Formulações e algoritmos para o problema das p-medianas heterogêneo livre de penalidade

This work presents a new model for the Heterogeneous p-median Problem (HPM), proposed to recover the hidden category structures present in the data provided by a sorting task procedure, a popular approach to understand heterogeneous individual’s perception of products and brands. This new model is named as the Penalty-free Heterogeneous p-median Problem (PFHPM), a single-objective version of the original problem, the HPM. The main parameter in the HPM is also eliminated, the penalty factor. It is responsible for the weighting of the objective function terms. The adjusting of this parameter controls the way that the model recovers the hidden category structures present in data, and depends on a broad knowledge of the problem. Additionally, two complementary formulations for the PFHPM are shown, both mixed integer linear programming problems. From these additional formulations lower-bounds were obtained for the PFHPM. These values were used to validate a specialized Variable Neighborhood Search (VNS) algorithm, proposed to solve the PFHPM. This algorithm provided good quality solutions for the PFHPM, solving artificial generated instances from a Monte Carlo Simulation and real data instances, even with limited computational resources. Statistical analyses presented in this work suggest that the new algorithm and model, the PFHPM, can recover more accurately the original category structures related to heterogeneous individual’s perceptions than the original model and algorithm, the HPM. Finally, an illustrative application of the PFHPM is presented, as well as some insights about some new possibilities for it, extending the new model to fuzzy environments

Practical engineering of hard spin-glass instances

Recent technological developments in the field of experimental quantum annealing have made prototypical annealing optimizers with hundreds of qubits commercially available. The experimental demonstration of a quantum speedup for optimization problems has since then become a coveted, albeit elusive goal. Recent studies have shown that the so far inconclusive results, regarding a quantum enhancement, may have been partly due to the benchmark problems used being unsuitable. In particular, these problems had inherently too simple a structure, allowing for both traditional resources and quantum annealers to solve them with no special efforts. The need therefore has arisen for the generation of harder benchmarks which would hopefully possess the discriminative power to separate classical scaling of performance with size from quantum. We introduce here a practical technique for the engineering of extremely hard spin-glass Ising-type problem instances that does not require "cherry picking" from large ensembles of randomly generated instances. We accomplish this by treating the generation of hard optimization problems itself as an optimization problem, for which we offer a heuristic algorithm that solves it. We demonstrate the genuine thermal hardness of our generated instances by examining them thermodynamically and analyzing their energy landscapes, as well as by testing the performance of various state-of-the-art algorithms on them. We argue that a proper characterization of the generated instances offers a practical, efficient way to properly benchmark experimental quantum annealers, as well as any other optimization algorithm.

Measurement resource planning:a methodology that uses quality characteristics mapping

Integration of the measurement activity into the production process is an essential rule in digital enterprise technology, especially for large volume product manufacturing, such as aerospace, shipbuilding, power generation and automotive industries. Measurement resource planning is a structured method of selecting and deploying necessary measurement resources to implement quality aims of product development. In this research, a new mapping approach for measurement resource planning is proposed. Firstly, quality aims are identified in the form of a number of specifications and engineering requirements of one quality characteristics (QCs) at a specific stage of product life cycle, and also measurement systems are classified according to the attribute of QCs. Secondly, a matrix mapping approach for measurement resource planning is outlined together with an optimization algorithm for combination between quality aims and measurement systems. Finally, the proposed methodology has been studied in shipbuilding to solve the problem of measurement resource planning, by which the measurement resources are deployed to satisfy all the quality aims. © Springer-Verlag Berlin Heidelberg 2010.

Linking Parametric CAD with Adjoint Surface Sensitivities

The goal of this work is to present an efficient CAD-based adjoint process chain for calculating parametric sensitivities (derivatives of the objective function with respect to the CAD parameters) in timescales acceptable for industrial design processes. The idea is based on linking parametric design velocities (geometric sensitivities computed from the CAD model) with adjoint surface sensitivities. A CAD-based design velocity computation method has been implemented based on distances between discrete representations of perturbed geometries. This approach differs from other methods due to the fact that it works with existing commercial CAD packages (unlike most analytical approaches) and it can cope with the changes in CAD model topology and face labeling. Use of the proposed method allows computation of parametric sensitivities using adjoint data at a computational cost which scales with the number of objective functions being considered, while it is essentially independent of the number of design variables. The gradient computation is demonstrated on test cases for a Nozzle Guide Vane (NGV) model and a Turbine Rotor Blade model. The results are validated against finite difference values and good agreement is shown. This gradient information can be passed to an optimization algorithm, which will use it to update the CAD model parameters.

