Bayesian Optimisation Algorithm for Nurse Scheduling, Scalable Optimization via Probabilistic Modeling

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.

Dynamic Design of Piezoelectric Laminated Sensors and Actuators using Topology Optimization

Sensors and actuators based on piezoelectric plates have shown increasing demand in the field of smart structures, including the development of actuators for cooling and fluid-pumping applications and transducers for novel energy-harvesting devices. This project involves the development of a topology optimization formulation for dynamic design of piezoelectric laminated plates aiming at piezoelectric sensors, actuators and energy-harvesting applications. It distributes piezoelectric material over a metallic plate in order to achieve a desired dynamic behavior with specified resonance frequencies, modes, and enhanced electromechanical coupling factor (EMCC). The finite element employs a piezoelectric plate based on the MITC formulation, which is reliable, efficient and avoids the shear locking problem. The topology optimization formulation is based on the PEMAP-P model combined with the RAMP model, where the design variables are the pseudo-densities that describe the amount of piezoelectric material at each finite element and its polarization sign. The design problem formulated aims at designing simultaneously an eigenshape, i.e., maximizing and minimizing vibration amplitudes at certain points of the structure in a given eigenmode, while tuning the eigenvalue to a desired value and also maximizing its EMCC, so that the energy conversion is maximized for that mode. The optimization problem is solved by using sequential linear programming. Through this formulation, a design with enhancing energy conversion in the low-frequency spectrum is obtained, by minimizing a set of first eigenvalues, enhancing their corresponding eigenshapes while maximizing their EMCCs, which can be considered an approach to the design of energy-harvesting devices. The implementation of the topology optimization algorithm and some results are presented to illustrate the method.

Topology optimization for designing strain-gauge load cells

Load cells are used extensively in engineering fields. This paper describes a novel structural optimization method for single- and multi-axis load cell structures. First, we briefly explain the topology optimization method that uses the solid isotropic material with penalization (SIMP) method. Next, we clarify the mechanical requirements and design specifications of the single- and multi-axis load cell structures, which are formulated as an objective function. In the case of multi-axis load cell structures, a methodology based on singular value decomposition is used. The sensitivities of the objective function with respect to the design variables are then formulated. On the basis of these formulations, an optimization algorithm is constructed using finite element methods and the method of moving asymptotes (MMA). Finally, we examine the characteristics of the optimization formulations and the resultant optimal configurations. We confirm the usefulness of our proposed methodology for the optimization of single- and multi-axis load cell structures.

Toward Optimal Design of Piezoelectric Transducers Based on Multifunctional and Smoothly Graded Hybrid Material Systems

This work explores the design of piezoelectric transducers based on functional material gradation, here named functionally graded piezoelectric transducer (FGPT). Depending on the applications, FGPTs must achieve several goals, which are essentially related to the transducer resonance frequency, vibration modes, and excitation strength at specific resonance frequencies. Several approaches can be used to achieve these goals; however, this work focuses on finding the optimal material gradation of FGPTs by means of topology optimization. Three objective functions are proposed: (i) to obtain the FGPT optimal material gradation for maximizing specified resonance frequencies; (ii) to design piezoelectric resonators, thus, the optimal material gradation is found for achieving desirable eigenvalues and eigenmodes; and (iii) to find the optimal material distribution of FGPTs, which maximizes specified excitation strength. To track the desirable vibration mode, a mode-tracking method utilizing the `modal assurance criterion` is applied. The continuous change of piezoelectric, dielectric, and elastic properties is achieved by using the graded finite element concept. The optimization algorithm is constructed based on sequential linear programming, and the concept of continuum approximation of material distribution. To illustrate the method, 2D FGPTs are designed for each objective function. In addition, the FGPT performance is compared with the non-FGPT one.

Piezoresistive sensor design using topology optimization

Piezoresistive materials, materials whose resistivity properties change when subjected to mechanical stresses, are widely utilized in many industries as sensors, including pressure sensors, accelerometers, inclinometers, and load cells. Basic piezoresistive sensors consist of piezoresistive devices bonded to a flexible structure, such as a cantilever or a membrane, where the flexible structure transmits pressure, force, or inertial force due to acceleration, thereby causing a stress that changes the resistivity of the piezoresistive devices. By applying a voltage to a piezoresistive device, its resistivity can be measured and correlated with the amplitude of an applied pressure or force. The performance of a piezoresistive sensor is closely related to the design of its flexible structure. In this research, we propose a generic topology optimization formulation for the design of piezoresistive sensors where the primary aim is high response. First, the concept of topology optimization is briefly discussed. Next, design requirements are clarified, and corresponding objective functions and the optimization problem are formulated. An optimization algorithm is constructed based on these formulations. Finally, several design examples of piezoresistive sensors are presented to confirm the usefulness of the proposed method.

