Modelling a reliability system governed by discrete phase-type distributions

We present an n-system with one online unit and the others in cold standby. There is a repairman. When the online fails it goes to repair, and instantaneously a standby unit becomes the online one. The operational and repair times follow discrete phase-type distributions. Given that any discrete distribution defined on the positive integers is a discrete phase-type distribution, the system can be considered a general one. A model with unlimited number of units is considered for approximating a system with a great number of units. We show that the process that governs the system is a quasi-birth-and-death process. For this system, performance reliability measures; the up and down periods, and the involved costs are calculated in a matrix and algorithmic form. We show that the discrete case is not a trivial case of the continuous one. The results given in this paper have been implemented computationally with Matlab.     

Image identification system based on an optical broadcast neural network processor

We describe the implementation of a vision system based on a hardware neural processor. The architecture of the neural network processor has been designed to exploit the computational characteristics of electronics and the communication characteristics of optics in an optimal manner, thus it is based on an optical broadcast of input signals to a dense array of processing elements. The vision system has been built by use of a prototype implementation of a neural network processor with discrete optic and optoelectronic devices. It has been adapted to work as a Hamming classifier of the images taken with a 128 x 128 complementary metal-oxide semiconductor image sensor. Its results, performance characteristics of the image classification system, and an analysis of its scalability in size and speed, with the improvement of the optoelectronic neural processor, are presented.     

Batching and scheduling to minimize the makespan in the two-machine flowshop

In this paper, we consider a class of batching and scheduling problems in the two-machine flowshop where one of the machines is a discrete processor and the other one is a batch processor. The jobs are processed separately on the discrete processor and processed in batches on the batch processor. The processing time of a batch is equal to the total processing time of the jobs contained in it, and the completion time of a job in a batch is defined as the completion time of the batch containing it. A constant setup time is incurred whenever a batch is formed on the batch processor. The problem is to find the optimal batch compositions and the optimal schedule of the batches so that the makespan is minimized. All problems in this class are shown to be NP-complete in the ordinary sense. We also identify some polynomially solvable cases by introducing their corresponding solution methods.     

High-capacity neural networks on nonideal hardware

We present a new training-out algorithm for neural networks that permits good performance on nonideal hardware with limited analog neuron and weight accuracy. Optical neural networks are emphasized with the error sources including nonuniform beam illumination and nonlinear device characteristics. We compensate for processor nonidealities during gated learning (off-line training); thus our algorithm does not require real-time neural networks with adaptive weights. This permits use of high-accuracy nonadaptive weights and reduced hardware complexity. The specific neural network we consider is the Ho-Kashyap associative processor because it provides the largest storage capacity. Simulation results and optical laboratory data are provided. The storage measure we use is the ratio M/N of the number of vectors stored (M) to the dimensionality of the vectors stored (N). We show a storage capacity of M/N = 1.5 on our optical laboratory system with excellent recall accuracy, > 95%. The theoretical maximum storage is M/N = 2 (as N approaches infinity), and thus the storage and performance we demonstrate are impressive considering the processor nonidealities we present. Our techniques can be applied to other neural network algorithms and other nonideal processing hardware.     

Flow shop scheduling with a batch processor and limited buffer

This paper addresses flow shop scheduling problem with a batch processor followed by a discrete processor. Incompatible job families and limited buffer size are considered, and the objective is to determine a schedule such that the total completion time is minimised. Flexible buffer service policy is designed, and a greedy heuristic together with the worst-case analysis is developed. We also propose a hybrid method involving a Differential Evolution algorithm. Moreover, two tight lower bounds are provided to measure the performances of the proposed algorithms. Numerical results demonstrate that the proposed algorithms are capable of providing high-quality solutions for large-scale problems within a reasonable computational time.     

A distributed measurement architecture for industrial applications

In this paper, a distributed digital measurement architecture for industrial applications is proposed. The architecture is arranged on three hierarchical communication levels: the fieldbus, the intranet, and the Internet. Particular attention has been paid to the lower level, the field level, implemented using a low-priced smart front-end. It is based on the H8/3048F Hitachi microcontroller and embodies a fieldbus interface (I/F). The same board can be linked to a VXIbus controller by means of a suitable register-based interface. The proposed network can embody a number of analog signal conditioning circuits, processor, and communication capabilities, to meet the industrial needs. We propose two applications of this distributed measurement architecture: the monitoring of power quality in an electrical distribution network and the management of a water distribution system. Experimental results showing the system performance are also included in the paper.     