Integrated Optimal Power Flow for Distribution Networks in Local and Urban Scales

This paper develops an integrated optimal power flow (OPF) tool for distribution networks in two spatial scales. In the local scale, the distribution network, the natural gas network, and the heat system are coordinated as a microgrid. In the urban scale, the impact of natural gas network is considered as constraints for the distribution network operation. The proposed approach incorporates unbalance three-phase electrical systems, natural gas systems, and combined cooling, heating, and power systems. The interactions among the above three energy systems are described by energy hub model combined with components capacity constraints. In order to efficiently accommodate the nonlinear constraint optimization problem, particle swarm optimization algorithm is employed to set the control variables in the OPF problem. Numerical studies indicate that by using the OPF method, the distribution network can be economically operated. Also, the tie-line power can be effectively managed.

Conception optimale d’une chaîne de traction électrique pour une voiture de type Formule SAE

La Formule SAE (Society of Automotive Engineers) est une compétition étudiante consistant en la conception et la fabrication d’une voiture de course monoplace. De nombreux événements sont organisés à chaque année au cours desquels plusieurs universités rivalisent entre elles lors d’épreuves dynamiques et statiques. Celles-ci comprennent l’évaluation de la conception, l’évaluation des coûts de fabrication, l’accélération de la voiture, etc. Avec plus de 500 universités participantes et des événements annuels sur tous les continents, il s’agit de la plus importante compétition d’ingénierie étudiante au monde. L’équipe ULaval Racing a participé pendant plus de 20 ans aux compétitions annuelles réservées aux voitures à combustion. Afin de s’adapter à l’électrification des transports et aux nouvelles compétitions destinées aux voitures électriques, l’équipe a conçu et fabriqué une chaîne de traction électrique haute performance destinée à leur voiture 2015. L’approche traditionnelle employée pour concevoir une motorisation électrique consiste à imposer les performances désirées. Ces critères comprennent l’inclinaison maximale que la voiture doit pouvoir gravir, l’autonomie désirée ainsi qu’un profil de vitesse en fonction du temps, ou tout simplement un cycle routier. Cette approche n’est malheureusement pas appropriée pour la conception d’une traction électrique pour une voiture de type Formule SAE. Ce véhicule n’étant pas destiné à la conduite urbaine ou à la conduite sur autoroute, les cycles routiers existants ne sont pas représentatifs des conditions d’opération du bolide à concevoir. Ainsi, la réalisation de ce projet a nécessité l’identification du cycle d’opération routier sur lequel le véhicule doit opérer. Il sert de point de départ à la conception de la chaîne de traction composée des moteurs, de la batterie ainsi que des onduleurs de tension. L’utilisation d’une méthode de dimensionnement du système basée sur un algorithme d’optimisation génétique, suivie d’une optimisation locale couplée à une analyse par éléments-finis a permis l’obtention d’une solution optimale pour les circuits de type Formule SAE. La chaîne de traction conçue a été fabriquée et intégrée dans un prototype de voiture de l’équipe ULaval Racing lors de la saison 2015 afin de participer à diverses compétitions de voitures électriques.

'Explicit Learning: an Effort towards Human Scheduling Algorithms'

