Efficient Function Implementation for Bit-Serial Parallel Processors

Many network-reliability analysis techniques define and compute a variety of reliability measures. Most techniques assume that network connectivity is the only determining factor in network reliability; and merely analyze an existing network structure but do not provide any methodology for reliable design. This paper presents a heuristic design algorithm to enhance the reliability of existing communication networks by modifying their topology. This algorithm improves the reliability of the least reliable node (reliability is the probability that messages transmitted from a given node reach their destination). To use this algorithm on large networks, a reliability analysis method is developed which determines approximate network reliability values in linear-time when an upper bound is placed upon the in-degree of all network nodes. The heuristic network design algorithm uses this approximate reliability analysis technique to place additional links. The goal of this link placement is to improve the reliability of the least-reliable node. The placement of additional links is a function of both the traffic distribution and the connectivity of the network. This process continues until either a desired level of network reliability is achieved or a maximum number of additional links has been added. A unique feature of this design strategy is that it has quadratic time-complexity when the maximum in-degree of all network nodes is limited. >

Objectives: In this paper, we consider the Flexible Flowshop Scheduling (FFS) problem with parallel machines. The main objective of this paper is to obtain a good schedule of jobs to minimize the makespan of FFS problem. Methods/Statistical analysis: In this study, two heuristic algorithms have been developed of FFS to reduce the makespan. First, we constructed the new heuristic algorithm based on Minimum Processing Time Selective Approach (MPTSA) and Longest Processing Times (LPT) approach to find the optimal or near optimal sequence for minimization of makespan of FFS problem with parallel machines. Next, we developed the heuristic algorithm using PALMER approach. In the PALMER approach we sequence the jobs based on Longest Slope Value (LSV) and obtained the value of objective function. Findings: We compared both the heuristic algorithms with the help of numerical illustrations. We solved the same numerical by both the heuristic algorithm and result show that our constructed heuristic algorithm has resulted in a better industrial production makespan. The percentage improvement of our constructive heuristic algorithm is also calculated. Gantt chart is also generated to verify the effectiveness of constructed heuristic algorithm. Application/Improvements: Our constructed heuristic algorithm is more effective to reduced the makespan of FFS problems as compare to classic heuristic algorithm as Palmer approach and provide an important tool for decision makers in production management. Keywords: Flexible Flowshop Scheduling, Gantt Chart, LPT Approach, Makespan, MPTSA Approach, Parallel Machines, Percentage Improvement

Process planning can be deemed as an important component in manufacturing systems. It bridges the gap between designing and manufacturing by specifying the manufacturing requirements and details to convert a part from raw materials to the finished form. For the purpose of low carbon emission, this paper pays attention to both technical performance measures and environmental impact criteria. In this problem, a part may have more than one process plans and only one process plan can finally be adopted. Due to the non-deterministic polynomial-hardness, it is rather difficult to conduct operation selection, machine determination, operation sequencing, and energy consumption reduction simultaneously with various constraints from technical requirements or the shop floor status. A novel position-based mixed integer linear programming model is developed to reduce total production time and the total energy consumption. The energy consumption coefficient matrix is proposed to evaluate the energy consumption in process planning. Because of the complexity in solving the model, this research proposes dynamic programming-like heuristic algorithm to tackle this problem. The weighted sum method is applied in multi-objective optimization and three typical instances with operation flexibility, sequencing flexibility, and processing flexibility are used to test the proposed algorithm. According to the results, both the total production time and the energy consumption criteria are optimized; in the best case, the energy consumption after optimization takes only 21.2% of the one before optimization. On average, about 40.9% of the total energy consumption can be reduced after optimization. Computational results also show that the proposed algorithm is generally better than the genetic algorithm. This research gives a novel perspective to reduce energy consumption in the process planning stage.