Optimisation of flow-shop scheduling with batch processor and limited buffer

This paper deals with a flow-shop scheduling problem with limited intermediate buffer. Jobs are grouped in incompatible job families. Each job has to be processed by a batch processor followed by a discrete processor in the same order. The batch processor can process several jobs simultaneously so that all jobs of the same batch start and complete together. We assume that the capacity of batch processor is bounded. The batch processing time is identical for batches of the same family. A batch which has completed processing on the batch processor may block the processor until there is a free unit in the buffer. The objective is to determine a batching and scheduling for all jobs so as to minimise mean completion time. A lower bound and two heuristics algorithm are developed. Moreover, a two-stage method embedded with a Differential Evolution (DE) algorithm is also developed. DE is one of the latest evolutionary computation algorithms, which implements mutation, crossover, and selection operators to improve the candidate solutions iteratively. Three variants of DE are first compared with a continuous Genetic Algorithm employing the random key representation. Then, one variant of the DE with the best convergence speed is selected. Numerical experiments are conducted to evaluate the performances of the selected two-stage meta-heuristic and two heuristics.     

Simulation of Wave Processes under Conditions of Friction of Rough Surfaces (Review)

We show that, under conditions of friction, there arise heat, deformation, and acoustic waves. We also describe the nature of their onset and analyze the statistical distribution of regular and irregular roughness on an actual discrete contact surface for different levels of structure and dimensions. We substantiate the possibility of development of new heat-dynamic models of temperature flashes with regard for the periodic discreteness of actual friction surfaces.     

On Generalized Fibonacci Cubes and Unitary Transforms

 We present a new interconnection topology called generalized Fibonacci topology, which unifies a wide range of connection topologies such as the Boolean cube (or hypercube), classical Fibonacci cube, etc. Some basic topological properties of generalized Fibonacci cubes are established. Finally, we developed new classes of the discrete orthogonal transforms, based on the generalized Fibonacci recursions. They can be implemented efficiently by butterfly-type networks (like the Fourier, or the Haar transforms). A generalized Fibonacci cube based processor architecture (generalizing the known SIMD architecture — hypercube processor) can be efficiently used for hardware implementation of the proposed discrete orthogonal transforms. Received: October 31, 1996     

Decentralized multi-product multi-stage systems with backorders

A multi-stage production/inventory system with decentralized two-card kanban control policies producing multiple product types is considered. The system involves one-at-a-time processing at each stage, finite target levels in the buffers and batch transfers between production stages. The demand process for each product is assumed to be Poisson and excess demand is backordered. Products have a priority structure and the processor at each stage is shared according to a switching rule. Our objective is to obtain steady-state performance measures such as the average inventory and service levels for each product type. We propose an approximation algorithm based on: (i) characterization of the delay by a product type before receiving the processor's attention at each stage; and (ii) creation of subsystems for all the storage activity and phase-type modeling of the remaining system's behavior. Numerical examples are presented to show the accuracy of the method and its potential use in system design.     

Dedicated optoelectronic stochastic parallel processor for real-time image processing: motion-detection demonstration and design of a hybrid complementary-metal-oxide semiconductor- self-electro-optic-device-based prototype

We report experimental results and performance analysis of a dedicated optoelectronic processor that implements stochastic optimization-based image-processing tasks in real time. We first show experimental results using a proof-of-principle-prototype demonstrator based on standard silicon-complementary-metal-oxide-semiconductor (CMOS) technology and liquid-crystal spatial light modulators. We then elaborate on the advantages of using a hybrid CMOS-self-electro-optic-device-based smart-pixel array to monolithically integrate photodetectors and modulators on the same chip, providing compact, high-bandwidth intrachip optoelectronic interconnects. We have modeled the operation of the monolithic processor, clearly showing system-performance improvement.     

Wavelet tests for the detection of transients in the VIRGO interferometric gravitational wave detector

Here we employ discrete wavelet transforms (DWTs) to develop test statistics for the detection of transients, i.e., signals with a short duration and unknown shape, embedded in Gaussian white noise. Distributions of the test statistics under both the and alternative hypotheses can be easily derived. We test performance on a set of 78 templates provided by theoretical studies on gravitational waves emitted in a supernova explosion where we seek the maximal detection distance of the source generating the signal at which the tests correctly reject the hypothesis. We discuss practical implementation issues and performance assessment methods. We compare results with both matched filtering, an optimal method that requires the prior knowledge of the signal shape, and with the slope filtering, that uses limited prior knowledge on the signal. The wavelet statistics show a good behavior for each of the considered waveforms, unlike other detection methods.     

Multi-domain two- and three-dimensional thermoelasticity by BEM

A computationally advantageous multi-domain boundary element formulation is presented for uncoupled thermoelastic analysis of two- and three-dimensional isotropic media. Particular integrals capable of representing exactly a quadratic temperature distribution in both two and three dimensions are employed. These particular integrals are derived using an orderly approach. When temperature data are provided at discrete points, the best fit quadratic distribution is generated in the least square error sense which, together with a multi-domain approach, makes it possible to represent an arbitrary temperature distribution within acceptable bounds. Numerical examples are given to show the efficiency and effectiveness of the present formulation, which is now incorporated in a general purpose boundary element code named GPBEST.     