Coarse-grid higher order finite-difference time-domain algorithm with low dispersion errors

Higher order (2,4) FDTD schemes used for numerical solutions of Maxwell`s equations are focused on diminishing the truncation errors caused by the Taylor series expansion of the spatial derivatives. These schemes use a larger computational stencil, which generally makes use of the two constant coefficients, C-1 and C-2, for the four-point central-difference operators. In this paper we propose a novel way to diminish these truncation errors, in order to obtain more accurate numerical solutions of Maxwell`s equations. For such purpose, we present a method to individually optimize the pair of coefficients, C-1 and C-2, based on any desired grid size resolution and size of time step. Particularly, we are interested in using coarser grid discretizations to be able to simulate electrically large domains. The results of our optimization algorithm show a significant reduction in dispersion error and numerical anisotropy for all modeled grid size resolutions. Numerical simulations of free-space propagation verifies the very promising theoretical results. The model is also shown to perform well in more complex, realistic scenarios.

Optimal location of the workpiece in a PKM-based machining robotic cell

Most machining tasks require high accuracy and are carried out by dedicated machine-tools. On the other hand, traditional robots are flexible and easy to program, but they are rather inaccurate for certain tasks. Parallel kinematic robots could combine the accuracy and flexibility that are usually needed in machining operations. Achieving this goal requires proper design of the parallel robot. In this chapter, a multi-objective particle swarm optimization algorithm is used to optimize the structure of a parallel robot according to specific criteria. Afterwards, for a chosen optimal structure, the best location of the workpiece with respect to the robot, in a machining robotic cell, is analyzed based on the power consumed by the manipulator during the machining process.

Hyperspectral unmixing based on mixtures of Dirichlet components

This paper introduces a new unsupervised hyperspectral unmixing method conceived to linear but highly mixed hyperspectral data sets, in which the simplex of minimum volume, usually estimated by the purely geometrically based algorithms, is far way from the true simplex associated with the endmembers. The proposed method, an extension of our previous studies, resorts to the statistical framework. The abundance fraction prior is a mixture of Dirichlet densities, thus automatically enforcing the constraints on the abundance fractions imposed by the acquisition process, namely, nonnegativity and sum-to-one. A cyclic minimization algorithm is developed where the following are observed: 1) The number of Dirichlet modes is inferred based on the minimum description length principle; 2) a generalized expectation maximization algorithm is derived to infer the model parameters; and 3) a sequence of augmented Lagrangian-based optimizations is used to compute the signatures of the endmembers. Experiments on simulated and real data are presented to show the effectiveness of the proposed algorithm in unmixing problems beyond the reach of the geometrically based state-of-the-art competitors.

Particle swarm design of digital circuits

Swarm Intelligence (SI) is the property of a system whereby the collective behaviors of (unsophisticated) agents interacting locally with their environment cause coherent functional global patterns to emerge. Particle swarm optimization (PSO) is a form of SI, and a population-based search algorithm that is initialized with a population of random solutions, called particles. These particles are flying through hyperspace and have two essential reasoning capabilities: their memory of their own best position and knowledge of the swarm's best position. In a PSO scheme each particle flies through the search space with a velocity that is adjusted dynamically according with its historical behavior. Therefore, the particles have a tendency to fly towards the best search area along the search process. This work proposes a PSO based algorithm for logic circuit synthesis. The results show the statistical characteristics of this algorithm with respect to number of generations required to achieve the solutions. It is also presented a comparison with other two Evolutionary Algorithms, namely Genetic and Memetic Algorithms.

A Clinical Support System Based on Quality of Life Estimation

Quality of life is a concept influenced by social, economic, psychological, spiritual or medical state factors. More specifically, the perceived quality of an individual's daily life is an assessment of their well-being or lack of it. In this context, information technologies may help on the management of services for healthcare of chronic patients such as estimating the patient quality of life and helping the medical staff to take appropriate measures to increase each patient quality of life. This paper describes a Quality of Life estimation system developed using information technologies and the application of data mining algorithms to access the information of clinical data of patients with cancer from Otorhinolaryngology and Head and Neck services of an oncology institution. The system was evaluated with a sample composed of 3013 patients. The results achieved show that there are variables that may be significant predictors for the Quality of Life of the patient: years of smoking (p value 0.049) and size of the tumor (p value < 0.001). In order to assign the variables to the classification of the quality of life the best accuracy was obtained by applying the John Platt's sequential minimal optimization algorithm for training a support vector classifier. In conclusion data mining techniques allow having access to patients additional information helping the physicians to be able to know the quality of life and produce a well-informed clinical decision.

