Advanced Materials Research Vols. 121-122

Abstract: Flexible job-sop scheduling problem (FJSP) is based on the classical job-shop scheduling problem (JSP). however, it is even harder than JSP because of the addition of machine selection process in FJSP. An improved artificial immune algorithm, which combines the stretching technique and clonal selection algorithm is proposed to solve the FJSP. The algorithm can keep workload balance among the machines, improve the quality of the initial population and accelerate the speed of the algorithm’s convergence. The details of implementation for the multi-objective FJSP and the corresponding computational experiments are reported. The results indicate that the proposed algorithm is an efficient approach for the multi-objective FJSP.

Abstract: The application of neural networks in Numerical Control (NC) machine reliability researching is introduced in this paper. Because of small reliability data sample, it is difficult to establish the reliability data distribution model, predict the reliability and allocate the reliability of NC machine. By using the neural networks, the reliability data of NC machine is extended. The reliability data distribution model can be established, the reliability prediction can also be done accurately. It is proved by the instance that the reliability prediction accuracy is improved 57% by the proposed method. The application of neural networks has a good reference to NC machine reliability researching.

Abstract: In this paper, we present a solution developed at rough set theory and weight value to make multiattribute decision with incomplete information. The paper defined the concepst of breaking points and recovered the incomplete information system according to the relationship between the condition attributes and decision attributes. And introduced OWGA(ordered weight geometric averaging) operators to calculate the aggregation value of each project. Finally, selected the project with the maximum aggregation value as the best decision making. The illustration and experiments were implemented and the results indicate that the method is effective and efficient.

Abstract: With rapid technology scaling, the leakage dissipation that begins to replace dynamitic dissipation is becoming a major source in CMOS circuits because of the increasing sub-threshold leakage current in nanometer CMOS processes. This paper introduces a MTCMOS power-gating technique, which is used for an adiabatic register file based on improved CAL (Clocked Adiabatic Logic) to reduce leakage dissipation in sleep mode. A 32 X 32 single-phase adiabatic register file are verified using HSPICE in different processes, threshold voltage, and active ratios, and BSIM4 model is adopted to reflect the leakage currents. Simulation results show that leakage losses are greatly reduced.

Abstract: Service oriented architecture presents us with a standards-based architecture for developing modular business applications and a lightweight framework for the business integration across heterogeneous platforms. However, due to the traits such as loose coupling, autonomy and distribution of web services, service oriented architecture is inherently vulnerable to exceptions. Moreover, the traditional exception handling strategies can hardly handle with the exceptions occurred in service oriented architecture. Based on the idea that exception management should be implemented in the system infrastructure, the paper presents a novel web services working framework with exception handling capability, including exception definition, exception handling policies, exception monitoring mechanism and handling process. The framework can dynamically monitor and handle with the exceptions that occur during the execution of web services.

Abstract: Existing enterprise information systems seldom consider different requirements from different users. To remedy this problem, the idea of manufacturing information active recommendation is put forward in this paper to deliver proper information to proper users correctly and timely. A new model called Sequence Behavior Access Pattern tree (SBAP-tree) is constructed based on the different requirements of user's identity, location, behavior habit and business needs for manufacturing information in Web environment. Using this SBAP-tree, historical situations similar to current situation could be sorted by their values, and the behavior could then be determined and output based on the association between the highly similar historical situations. Finally, an example is provided to demonstrate the effectiveness of SBAP-tree and the manufacturing information recommendation model based on SBAP-tree.

Abstract: In the practical applications, there is not only one critical path between nodes in the net. Therefore, it needs to find all critical paths among path selections under other restrictions, and then compare and select optimal or better path in these critical paths. Based on introduction of definition and nature theorem of main critical path, the paper brought out an expanded critical path algorithm in the view of application, which solved the storage problem of duality relation with duality table of net, so as to provide a good basis for determine number of critical path with main critical path algorithm.

Abstract: In the application of Genetic Algorithm (GA) to solve the function optimization problem, different encoding methods have different effect on performance of GA. Aiming at the global optimization problem of a class of nonlinear multi-peak function, the paper utilized binary coding and floating coding methods for genetic optimization and analyzed their performance. The experimental result of four kinds of typical nonlinear multi-peak function showed that under the precondition of given genetic operator, the optimizing performance of floating coding method to optimize nonlinear multi-peak function with isolated extreme points is less that the binary coding. The tuning ability of floating coding is stronger. As to the ordinary multi-peak function, the search affect is better than binary coding.

Abstract: Effective Supply Chain Management (SCM) approach must focus on flexible supply and production processes as well as rapidly respond to change of customer demands. To make up of existing drawbacks of association rule data mining algorithm, the paper brought out an improved algorithm and applies it in the product relativity analysis of SCM. Based on the algorithm, the solution of how the parts be arranged can achieve more cost-effective and higher profits can be achieved by data mining. The mining result can not only guide customers to correctly shopping, but also help manufacturers to design and produce goods, so that the companies can be in a better competitive position.

Abstract: As the consumptive concept increasingly matured, the product quality and the economics of quality has been an important aspect in the economical development. The paper laid emphasis on the research of comprehensive evaluation of economics of quality. It established the evaluation index system that can effectively indicate the conditions of economics of quality, measured the economics of quality comprehensively and quantitatively. A fuzzy evaluation model which can quantitatively evaluate the economics of quality was also built up. The feasibility and effectiveness of the model was also proved by a material example.