Gengui Zhou

A new method for VE partner selection and evaluation based on AHP and fuzzy theory

In view of inconsistency of the judgment matrix gotten by pairwise comparison between increasing alternatives in nowadays virtual enterprise (VE) partner selection & evaluation problem by the analytic hierarchy process (AHP), a new Web-based method combined the AHP and other basic fuzzy theory is proposed, in which the priority weights of decision criteria in every level of hierarchy structure are identified by the AHP and attribute-values of every alternative are determined by fuzzy relation matrix, then the evaluation result is acquired by synthesizing them finally. An application system based on this method has been developed, and a real case study of selecting VE partners demonstrated its application and examined its effectiveness. The use of the proposed method indicates that it is not only suited to solving problems such as regional restriction, long treating time and inflexible processing way, etc., which are produced by applying traditional partner selection method, but also suited to making up drawbacks including limited number of alternatives, unreasonable mathematical model structure, etc., which are caused by the use of the AHP or fuzzy-AHP. This applicable method makes it easier for the evaluation team to arrive at a consensus decision, obtain fair and reasonable conclusions and examine the strengths and weaknesses of alternatives by comparing them with respect to appropriate criteria.

A Smoothing Support Vector Machine Based on Quarter Penalty Function

It is very important to find out a smoothing support vec- tor machine. This paper studies a smoothing support vec- tor machine (SVM) by using quarter penalty function. We introduce the optimization problem of SVM with an uncon- strained and nonsmooth optimization problem via quarter penalty function. Then, we define a one-order differentiable function to approximately smooth the penalty function, and get an unconstrained and smooth optimization problem. By error analysis, we may obtain approximate solution of SVM by solving its approximately smooth penalty optimization problem without constraints. The numerical experiment shows that our smoothing SVM is efficient.

Multiobjective Interaction Programming Problem with Interaction Constraint for Two Players

This paper extends an existing cooperative multi-objective interaction programming problem with interaction constraint for two playersor two agents� . First, we define an s-optimal joint solution with weight vector to multi-objective interaction programming problem with interaction constraint for two players and get some properties of it. It is proved that thes-optimal joint solution with weight vector to the multi-objective interaction programming problem can be obtained by solving a corresponding mathematical programming problem. Then, we define another s- optimal joint solution with weight value to multi-objective interaction programming problem with interaction constraint for two players and get some of its properties. It is proved that the s- optimal joint solution with weight vector to multi-objective interaction programming problem can be obtained by solving a corresponding mathematical programming problem. Finally, we build a pricing multi-objective interaction programming model for a bi-level supply chain. Numerical results show that the interaction programming pricing model is better than Stackelberg pricing model and the joint pricing model.

A New Tree Encoding for the Degree-Constrained Spanning Tree Problem

The degree-constrained minimum spanning tree (dc- MST) problem is of high practical importance. Up to now there are few effective algorithms to solve this problem because of its NP-hard complexity. More recently, a ge- netic algorithm (GA) approach for this problem was tried by using Pr¨ufer number to encode a spanning tree. The Pr¨ufer number is a skillful encoding for tree but not efficient enough to deal with the dc-MST problem. In this paper, a new tree encoding is developed directly based on the de- gree constraint on each vertex and the connectivity among vertices. We denote it as degree-based permutation encod- ing and apply it to the dc-MST problem by using the GA approach. Compared with the numerical results and CPU runtimes between two encodings, the new degree-based per- mutation is effective to deal with the dc-MST problem and even more efficient than the Pr¨ufer number to evolve to the optimal or near-optimal solutions.

A Genetic Algorithm Approach on Capacitated Minimum Spanning Tree Problem

For the capacitated minimum spanning tree problem (CMST), there are still no effective algorithms to solve this problem up to now. In this paper, we present a completely new approach by using the genetic algorithms (GAs). For the adaptation to the evolutionary process, we developed a tree-based genetic representation to code the candidate solution of the CMST problem. Numerical analysis shows the effectiveness of the proposed GA approach on the CMST problem

The Risk-Evaluation Model in Customs Based on BP Neural Networks

The basic learning algorithm of BP neural network based on delta learning rule is introduced, and the Levengberg-Marquardt algorithm is described. Then the BP neural network model of risk-evaluation in China Customs is presented. It includes the design of the input layer, the output layer and the hidden layer and the confirmation of the initial weights and the learning rate. On the basis of the data warehouse of risk management of China Customs, The BP neural network of the risk-evaluation is implemented, and the comparison of performance between the momentum factorial algorithm and the Levenberg-Marquardt algorithm is indicated. At last, the further application of BP neural network in China Customs is discussed.

Partner Selection and Evaluation in Agile Virtual Enterprise Based upon TFN-AHP Model

To solve some problems caused by the analytic hierarchy process (AHP) or other fuzzy-AHP approaches to partner selection and evaluation in the formation of agile virtual enterprises, a model based on the AHP and basic theory of the trigonometrical fuzzy number (TFN) was given and corresponding evaluation process was described. Compared with the AHP, the proposed model combines subjective analysis with quantitative analysis more reasonably, synthesizes group opinions more adequately. The model derives priorities from TFN-based judgment matrices effectively, regardless of their inconsistency. In addition, the model can be programmed easily due to its simple, normative arithmetic formulas. The results of a case study on selecting suppliers in an agile virtual enterprise indicate that, by applying the model-based software system, not only fair and reasonable conclusions, but also the consistent degree of opinions provided by different experts can be acquired clearly

A genetic algorithm approach on reverse logistics optimization for product return distribution network

Traditionally, product returns have been viewed as an unavoidable cost of distribution systems. Up to now there are few studies to address the problem of determining the number and location of centralized product return centers where returned products from retailers or end-customers are collected for manufacturers’ or distributors’ repair facilities while considering the distribution system. To fill the void in such a line of research, this paper proposes a nonlinear mixed-integer programming model and a genetic algorithm that can solve the distribution problem with forward and reverse logistics simultaneously. Compared with a partly enumeration method, the numerical analysis shows the effectiveness of the proposed model and its genetic algorithm approach.

A group-evaluation method based on AHP and theory of triangular fuzzy number

In view of uncertainty degree of judgment in group-evaluation problem, a new method combined the analytic hierarchy process (AHP) with the basic theory of triangular fuzzy number was given in this paper. A real case study of personnel selection in Matlab environment demonstrated its application and examined its effectiveness. The use of this proposed method indicates that it is easier for the evaluation team not only to obtain impartial and reasonable resulting value, but also to arrive at a consensus decision. Moreover, the consistent degree of decision or opinion of each decision element from different expert can be showed clearly.

The generalized approaches of genetic algorithms on constrained minimum spanning tree problems

This paper explores the generalization of problem-solving by genetic algorithms (GAs) for a class of constrained minimum spanning tree (MST) problems. These constrained MST problems are quite different from each other in constraints and usually NP-hard but of high practical importance. The paper emphasizes that it is possible for GAs to generalize the problem-solving as long as the problems at hand can be generalized in data structure of solutions. The proposed method adopts only one solution encoding and one algorithm to deal with three constrained MST problems. Numerical experiments show that the proposed GAs approach can respectively compete with the existing heuristic algorithms on these constrained MST problems.