Abstract

Since the emergence of the Internet there has been continuous evolution in interface design, with designers trying to offer a good experience to users. Website developers regularly turn to heuristic indicators that have been collected throughout years of research and praxis. Nevertheless, these indicators are diverse and it is complicated locating which ones are applicable in every case. The objective of this work is to identify and analyze the most scientifically and geographically relevant heuristic indicators through a mapping review, in order to classify them according to formal and content components. The results are two taxonomies which make it easier to identify the location, and analyze and apply the guidelines.

Abstract

We present two heuristics based on constraint technology that solve the problem of generating air traffic management contingency plans, which are used in the case of a catastrophic infrastructure failure within EUROCONTROL, the European Organisation for the Safety of Air Navigation. Of the heuristics presented, one is based on constraint-based local search and tabu search, and the other one is a constraint programming and large neighbourhood search hybrid algorithm. The heuristics show that it is feasible to automate the development of contingency plans, which is currently done by human experts; this is desirable for several reasons, for example it would allow the contingency plans to be generated with an increased frequency. The generated plans were evaluated, by EUROCONTROL, to be at least as good as the human-made ones. Document type: Part of book or chapter of book

Abstract

In this report, we study the problem of optimizing the throughput of applications for heterogeneous platforms subject to failures. The considered applications are composed of a sequence of consecutive tasks linked as a linear graph (pipeline), with a type associated to each task. The challenge is to specialize the machines of a target platform to process only one task type, given that every machine is able to process all the types before being specialized, to avoid costly context or setup changes. Each instance can thus be performed by any machine specialized in its type and the workload of the system can be shared among a set of specialized machines. For identical machines, we prove that an optimal solution can be computed in polynomial time. However, the problem becomes NP-hard when two machines can compute the same task type at different speeds. Several polynomial time heuristics are presented for the most realistic specialized settings. Experimental results show that the best heuristics obtain a good throughput, much better than the throughput obtained with a random mapping, and close to the optimal throughput in the particular cases on which the optimal throughput can be computed.; Dans ce rapport, nous étudions le problème de l'optimisation du débit d'applications de type pipeline dans un environnement hétérogène sujet à des pannes. Les applications sont constituées d'un ensemble de tâches consécu\\-tives typées et organisées selon une chaîne linéaire ou pipeline. Le but est ici de spécialiser les machines de la plate-forme d'exécution afin qu'elles ne traitent qu'un seul type de tâches, sachant qu'au départ elles peuvent exécuter tous les types. Cela permet d'éviter des reconfigurations coûteuses entre tâches de types différents sur une même machine. Ainsi, les instances d'une même tâche peuvent être distribuées sur plusieurs machines spécialisées pour le type de cette tâche, ce qui permet une répartition de la charge du système sur un ensemble de machines spécialisées. Lorsque la plate-forme est composée de machines identiques, nous prouvons qu'une solution optimale peut être trouvée en temps polynomial. Par contre, le problème devient NP-complet dès lors que deux machines peuvent traiter une même tâche à des vitesses différentes. Ce faisant, plusieurs heuristiques sont présentées dans le cas le plus réaliste d'un système spécialisé. Les résultats expérimentaux montrent que les meilleures heuristiques obtiennent de bons résultats en terme de débit, meilleurs qu'avec une allocation aléatoire, et que les débits atteints sont très proches des débits optimaux dans les cas particuliers pour lesquels une solution optimale peut être calculée. Document type: Part of book or chapter of book

Abstract

This paper presents a traffic engineering framework able to optimize OSPF weight setting administrative procedures. Using the proposed framework, enhanced OSPF configurations are now provided to network administrators in order to effectively improve the QoS performance of the corresponding network domain. The envisaged NP-hard optimization problem is faced resorting to Evolutionary Algorithms, which allocate OSPF weights guided by a bi-objective function. The results presented in this work show that the proposed optimization tool clearly outperforms common weight setting heuristics. Document type: Part of book or chapter of book