With recognition of quantum computer's enormous computational ability, it is of paramount importance to develop fault-tolerant quantum computing systems for their practical use. Recently, it has been shown that fault-tolerant systems can be achieved using a small set of basic quantum operations. This, however, incurs technical difficulties in finding an optimal sequence of basic operations toward a specific target computation and may limit possible quantum computations. In this work, we aim to achieve arbitrary target quantum computations under the restriction of four universal quantum gates of Pauli-X, -Y, -Z and SWAP. We develop two gate-sequence search methods based on the fidelity measure and deep neural networks. We verify the performance of our proposed methods through numerical results comparing total search space and the number of searched nodes.

Concurrent object-oriented programming languages (concurrent OOPLs) are promising approach for building large and complex applications on multicomputers, but their dynamic/asynchronous nature makes efficient implementation difficult--support of dynamic concurrent object creation involves scheduling of multiple objects on each node, and sending of asynchronous messages between objects cause buffering of messages, both of which could result in significant overhead. One possible direction to achieve efficiency is to combine dedicated hardware[3, 1] and software tailored for it[2, 7]. The hardware has crucial support for concurrent OOPLs--low overhead remote communication and efficient multithreading. While they are attractive, the majority of current commercial multicomputers such as AP1000[4] and CM5 have no such support. This note summarizes our implementation scheme[5, 8] on these conventional multicomputers, and shows that concurrent OOPLs can greatly benefit from their superior node-local performance while communication performance can be made comparable to dedicated architectures. Basic operations such as message passing or object creation were implemented very efficiently. Intra-node message passing takes about 15 instructions in addition to the cost of procedure call, unless the receiver object is already executing some other method. Inter-node message passing latency is within 10ps for smM1 messages. An application benchmark showed good speedup (440 times speedup on 512 nodes system) for a large program.

This paper aims to provide a smart design to improve the efficiency of surface permanent magnet synchronous motor. An efficient design strategy involving penalty approaches are considered for extracting all the possible parameter combinations and the solutions in the infeasible region. We compare the performance of tree heuristic optimization algorithms and six penalty methods. The heuristic optimization algorithms are: particle swarm optimization, differential search algorithm, and tree seed algorithm. The penalty methods are: three of which are static penalty approaches, two of dynamic penalty approaches, and Deb’s rule. Besides, the optimized motor design is tested with finite element analysis. Two conclusions were drawn from the experiments. First, heuristic algorithms using penalty approaches had significantly better performance compared to standard and popular heuristic algorithms. This emphasizes the importance of using heuristic algorithms with penalty approaches in SPMSM design optimization. Second, the compatibility of design optimization and numerical analysis results are acceptable and highly satisfactory for surface permanent magnet synchronous motor design. According to the analytical design, a 4% improvement in efficiency was achieved with the proposed approach.

Particle swarm optimization (PSO) is one of the modern heuristic algorithms. PSO has attracted great attention due to its features of easy implementation, robustness to control parameters and computation efficiency compared with other existing heuristic algorithms. The performance of a PSO can depend on its parameters such as the inertia weight factors and two acceleration coefficients. The value of inertia weight can treat the balance between exploration and exploitation, so a proper control of inertia weight is very important to find the optimum solution efficiently. This paper compared several kinds of inertia weights. Experimental results demonstrate that simulated annealing inertia weight and linearly decreasing inertia weight have better convergence performance in the searching procedure. The dimension of the solution has no effect on the performance of inertia weight. The performance of the two kinds of inertia weight will change when the benchmark functions and acceleration coefficients change.

In this study, crabs mating optimization (CRAB) algorithm that is one of the heuristic algorithms, has been developed and a monogamous crab mating optimization (MCO) algorithm has been proposed. In development, the main goal is to develop an algorithm that runs faster than the CRAB algorithm, to ensure obtaining good results like the CRAB algorithm. The developed MCO algorithm is compared with the popular heuristic algorithms such as CRAB, PSO, ABC, SA, TACO with the different metrics on five benchmark functions.