Abstract Scheduling problems are generally NP-hard combinatorial problems, and a lot of research has been done to solve these problems heuristically. However, most of the previous approaches are problem-specific and research into the development of a general scheduling algorithm is still in its infancy. Mimicking the natural evolutionary process of the survival of the fittest, Genetic Algorithms (GAs) have attracted much attention in solving difficult scheduling problems in recent years. Some obstacles exist when using GAs: there is no canonical mechanism to deal with constraints, which are commonly met in most real-world scheduling problems, and small changes to a solution are difficult. To overcome both difficulties, indirect approaches have been presented (in [1] and [2]) for nurse scheduling and driver scheduling, where GAs are used by mapping the solution space, and separate decoding routines then build solutions to the original problem. In our previous indirect GAs, learning is implicit and is restricted to the efficient adjustment of weights for a set of rules that are used to construct schedules. The major limitation of those approaches is that they learn in a non-human way: like most existing construction algorithms, once the best weight combination is found, the rules used in the construction process are fixed at each iteration. However, normally a long sequence of moves is needed to construct a schedule and using fixed rules at each move is thus unreasonable and not coherent with human learning processes. When a human scheduler is working, he normally builds a schedule step by step following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not completed yet, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this research we intend to design more human-like scheduling algorithms, by using ideas derived from Bayesian Optimization Algorithms (BOA) and Learning Classifier Systems (LCS) to implement explicit learning from past solutions. BOA can be applied to learn to identify good partial solutions and to complete them by building a Bayesian network of the joint distribution of solutions [3]. A Bayesian network is a directed acyclic graph with each node corresponding to one variable, and each variable corresponding to individual rule by which a schedule will be constructed step by step. The conditional probabilities are computed according to an initial set of promising solutions. Subsequently, each new instance for each node is generated by using the corresponding conditional probabilities, until values for all nodes have been generated. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the Bayesian network is updated again using the current set of good rule strings. The algorithm thereby tries to explicitly identify and mix promising building blocks. It should be noted that for most scheduling problems the structure of the network model is known and all the variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus learning can amount to 'counting' in the case of multinomial distributions. In the LCS approach, each rule has its strength showing its current usefulness in the system, and this strength is constantly assessed [4]. To implement sophisticated learning based on previous solutions, an improved LCS-based algorithm is designed, which consists of the following three steps. The initialization step is to assign each rule at each stage a constant initial strength. Then rules are selected by using the Roulette Wheel strategy. The next step is to reinforce the strengths of the rules used in the previous solution, keeping the strength of unused rules unchanged. The selection step is to select fitter rules for the next generation. It is envisaged that the LCS part of the algorithm will be used as a hill climber to the BOA algorithm. This is exciting and ambitious research, which might provide the stepping-stone for a new class of scheduling algorithms. Data sets from nurse scheduling and mall problems will be used as test-beds. It is envisaged that once the concept has been proven successful, it will be implemented into general scheduling algorithms. It is also hoped that this research will give some preliminary answers about how to include human-like learning into scheduling algorithms and may therefore be of interest to researchers and practitioners in areas of scheduling and evolutionary computation. References 1. Aickelin, U. and Dowsland, K. (2003) 'Indirect Genetic Algorithm for a Nurse Scheduling Problem', Computer & Operational Research (in print). 2. Li, J. and Kwan, R.S.K. (2003), 'Fuzzy Genetic Algorithm for Driver Scheduling', European Journal of Operational Research 147(2): 334-344. 3. Pelikan, M., Goldberg, D. and Cantu-Paz, E. (1999) 'BOA: The Bayesian Optimization Algorithm', IlliGAL Report No 99003, University of Illinois. 4. Wilson, S. (1994) 'ZCS: A Zeroth-level Classifier System', Evolutionary Computation 2(1), pp 1-18.

'Explicit Learning: an Effort towards Human Scheduling Algorithms'

Abstract Scheduling problems are generally NP-hard combinatorial problems, and a lot of research has been done to solve these problems heuristically. However, most of the previous approaches are problem-specific and research into the development of a general scheduling algorithm is still in its infancy. Mimicking the natural evolutionary process of the survival of the fittest, Genetic Algorithms (GAs) have attracted much attention in solving difficult scheduling problems in recent years. Some obstacles exist when using GAs: there is no canonical mechanism to deal with constraints, which are commonly met in most real-world scheduling problems, and small changes to a solution are difficult. To overcome both difficulties, indirect approaches have been presented (in [1] and [2]) for nurse scheduling and driver scheduling, where GAs are used by mapping the solution space, and separate decoding routines then build solutions to the original problem. In our previous indirect GAs, learning is implicit and is restricted to the efficient adjustment of weights for a set of rules that are used to construct schedules. The major limitation of those approaches is that they learn in a non-human way: like most existing construction algorithms, once the best weight combination is found, the rules used in the construction process are fixed at each iteration. However, normally a long sequence of moves is needed to construct a schedule and using fixed rules at each move is thus unreasonable and not coherent with human learning processes. When a human scheduler is working, he normally builds a schedule step by step following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not completed yet, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this research we intend to design more human-like scheduling algorithms, by using ideas derived from Bayesian Optimization Algorithms (BOA) and Learning Classifier Systems (LCS) to implement explicit learning from past solutions. BOA can be applied to learn to identify good partial solutions and to complete them by building a Bayesian network of the joint distribution of solutions [3]. A Bayesian network is a directed acyclic graph with each node corresponding to one variable, and each variable corresponding to individual rule by which a schedule will be constructed step by step. The conditional probabilities are computed according to an initial set of promising solutions. Subsequently, each new instance for each node is generated by using the corresponding conditional probabilities, until values for all nodes have been generated. Another set of rule strings will be generated in this way, some of which will replace previous strings based on fitness selection. If stopping conditions are not met, the Bayesian network is updated again using the current set of good rule strings. The algorithm thereby tries to explicitly identify and mix promising building blocks. It should be noted that for most scheduling problems the structure of the network model is known and all the variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus learning can amount to 'counting' in the case of multinomial distributions. In the LCS approach, each rule has its strength showing its current usefulness in the system, and this strength is constantly assessed [4]. To implement sophisticated learning based on previous solutions, an improved LCS-based algorithm is designed, which consists of the following three steps. The initialization step is to assign each rule at each stage a constant initial strength. Then rules are selected by using the Roulette Wheel strategy. The next step is to reinforce the strengths of the rules used in the previous solution, keeping the strength of unused rules unchanged. The selection step is to select fitter rules for the next generation. It is envisaged that the LCS part of the algorithm will be used as a hill climber to the BOA algorithm. This is exciting and ambitious research, which might provide the stepping-stone for a new class of scheduling algorithms. Data sets from nurse scheduling and mall problems will be used as test-beds. It is envisaged that once the concept has been proven successful, it will be implemented into general scheduling algorithms. It is also hoped that this research will give some preliminary answers about how to include human-like learning into scheduling algorithms and may therefore be of interest to researchers and practitioners in areas of scheduling and evolutionary computation. References 1. Aickelin, U. and Dowsland, K. (2003) 'Indirect Genetic Algorithm for a Nurse Scheduling Problem', Computer & Operational Research (in print). 2. Li, J. and Kwan, R.S.K. (2003), 'Fuzzy Genetic Algorithm for Driver Scheduling', European Journal of Operational Research 147(2): 334-344. 3. Pelikan, M., Goldberg, D. and Cantu-Paz, E. (1999) 'BOA: The Bayesian Optimization Algorithm', IlliGAL Report No 99003, University of Illinois. 4. Wilson, S. (1994) 'ZCS: A Zeroth-level Classifier System', Evolutionary Computation 2(1), pp 1-18.