Learning effective dispatching rules for batch processor scheduling

Batch processor scheduling, where machines can process multiple jobs simultaneously, is frequently harder than its unit-capacity counterpart because an effective scheduling procedure must not only decide how to group the individual jobs into batches, but also determine the sequence in which the batches are to be processed. We extend a previously developed genetic learning approach to automatically discover effective dispatching policies for several batch scheduling environments, and show that these rules yield good system performance. Computational results show the competitiveness of the learned rules with existing rules for different performance measures. The autonomous learning approach addresses a growing practical need for rapidly developing effective dispatching rules for these environments by automating the discovery of effective job dispatching procedures.     

Mie light-scattering granulometer with adaptive numerical filtering. I. Theory

A search procedure based on a least-squares method including a regularization scheme constructed from numerical filtering is presented. This method, with the addition of a nephelometer, can be used to determine the particle-size distributions of various scattering media (aerosols, fogs, rocket exhausts, motor plumes) from angular static light-scattering measurements. For retrieval of the distribution function, the experimental data are matched with theoretical patterns derived from Mie theory. The method is numerically investigated with simulated data, and the performance of the inverse procedure is evaluated. The results show that the retrieved distribution function is quite reliable, even for strong levels of noise.     

Parallel load‐balanced simulation for short‐range interaction particle methods with hierarchical particle grouping based on orthogonal recursive bisection

We describe an efficient load‐balancing algorithm for parallel simulations of particle‐based discretization methods such as the discrete element method or smoothed particle hydrodynamics. Our approach is based on an orthogonal recursive bisection of the simulation domain that is the basis for recursive particle grouping and assignment of particle groups to the parallel processors. Particle grouping is carried out based on sampled discrete particle distribution functions. For interaction detection and computation, which is the core part of particle simulations, we employ a hierarchical pruning algorithm for an efficient exclusion of non‐interacting particles via the detection of non‐overlapping bounding boxes. Load balancing is based on a hierarchical PI‐controller approach, where the differences of processor per time step waiting times serve as controller input. Copyright © 2007 John Wiley & Sons, Ltd.     

Adaptive learning algorithms for nernst potential and I-V curves in nerve cell membrane ion channels modeled as hidden Markov models

We present discrete stochastic optimization algorithms that adaptively learn the Nernst potential in membrane ion channels. The proposed algorithms dynamically control both the ion channel experiment and the resulting hidden Markov model signal processor and can adapt to time-varying behavior of ion channels. One of the most important properties of the proposed algorithms is their its self-learning capability-they spend most of the computational effort at the global optimizer (Nernst potential). Numerical examples illustrate the performance of the algorithms on computer-generated synthetic data.     

Discrete Green's methods and their application to two-dimensional phase unwrapping

A fully self-contained discrete framework with discrete equivalents of Stokes's, Gauss's, and Green's theorems is presented. The formulation is analogous to that of continuous operators, but totally discrete in nature, and the exact relationships derived are shown to hold provided that a set of predefined rules is followed in building discrete contours and domains. The method allows for an analytical rigor that is not guaranteed if one translates the classical continuous formulations onto a discretized approximated framework. We clarify several issues related to the use of discrete operators, which may play a crucial role in specific applications such as the two-dimensional phase-unwrapping problem, chosen as our main application example, and we show that reconstruction on irregular domains and/or in the presence of undersampling and noise is better formulated in the discrete framework than in the continuous domain.     

混凝土多孔砖砌体房屋抗震可靠度分析

通过对混凝土多孔砖砌体房屋结构的抗震可靠性研究,提出了基于概率地震烈度所对应的地面运动加速度峰值分布的可靠度计算公式和采用将各烈度地震对应的峰值加速度概率进行离散化处理的方法,基于作者早期的混凝土多孔砖墙片试验的结果推导出该结构水平最大位移和累积能量耗损的双参数破坏准则,分析了混凝土多孔砖砌体房屋结构在不同地震设防区的抗震安全性能及其影响因素。     

Markovian models for deadlock analysis in automated manufacturing systems

Deadlocks constitute a major issue in the desing and operation of discrete event systems. In automated manufacturing systems, deadlocks assume even greater importance in view of the automated operation. In this paper, we show that Markov chains with absorbing states provide a natural model of manufacturing systems with deadlocks. With illustrative examples, we show that performance indices such as mean time to deadlock and mean number of finished parts before deadlock can be efficiently computed in the modelling framework of Markov chains with absorbing states. We also show that the distribution of time to deadlock can be computed by conducting a transient analysis of the Markov chain model.