Testing Nelder-Mead Based Repulsion Algorithms for Multiple Roots of Nonlinear Systems via a Two-Level Factorial Design of Experiments

This paper addresses the challenging task of computing multiple roots of a system of nonlinear equations. A repulsion algorithm that invokes the Nelder-Mead (N-M) local search method and uses a penalty-type merit function based on the error function, known as 'erf', is presented. In the N-M algorithm context, different strategies are proposed to enhance the quality of the solutions and improve the overall efficiency. The main goal of this paper is to use a two-level factorial design of experiments to analyze the statistical significance of the observed differences in selected performance criteria produced when testing different strategies in the N-M based repulsion algorithm. The main goal of this paper is to use a two-level factorial design of experiments to analyze the statistical significance of the observed differences in selected performance criteria produced when testing different strategies in the N-M based repulsion algorithm.

Modern optimization in earthwork construction

Earthworks tasks are often regarded in transportation projects as some of the most demanding processes. In fact, sequential tasks such as excavation, transportation, spreading and compaction are strongly based on heavy mechanical equipment and repetitive processes, thus becoming as economically demanding as they are time-consuming. Moreover, actual construction requirements originate higher demands for productivity and safety in earthwork constructions. Given the percentual weight of costs and duration of earthworks in infrastructure construction, the optimal usage of every resource in these tasks is paramount. Considering the characteristics of an earthwork construction, it can be looked at as a production line based on resources (mechanical equipment) and dependency relations between sequential tasks, hence being susceptible to optimization. Up to the present, the steady development of Information Technology areas, such as databases, artificial intelligence and operations research, has resulted in the emergence of several technologies with potential application bearing that purpose in mind. Among these, modern optimization methods (also known as metaheuristics), such as evolutionary computation, have the potential to find high quality optimal solutions with a reasonable use of computational resources. In this context, this work describes an optimization algorithm for earthworks equipment allocation based on a modern optimization approach, which takes advantage of the concept that an earthwork construction can be regarded as a production line.

An innovative framework for probabilistic-based structural assessment with an application to existing reinforced concrete structures

A novel framework for probabilistic-based structural assessment of existing structures, which combines model identification and reliability assessment procedures, considering in an objective way different sources of uncertainty, is presented in this paper. A short description of structural assessment applications, provided in literature, is initially given. Then, the developed model identification procedure, supported in a robust optimization algorithm, is presented. Special attention is given to both experimental and numerical errors, to be considered in this algorithm convergence criterion. An updated numerical model is obtained from this process. The reliability assessment procedure, which considers a probabilistic model for the structure in analysis, is then introduced, incorporating the results of the model identification procedure. The developed model is then updated, as new data is acquired, through a Bayesian inference algorithm, explicitly addressing statistical uncertainty. Finally, the developed framework is validated with a set of reinforced concrete beams, which were loaded up to failure in laboratory.

Semi-supervised segmentation of ultrasound images based on patch representation and continuous min cut.

Ultrasound segmentation is a challenging problem due to the inherent speckle and some artifacts like shadows, attenuation and signal dropout. Existing methods need to include strong priors like shape priors or analytical intensity models to succeed in the segmentation. However, such priors tend to limit these methods to a specific target or imaging settings, and they are not always applicable to pathological cases. This work introduces a semi-supervised segmentation framework for ultrasound imaging that alleviates the limitation of fully automatic segmentation, that is, it is applicable to any kind of target and imaging settings. Our methodology uses a graph of image patches to represent the ultrasound image and user-assisted initialization with labels, which acts as soft priors. The segmentation problem is formulated as a continuous minimum cut problem and solved with an efficient optimization algorithm. We validate our segmentation framework on clinical ultrasound imaging (prostate, fetus, and tumors of the liver and eye). We obtain high similarity agreement with the ground truth provided by medical expert delineations in all applications (94% DICE values in average) and the proposed algorithm performs favorably with the literature.

A Monte Carlo-based procedure for independent monitor unit calculation in IMRT treatment plans.

Intensity-modulated radiotherapy (IMRT) treatment plan verification by comparison with measured data requires having access to the linear accelerator and is time consuming. In this paper, we propose a method for monitor unit (MU) calculation and plan comparison for step and shoot IMRT based on the Monte Carlo code EGSnrc/BEAMnrc. The beamlets of an IMRT treatment plan are individually simulated using Monte Carlo and converted into absorbed dose to water per MU. The dose of the whole treatment can be expressed through a linear matrix equation of the MU and dose per MU of every beamlet. Due to the positivity of the absorbed dose and MU values, this equation is solved for the MU values using a non-negative least-squares fit optimization algorithm (NNLS). The Monte Carlo plan is formed by multiplying the Monte Carlo absorbed dose to water per MU with the Monte Carlo/NNLS MU. Several treatment plan localizations calculated with a commercial treatment planning system (TPS) are compared with the proposed method for validation. The Monte Carlo/NNLS MUs are close to the ones calculated by the TPS and lead to a treatment dose distribution which is clinically equivalent to the one calculated by the TPS. This procedure can be used as an IMRT QA and further development could allow this technique to be used for other radiotherapy techniques like tomotherapy or volumetric modulated arc therapy.