univers et l’exploration interplanétaire avec de célèbres satellites comme le télescope spatial international Hubble ou des sondes comme Rosetta, Voyager 1 et 2, ... Cependant de nos jours, même si l’exploration spatiale reste un enjeu majeur de l’humanité, la plupart des satellites ont des missions plus modestes qui ont pourtant un impact important sur la vie économique et politique. Les satellites en question ont aujourd’hui deux buts : la défense et le commercial. Dans les deux cas on peut diviser les satellites en deux groupes distincts : les satellites d’observation et les satellites de télécommunication. Pour fonctionner tous ces satellites utilisent un harnais électrique. Le harnais électrique regroupe tous les câbles présents dans le satellite et qui ne transportent pas de données client. Dans le cadre de cette thèse nous nous intéressons à l’optimisation de la conception du harnais électrique des satellites. / Sixty years ago, on October 4, 1957, Sputnik, the first man-made artificial satellite, was sent into space. Its only function is to emit a radio beep at frequencies of 20 and 40 MHz to demonstrate the space power of the USSR. Since then, satellites have been multiple and their missions have diversified. Today, the missions of the satellites are so varied that some leave Earth's orbit. We speak in the case of probes, even if, in the rest of this thesis, they will be included in the term "satellite". The best-known satellite mission of the general public is the discovery of the universe and interplanetary exploration with satellite satellites such as the Hubble International Space Telescope or probes such as Rosetta, Voyager 1 and 2, ... nevertheless nowadays Although space exploration remains a major issue for humanity, most satellites have smaller missions that have a significant impact on economic and political life. The satellites in question today have two goals: defense and commercial. In both cases the satellites can be divided into two distinct groups: observation satellites and telecommunication satellites. To operate all these satellites, use an electrical harness. The electrical harness includes all the cables present in the satellite and which does not carry any customer data. As part of this we are interested in optimizing the design of the electrical harness of satellites.

A estrutura tributária brasileira proporciona uma dificuldade a mais no planejamento de redes logísticas, principalmente no caso do Imposto sobre Circulação de Mercadorias e Serviços (ICMS). A omissão do fator tributário na modelagem de redes logísticas no cenário brasileiro atual poderá resultar em cenários não ótimos, devido à marcante influência da mesma na localização das instalações e na definição dos fluxos de origem e destino. Devido aos impactos da legislação brasileira na perda de competitividade das empresas brasileiras atualmente, a Reforma Tributária se faz necessária por estas e outras razões não logísticas, alterando a forma como este imposto é cobrado. Alguns estudos foram realizados para avaliar o impacto da reforma tributária no planejamento de redes logísticas, porém o presente estudo atualizou a demanda e os custos logísticos para os valores atuais de mercado, buscando aprofundar a compreensão sobre as principais modificações da rede. Além disso, procurou-se avaliar o impacto da modificação da alíquota interna e externa na configuração de redes logísticas. Para estudar o impacto do ICMS na rede logística de uma empresa, recorreu-se ao modelo de Programação Linear Inteira Mista (PLIM). Os resultados do estudo evidenciam que a reforma tributária irá trazer melhorias significativas no aspecto de fluxo origem e destino e na localização das instalações. Porém, a estruturação da rede logística ainda irá sofrer interferência do ICMS, contribuindo assim para o aumento do Custo Brasil pela permanência do turismo fiscal. / The Brazilian tax structure provides a further difficulty in the planning of logistics networks, especially in the case of the Tax on Circulation of Goods and Services. The omission of the tax factor in modeling logistics networks in the current Brazilian scenario could result in non-optimal, due to the strong influence in the same location of facilities and on defining the flows of origin and destination. Due to the impacts of the Brazilian legislation in the loss of competitiveness of Brazilian companies currently the Tax Reform is needed for these and other logistical reasons not changing the way this tax is levied. Some studies were conducted to evaluate the impact of tax reform in the planning of logistics networks, but the present study updated the demand and logistics costs for the current market values, seeking to deepen the understanding of the main changes to the network. In addition, we sought to evaluate the impact of changes of the internal rate in the configuration of logistics networks. To study the impact of the ICMS in the logistics network of a company, we used the model of Mixed Integer Linear Programming (MILP). The study results show that tax reform will bring significant improvements in the appearance of flow origin and destination and the location of facilities. However, the structuring of the logistics network will still suffer interference from the ICMS, thus contributing to the increase in Brazil cost the tax residence of tourism.

O presente trabalho tem por objetivo definir o modal de transporte ótimo para componentes comprados que abastecem a linha de produção de uma empresa aeronáutica, ao menor custo logístico. Devido a característica do material e das exigências de órgãos e entidades responsáveis pelo ambiente industrial aeronáutico, os fornecedores são especializados e de origem internacional, pois as indústrias nacionais ainda não têm a certificação exigida para produzir tais materiais, salvo algumas poucas exceções, o que onera e torna a cadeia de suprimentos muito mais complexa. Portanto, para atingir o objetivo proposto utilizou-se um modelo heurístico, que decide o modal de transporte ótimo para transportar estes materiais de sua origem, EUA ou Europa, até seu destino que é o Brasil. / The present paper aims to define an optimum modal of transportation for acquired components used to supply the production line in a Aeronautical Company, based on the lowest logistic cost. Due to the caracteristics of such material and the special requirements of the Governamental agencies as well as the specific agencies that rule the aeronautical industrial environment, the suppliers which are in its vast majority of this net are from abroad. It happens due to the fact that the local Industry do not hold the required certification to manufacture or assemble such materials which throws the costs and its complexity much more present. Some few isolated companies constitute a exception of this general frame. Based on the exposed, to achieve the proposed goal a heuristic model was used, which defines that the best modal of transportation to transport those specific materials from its origin, USA or Europe, to your final destination in Brazil.