A new technique for approximating the entire solution set, in any bounding box, for a nonlinear system of relations (nonlinear equations, inequalities, etc. involving algebraic, smooth, or even continuous functions) is presented. The technique is to first plot each function as a pixel matrix, and to then perform a sequence of basic matrix operations, as dictated by how variables are shared by the relations in the system. The result is a pixel matrix graphing the approximated simultaneous solution set for the system.

In this article, a simple but novel technique for extracting a linear subspace related to event related potentials (ERPs) from ElectroEncephaloGraphy (EEG) data is introduced. The technique consists of a sequence of basic linear operations applied to multidimensional EEG data in a problem-specific manner. The derivation of the proposed technique is given and results with real data are described together with overall conclusions.

Graph edit distance is an error tolerant matching technique emerged as a powerful and flexible graph matching paradigm that can be used to address different tasks in pattern recognition, machine learning and data mining; it represents the minimum-cost sequence of basic edit operations to transform one graph into another by means of insertion, deletion and substitution of vertices and/or edges. A widely used method for exact graph edit distance computation is based on the A* algorithm. To overcome its high memory load while traversing the search tree for storing pending solutions to be explored, we propose a depth-first graph edit distance algorithm which requires less memory and searching time. An evaluation of all possible solutions is performed without explicitly enumerating them all. Candidates are discarded using an upper and lower bounds strategy. A solid experimental study is proposed; experiments on a publicly available database empirically demonstrated that our approach is better than the A* graph edit distance computation in terms of speed, accuracy and classification rate.

The presented paper comes as a summary of a theoretical work presenting an original algorithm describing International Multimodal Transport (IMT) mechanisms and processes. Since containers appeared in early 1960s on the regional, and later on, on global scale in logistic operations, the IMT has become the fastest growing combination of an international through-transport. With use of various modes of transport IMT has significant influence on global trade development. Due to its characteristic the IMT process is exposed on the global unitization, standardization, safety, security and economization (optimization) challenges. Thus, there is a need for scientific approach in building a standardized conceptual model of IMT referred in this paper to CAIMT. CAIMT is an original, deterministic algorithm, presented in the block notation, identifying typical procedures as a logical continuation of the programmed sequence of the basic IMT operations. The suggested CAIMT has a potential in further standardization of IMT procedures and can be scientifically developed through the modeling of the IMT processes and safety/security procedures. In addition, in terms of empirical aspects, algorithmization of the main IMT processes can lead to their application in the various expert systems. Summing up, this method (CAIMT) comes as an optimal instrument that can be utilized in order to build further pillars for IMT development, standardization and optimization.

Message Pattern Routing in Hypercubes: A Distribute-Concentrate Approach

A unidirectional mutation-selection approach for the general job shop scheduling problem is presented. The underlying algorithm uses threshold accepting as an iterative search technique and incorporates a special non-monotonic threshold function (threshold waving). Search is performed in the space of operation sequences and, for the purpose of evaluation, is combined with a simple heuristic transforming operation sequences into schedules. Mutation of solution candidates is based on a critical path analysis and, additionally, on a probabilistically employed local hill-climbing operator. Effectiveness of the approach is demonstrated within a comparison with some well-known heuristic and exact algorithms.

In Model-Driven Engineering (MDE) adequate means for collaborative modeling among multiple team members is crucial for large projects. To this end, several approaches exist to identify the operations applied in parallel, to detect conflicts among them, as well as to construct a merged model by incorporating all non-conflicting operations. Conflicts often denote situations where the application of one operation disables the applicability of another operation. Whether one operation disables the other, however, often depends on their application order. To obtain a merged model that maximizes the combined effect of all parallel operations, we propose an automated approach for finding the optimal merging sequence that maximizes the number of successfully applied operations. Therefore, we adapted and used a heuristic search algorithm to explore the huge search space of all possible operation sequences. The validation results on merging various versions of real-world models confirm that our approach finds operation sequences that successfully incorporate a high number of conflicting operations, which are otherwise not reflected in the merge by current approaches.