Otimização de controle de tráfego em grupo de elevadores com algoritmos bioinspirados

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Tecnologia, Departamento de Engenharia Mecânica, 2015.

BOA for nurse scheduling

Our research has shown that schedules can be built mimicking a human scheduler by using a set of rules that involve domain knowledge. This chapter presents a Bayesian Optimization Algorithm (BOA)for the nurse scheduling problem that chooses such suitable scheduling rules from a set for each nurse’s assignment. Based on the idea of using probabilistic models, the BOA builds a Bayesian network for the set of promising solutions and samples these networks to generate new candidate solutions. Computational results from 52 real data instances demonstrate the success of this approach. It is also suggested that the learning mechanism in the proposed algorithm may be suitable for other scheduling problems.

MODELING OF TRANSFER PATH FOR DETERMINATION OF COMBUSTION AND NOISE METRICS ON DIESEL ENGINES

Determination of combustion metrics for a diesel engine has the potential of providing feedback for closed-loop combustion phasing control to meet current and upcoming emission and fuel consumption regulations. This thesis focused on the estimation of combustion metrics including start of combustion (SOC), crank angle location of 50% cumulative heat release (CA50), peak pressure crank angle location (PPCL), and peak pressure amplitude (PPA), peak apparent heat release rate crank angle location (PACL), mean absolute pressure error (MAPE), and peak apparent heat release rate amplitude (PAA). In-cylinder pressure has been used in the laboratory as the primary mechanism for characterization of combustion rates and more recently in-cylinder pressure has been used in series production vehicles for feedback control. However, the intrusive measurement with the in-cylinder pressure sensor is expensive and requires special mounting process and engine structure modification. As an alternative method, this work investigated block mounted accelerometers to estimate combustion metrics in a 9L I6 diesel engine. So the transfer path between the accelerometer signal and the in-cylinder pressure signal needs to be modeled. Depending on the transfer path, the in-cylinder pressure signal and the combustion metrics can be accurately estimated - recovered from accelerometer signals. The method and applicability for determining the transfer path is critical in utilizing an accelerometer(s) for feedback. Single-input single-output (SISO) frequency response function (FRF) is the most common transfer path model; however, it is shown here to have low robustness for varying engine operating conditions. This thesis examines mechanisms to improve the robustness of FRF for combustion metrics estimation. First, an adaptation process based on the particle swarm optimization algorithm was developed and added to the single-input single-output model. Second, a multiple-input single-output (MISO) FRF model coupled with principal component analysis and an offset compensation process was investigated and applied. Improvement of the FRF robustness was achieved based on these two approaches. Furthermore a neural network as a nonlinear model of the transfer path between the accelerometer signal and the apparent heat release rate was also investigated. Transfer path between the acoustical emissions and the in-cylinder pressure signal was also investigated in this dissertation on a high pressure common rail (HPCR) 1.9L TDI diesel engine. The acoustical emissions are an important factor in the powertrain development process. In this part of the research a transfer path was developed between the two and then used to predict the engine noise level with the measured in-cylinder pressure as the input. Three methods for transfer path modeling were applied and the method based on the cepstral smoothing technique led to the most accurate results with averaged estimation errors of 2 dBA and a root mean square error of 1.5dBA. Finally, a linear model for engine noise level estimation was proposed with the in-cylinder pressure signal and the engine speed as components.