Abstract

We consider a stochastic routing model in which the goal is to find the optimal route that incorporates a measure of risk The problem arises in traffic engineering, transportation and even more abstract settings such as task planning (where the time to execute tasks is uncertain), etc The stochasticity is specified in terms of arbitrary edge length distributions with given mean and variance values in a graph The objective function is a positive linear combination of the mean and standard deviation of the route Both the nonconvex objective and exponentially sized feasible set of available routes present a challenging optimization problem for which no efficient algorithms are known In this paper we evaluate the practical performance of algorithms and heuristic approaches which show very promising results in terms of both running time and solution accuracy. Document type: Part of book or chapter of book

Abstract

Recently, the quality and the diversity of transport services are more and more required. Moreover, in case of a great deal of services and selling goods, a significant part of price is transport cost. Thus, the design of models and applications which make possible efficient transport planning and scheduling becomes important. A great deal of real transport problems may be modelled by using Pickup and Delivery Problem with Time Windows (PDPTW) and capacity constraints, which is based on the realization of a set of transport requests by a fleet of vehicles with given capacities. Each request is described by pickup and delivery locations, time periods when pickup and delivery operations should be performed and needed load. Application of evolutionary approach has brought good results in case of another, simpler transport problem – the Vehicle Routing Problem with Time Windows (VRPTW). This paper is aimed at proposing a straightforward extension of VRPTW based heuristics for the PDPTW. Document type: Part of book or chapter of book

Abstract

International audience; To sustain the rapidly increasing air traffic demand, the future air traffic management system will rely on a concept, called Trajectory-Based Operations (TBO), that will require aircraft to follow an assigned 4D trajectory (time-constrained trajectory) with high precision. TBO involves separating aircraft via strategic (long-term) trajectory deconfliction rather than the currently-practicing tactical (short-term) conflict resolution. In this context, this chapter presents a strategic trajectory planning approach aiming at minimizing the number of conflicts between aircraft trajectories for a given day. The proposed methodology allocates an alternative departure time, a horizontal flight path, and a flight level to each aircraft at a nation-wide scale.In real-life situations, aircraft may arrive at a given position with some uncertainties on its curvilinear abscissa due to external events. To ensure robustness of the strategic trajectory plan, the aircraft arrival time to any given position will be represented here by a probabilistic distribution over its nominal assigned arrival time.The proposed approach optimizes the 4D trajectory of each aircraft so as to minimize the probability of potential conflicts between trajectories. A hybrid-metaheuristic optimization algorithm has been developed to solve this large-scale mixed-variable optimization problem. The algorithm is implemented and tested with real air traffic data taking into account uncertainty over the French airspace for which a conflict-free and robust 4D trajectory plan is produced.

