Applied Mechanics and Materials Vols. 63-64

Abstract: To solve the dynamic and complex problem of production scheduling, depending on the introduction between multi-agent and hybrid genetic algorithm in mixed-model flow production scheduling, this paper proposed a mixed-model flow production scheduling method based on multi-agent and hybrid genetic algorithm. On the basis of this model, the mixed-model flow production scheduling procedure and strategy based on multi-agent and hybrid genetic algorithm were established. Finally, mixed-model flow production scheduling simulation system based on multi-agent and hybrid genetic algorithm was demonstrated and validated by QUEST software. It has shown the proposed method can improve the benefit of production scheduling, and provide a support for adapting to complex and dynamic production scheduling in mixed-model flow production.

Abstract: Risk management and its accurate analysis are very important for project management. RBF and MLP Neural Network Model are common methods of risk management and analysis, which are not accurate enough. In this paper a new method based on LS-SVM is introduced. Analytical models of risky projects are investigated and function approximation results are compared. Experimental results show that the regression analysis of risk based on LS-SVM method has higher prediction accuracy and better generalization ability.

Abstract: Evaluating the performance of activities or organizations by traditional data envelopment analysis model requires crisp input/output data. However, in real-world problems inputs and outputs are often with some fuzziness. To evaluate DMU with fuzzy input/output data, researchers provided fuzzy data envelopment analysis (FDEA) model and proposed related evaluating method. But up to now, we still cannot evaluate a fuzzy sample decision making unit (SDMU) for FDEA model. So this paper proposes a generalized fuzzy DEA model which can evaluate a sample decision making unit and a numerical experiment is used to illustrate this model.

Abstract: The surface roughness is difficult to estimate in machining, especially for weak stiffness workpiece. So, prediction model of surface roughness using artificial neural network (ANN) is developed. This model investigates the effects of cutting parameters during milling Aluminum alloy 6061. The experiments are planned with four factors and four levels for developing the knowledge base for ANN training. Three-dimensional surface plots are generated using ANN model to study the effects of cutting parameters on surface roughness. The analysis reveals that cutting speed and feed rate have significant effects in reducing the surface roughness, while the axial and radial depth of cut has less effect. And the variations of surface roughness are highly non-linear with the cutting parameters.

Abstract: The research on Communication based on Ad hoc networks between vehicles is highly concerned about in many countries. However, Vehicular Ad hoc Networks (VANETs) is in the environment in which traditional methods have certain limitations. In this paper, vehicular network scenarios and the probability of penetration model are given. Through adding obstacles and the probability of penetration model to NS2, GPSR performs better in the new simulation scenarios.

Abstract: Based on scheme of Dalian gulf cross-sea bridge, in this paper, a 3-dimensional FE model for Self-anchored cable-stayed suspension bridge is established with finite element program and pile-soil-structure interaction is simulated by use of the equivalent embed fixation model. Based on the FE model, model analysis is carried out and the effects of pile-soil-structure interaction on dynamic behavior of long-span self-anchored cable-stayed suspension bridge are specially studied. The seismic response analysis result considering that pile-soil-structure interaction was compared with that of without considering such interaction. The analysis result show that interaction reduced horizontal displacement in middle span of stiffening beam and top of tower, horizontal moment not only at bottom of tower, but also assistant piers. The research results provide some theoretical foundation to composite structure system.

Abstract: Pattern growth approach is an important method in sequential pattern mining. Projection database based on the method is introduced in PrefixSpan, and the PrefixSpan algorithm can solve the problem of mining sequential patterns. But relative to large projection database, the performance of PrefixSpan is affected. Inspired by the prefix-divide method and MH structure, this paper proposed a new algorithm MHSP for sequential pattern mining. Based on the real datasets, experimental results show that the performance of MHSP algorithm is more than twice as fast as PrefixSpan.

Abstract: This article firstly established two finite element models of seam weld and spot-weld for the front sub-frame, then used the two models for seam weld and spot weld fatigue simulation analysis, next cooperated with bench test results for verification. After the above steps, we not only found the weak portion of the front sub-frame, but also provide reference for both optimum design and improving the bench test design.

Abstract: Three dimensional finite element model of roller and sheet rolling process was first built according to roller and flattening sheet rolling wear characteristics by using ANSYS/LS-DYNA software. Total 10 rolling models including 1 smooth surface and 9 non-smooth surfaces were established to compare with wear resistance characteristics difference. Simulation results showed that comparing with smooth surface roll sample, the non-smooth roller surface with concaves possessed had higher Von Mises stress value with non-uniform fluctuation at mean value in favor of enforcing rolling sheet fluidity. On the other hand, some air concentrated on concave circumjacent area will low adhesion wear appearance between plate and roller. Thirdly, Concave on the roller surface was also able to protect from sliding at friction interfaces. It was these factors above that were in favor of improving wear resistance capability of the rollers. The result of the model tests and the simulation are identical.

Abstract: With the development of knowledge economy, E-commerce is facing more and more challenges, therefore, it is necessary to formulate a new developing strategy of e-commerce through the introduction of Knowledge Management (KM). This paper discusses the relationship between KM and e-commerce, points out that it is very necessary to introduce KM into e-commerce. Afterward, on the basis of analysis of knowledge contents concerned in e-commerce, it proposed the KM-based developing strategy of e-commerce from three aspects: the premise of development, the developing path, and the guaranteeing measure.