Vehicle routing and location routing problems with minimum latency: algorithms and models

Latency can be defined as the sum of the arrival times at the customers. Minimum latency problems are specially relevant in applications related to humanitarian logistics. This thesis presents algorithms for solving a family of vehicle routing problems with minimum latency. First the latency location routing problem (LLRP) is considered. It consists of determining the subset of depots to be opened, and the routes that a set of homogeneous capacitated vehicles must perform in order to visit a set of customers such that the sum of the demands of the customers assigned to each vehicle does not exceed the capacity of the vehicle. For solving this problem three metaheuristic algorithms combining simulated annealing and variable neighborhood descent, and an iterated local search (ILS) algorithm, are proposed. Furthermore, the multi-depot cumulative capacitated vehicle routing problem (MDCCVRP) and the multi-depot k-traveling repairman problem (MDk-TRP) are solved with the proposed ILS algorithm. The MDCCVRP is a special case of the LLRP in which all the depots can be opened, and the MDk-TRP is a special case of the MDCCVRP in which the capacity constraints are relaxed. Finally, a LLRP with stochastic travel times is studied. A two-stage stochastic programming model and a variable neighborhood search algorithm are proposed for solving the problem. Furthermore a sampling method is developed for tackling instances with an infinite number of scenarios. Extensive computational experiments show that the proposed methods are effective for solving the problems under study.

Optimization of cruciform specimens for biaxial fatigue loading with direct multi search

In order to correctly assess the biaxial fatigue material properties one must experimentally test different load conditions and stress levels. With the rise of new in-plane biaxial fatigue testing machines, using smaller and more efficient electrical motors, instead of the conventional hydraulic machines, it is necessary to reduce the specimen size and to ensure that the specimen geometry is appropriate for the load capacity installed. At the present time there are no standard specimen's geometries and the indications on literature how to design an efficient test specimen are insufficient. The main goal of this paper is to present the methodology on how to obtain an optimal cruciform specimen geometry, with thickness reduction in the gauge area, appropriate for fatigue crack initiation, as a function of the base material sheet thickness used to build the specimen. The geometry is optimized for maximum stress using several parameters, ensuring that in the gauge area the stress distributions on the loading directions are uniform and maximum with two limit phase shift loading conditions (delta = 0 degrees and (delta = 180 degrees). Therefore the fatigue damage will always initiate on the center of the specimen, avoiding failure outside this region. Using the Renard Series of preferred numbers for the base material sheet thickness as a reference, the reaming geometry parameters are optimized using a derivative-free methodology, called direct multi search (DMS) method. The final optimal geometry as a function of the base material sheet thickness is proposed, as a guide line for cruciform specimens design, and as a possible contribution for a future standard on in-plane biaxial fatigue tests

Efficient robust optimization based on polynomial chaos and tabu search algorithm

The study of robust design methodologies and techniques has become a new topical area in design optimizations in nearly all engineering and applied science disciplines in the last 10 years due to inevitable and unavoidable imprecision or uncertainty which is existed in real word design problems. To develop a fast optimizer for robust designs, a methodology based on polynomial chaos and tabu search algorithm is proposed. In the methodology, the polynomial chaos is employed as a stochastic response surface model of the objective function to efficiently evaluate the robust performance parameter while a mechanism to assign expected fitness only to promising solutions is introduced in tabu search algorithm to minimize the requirement for determining robust metrics of intermediate solutions. The proposed methodology is applied to the robust design of a practical inverse problem with satisfactory results.

Multi-Objective Optimization Design of Tanker Ships via a Genetic Algorithm

The cost of a new ship design heavily depends on the principal dimensions of the ship; however, dimensions minimization often conflicts with the minimum oil outflow (in the event of an accidental spill). This study demonstrates one rational methodology for selecting the optimal dimensions and coefficients of form of tankers via the use of a genetic algorithm. Therein, a multi-objective optimization problem was formulated by using two objective attributes in the evaluation of each design, specifically, total cost and mean oil outflow. In addition, a procedure that can be used to balance the designs in terms of weight and useful space is proposed. A genetic algorithm was implemented to search for optimal design parameters and to identify the nondominated Pareto frontier. At the end of this study, three real ships are used as case studies. [DOI:10.1115/1.4002740]