Abstract

In this thesis we study models and algorithms for the optimal design of bus routes in urban public transportation systems. The problem known as TNDP (Transit Network Design Problem) consists in determining the number and itinerary of public transportation lines and their corresponding frequencies, in terms of a given infrastructure of streets and stops. The solutions should satisfy a given origin-destination demand and should take into account the interests of users and operators and a given set of physical, policy and budgetary constraints. We propose an explicit mixed integer linear programming formulation which incorporates the waiting time and the existence of multiple lines in the behavior of the passengers.Then, we discuss the impact in the structure of the model of adding transfer, infrastructure and bus capacity constraints. We apply the model (using a standard solver) to very small test cases as well as to a real one, related to a small-sized city comprising 13 bus lines. In order to deal with cases of larger sizes, we propose a greedy constructive algorithm that produces a set of routes that are convenient for both users and operators, taking into account constraints related to transfers. By using a real test case, we show that the proposed algorithm improves results from the state of the art. As a further extension, we represent the existence of the conflicting objectives of users and operators using a multi-objective combinatorial optimization model for the TNDP. This new model is solved by a metaheuristic that exploits the multi-objective nature of the problem in order to solve it eficiently. By using a benchmark test case and a real one, we show that the proposed algorithm improves results from the state of the art and produces solutions with characteristics comparable to the real one. Objective values of both constructive and metaheuristic algorithms are compared with values corresponding to reference solutions; for the first one we compare against optimal solutions obtained with the mathematical formulation, while for the second one we compare with the solution operating the public transportation system of the city corresponding to the real test case. Finally we discuss the relationships between the diferent contributions of this thesis and we comment several issues related to the application of the proposed methodologies to real cases. We also give some opinions and recommendations concerning future developments in this research field. En esta tesis se estudian modelos y algoritmos para el diseño óptimo de recorridos de buses en sistemas de transporte público urbano colectivo. El problema conocido como TNDP (Transit Network Design Problem) consiste en determinar el número y el itinerario de líneas de transporte público y sus correspondientes frecuencias, en términos de una infraestructura dada de calles y paradas. Las soluciones deben satisfacer una demanda origen-destino dada y deben tener en cuenta los intereses de los usuarios y de los operadores y un conjunto dado de restricciones físicas, políticas y de presupuesto. Se propone una formulación explícita de programación lineal entera mixta, que incorpora el tiempo de espera y la existencia de múltiples líneas en el comportamiento de los pasajeros. Seguidamente se discute el impacto en la estructura del modelo, al agregar restricciones de transbordos y de capacidad de la infraestructura y de los buses. El modelo se aplica (usando un solver estándar) a casos de prueba muy pequeños, así como a uno real relativo a una ciudad pequeña que consta de 13 líneas de buses. Con el propósito de atacar casos de mayor tamaño, se propone un algoritmo constructivo ávido que produce un conjunto de recorridos que son convenientes tanto para los usuarios como para los operadores, teniendo en cuenta restricciones de transbordos. Utilizando un caso de prueba real, se muestra que el algoritmo propuesto mejora resultados del estado del arte. Como una extensión del algoritmo constructivo, se representa la existencia de los objetivos en conflicto de usuarios y operadores usando un modelo de optimización combinatoria multi-objetivo para el TNDP. Este nuevo modelo se resuelve con una metaheurística que explota la naturaleza multi-objetivo del problema para resolverlo eficientemente. Utilizando un caso de prueba de referencia existente en la literatura y uno real, se muestra que el algoritmo propuesto mejora resultados del estado del arte y produce soluciones de características comparables a las del sistema real. Los valores objetivo del algoritmo constructivo y de la metaheurística se comparan con valores correspondientes a soluciones de referencia; en el primer caso se compara contra soluciones óptimas obtenidas con la formulación matemática, mientras que para el segundo se compara contra la solución que opera el sistema de transporte público de la ciudad correspondiente al caso de prueba real. Finalmente se discuten las relaciones entre las diferentes contribuciones de esta tesis y se comentan varias cuestiones relacionadas a la aplicación de las metodologías propuestas a casos reales. También se formulan algunas opiniones y recomendaciones en relación a futuros desarrollos de éste tópico de investigación.

Abstract

There is substantial evidence to indicate that route choice in urban areas is complex cognitive process, conducted under uncertainty and formed on partial perspectives. Yet, conventional route choice models continue make simplistic assumptions around the nature of human cognitive ability, memory and preference. In this paper, a novel framework for route choice in urban areas is introduced, aiming to more accurately reflect the uncertain, bounded nature of route choice decision making. Two main advances are introduced. The first involves the definition of a hierarchical model of space representing the relationship between urban features and human cognition, combining findings from both the extensive previous literature on spatial cognition and a large route choice dataset. The second advance involves the development of heuristic rules for route choice decisions, building upon the hierarchical model of urban space. The heuristics describe the process by which quick, ‘good enough’ decisions are made when individuals are faced with uncertainty. This element of the model is once more constructed and parameterised according to findings from prior research and the trends identified within a large routing dataset. The paper outlines the implementation of the framework within a real-world context, validating the results against observed behaviours. Conclusions are offered as to the extension and improvement of this approach, outlining its potential as an alternative to other route choice modelling frameworks. Document type: Article

Abstract

utomatic Identification System (AIS) is primarily used as a tracking system for ships, but with the launch of satellites to collect these data, new and previously untested possibilities are emerging. This paper presents the development of heuristics for establishing the specific ship type using information retrieved from AIS data alone. These heuristics expand the possibilities of AIS data, as the specific ship type is vital for several transportation research cases, such as emission analyses of ship traffic and studies on slow steaming. The presented method for developing heuristics can be used for a wider range of vessels. These heuristics may form the basis of large-scale studies on ship traffic using AIS data when it is not feasible or desirable to use commercial ship data registers. The Transnav - International Journal on Marine Navigation and Safety of Sea Transportation articles are distributed under the terms of the Creative Commons Attribution License (CC BY-NC), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. For licence details please see http://creativecommons.org/licenses/by-nc/3.0 Document type: Article