The MOE Key Laboratory of Process Optimization and Intelligent Decision making

Hefei, China

The MOE Key Laboratory of Process Optimization and Intelligent Decision making

Hefei, China
SEARCH FILTERS
Time filter
Source Type

Yu Z.-J.,Hefei University of Technology | Yu Z.-J.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | Yang S.-L.,Hefei University of Technology | Yang S.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | And 4 more authors.
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2015

It is of great significance to improve the prediction effect of forecasting method, however experience shows that it is very difficult for improving the accuracy of forecasting by setting up single forecasting model. This article describes the deficiencies of the present forecasting methods and puts forward a new approach for the improvement of prediction accuracy by introducing error correction. First, the fuzzy neural network forecasting model is established for a preliminary prediction by using the training sample data. Second, the data transformation is introduced to process the error sequence. On the basis of the processed data, the GM (1, 1) model is constructed and is used to predict the subsequent error. Third, the correction of preliminary prediction values is calibrated. The numerical example based on the historical data of the Shanghai composite index shows that the presented approach improves the accuracy of forecasting significantly compared with the prediction accuracy before correction, and then the validity of the model is verified. ©, 2015, Systems Engineering Society of China. All right reserved.


Zhang C.,Hefei University of Technology | Zhang C.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | Ni Z.-W.,Hefei University of Technology | Ni Z.-W.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | And 3 more authors.
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2015

PM2.5 is the main pollutant affecting the air quality, the concentration of PM2.5 is closed related to meteorological conditions, studying the influence of meteorological conditions on the concentration of PM2.5 has important significance for improving urban air quality. As fractal and wavelet have lots of advantages when dealing with complex nonlinear system, the calculating method of joint multifractal based on wavelet packet transform modulus maxima (WPTMM) has been proposed, first the variable sequences are decomposed by wavelet packet, this paper uses modulus maxima to denoise, then constructs the joint distribution function, finally calculates the joint multifractal spectrum, and analyzes the fractal correlation between two variables. This proposed method has extended single multifractal to the joint multifractal of two interacting variables, calculating joint multifractal spectra based on WPTMM can reduce computational complexity, meanwhile avoid the effects of noise. The paper has analyzed the relationship between the concentration of PM2.5 and the meteorological factors of Beijing and Hong Kong, experiment results show that this method can effectively analyze each meteorological factor on the impact of PM2.5 concentration in different seasons. ©, 2015, Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice. All right reserved.


Ni Z.-W.,Hefei University of Technology | Ni Z.-W.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | Xue Y.-J.,Hefei University of Technology | Xue Y.-J.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | And 4 more authors.
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

In order to reflect the financial status of enterprises objectively and comprehensively, more indexes will be included and the dimensionality of data set will be too large that traditional methods can't perform well in the process of financial distress prediction. Manifold learning performs well on high dimensional data set, and multiple kernel SVM has excellent classification performance on non-flat data set. Therefore, a hybrid algorithm of financial distress prediction model which integrates multiple kernel learning with manifold learning is proposed, and it can be used in the situation of prediction research with large number of indexes. Experiment results show that this model has better performance than traditional methods. ©, 2014, Systems Engineering Society of China. All right reserved.


Zhou K.-L.,Hefei University of Technology | Zhou K.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | Yang S.-L.,Hefei University of Technology | Yang S.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | And 2 more authors.
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

This paper proposes an adaptive fuzziness parameter selection method of fuzzy c-means (FCM) algorithm based on the establishment of five-stage load classification process model. The evaluation index of adaptive fuzziness parameter selection is the ratio of the sum of within-class distances and the sum of between-class distances. At the same time, simulated annealing algorithm and genetic algorithm are utilized to optimize the global search capability of FCM algorithm. Experimental results show that the widely used fuzziness parameter of FCM algorithm in load classification m=2 is not optimal, and we give the optimum range that is [2.6, 3.2]. The modified algorithm enhances the global search capability of traditional FCM algorithm, thus enhancing the accuracy and effectiveness of load classification.


Liu Y.-Z.,Hefei University of Technology | Liu Y.-Z.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | Zhou Y.-L.,Hefei University of Technology | Zhou Y.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

Studying the topology of scale-free networks mainly concentrates on computing clustering coefficient and average path length, and analyzing degree distribution. This paper argues that in real world, the three parameters are interrelated. A parameter can be replaced by the other two parameters. According to the viewpoint, this paper gives a formula to compute the average path length l>SF of large scale-free networks based on a tree structure model, and analyzes the impact of network scale and junction between nodes on l>SF. The results indicate that l>SF is related to average degreek, average clustering coefficient C and power exponent γ which are three parameters characterizing the topology of scale-free networks. Therefore, the complexity can be reduced by transferring computing average path length directly to indirectly. The experiments' results show that the formula is valid and the efficiency of studying the topology of large scale-free networks is greatly raised.


Peng Z.-L.,Hefei University of Technology | Peng Z.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | Zhang Q.,Hefei University of Technology | Zhang Q.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | And 4 more authors.
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

In the process of multiple attribute evaluation using traditional multi-attribute decision method, some problems such as that the evaluation results are close to each other, the evaluation values differ to each other scarcely, the difference between these evaluations is not obvious and so on can easily occur in practice. Although the multiple attribute decision model based on maximizing deviation can enlarge the evaluation values discrepancy for the optimizing and ranking of the decision making schemes, it only considers the variable weight caused by the discrepancy of the attribute values and neglects the weight of the evaluation indicator itself in practice, causing that the evaluation results easily deviate the actual results. In view of this, the paper makes further amendments to the multi-attribute decision making model based on maximizing deviation. After considering the weighted weight and the weight of the evaluation attribute completely, the author designs an improved maximizing deviation decision making model and applies the model to the comprehensive effectiveness evaluation of the multi-mission planning of near space system, which can provide decision support for the selection and optimization of the multi-mission planning schemes of near space system. Finally, the model is proved to have certain applicability by the case analysis.


Xue Y.-J.,Hefei University of Technology | Xue Y.-J.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | Ni Z.-W.,Hefei University of Technology | Ni Z.-W.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

With the rapid development of the information technology, it is challenging for the traditional machine learning and data mining algorithms to deal with large scale explosive growth data. Manifold learning is a dimensionality reduction algorithm which can overcome some shortages of traditional linear dimensionality reduction methods. However, it is not useful for large scale data because of high complexity. In order to deal with the dimensionality reduction of large scale data, a distributed manifold learning algorithm is proposed based on MapReduce. Local sensitive hash functions are used to map the similarity points to the same bucket, then the geodesic distance between points in the same bucket can be computed by Euclidean norm according to the local homeomorphisms of Euclidean spaces of manifold and the geodesic distance among points between buckets can be computed by the modified geodesic distance formula which takes use of central points and edge points. Experiments on large scale of manmade dataset and real dataset show that this distributed manifold learning algorithm can approximate the geodesic distance between points effectively and it is useful for large scale dimensionality reduction.


Zhang Q.-P.,Hefei University of Technology | Zhang Q.-P.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | Liu Y.-Z.,Hefei University of Technology | Liu Y.-Z.,The MOE Key Laboratory of Process Optimization and Intelligent Decision making | Li Y.-J.,Hefei University of Technology
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

This paper has studied the problem of allocating the fixed cost among decision making units. Suppose the production processes of decision making units during the two continuous periods which are before and after the fixed cost input are available, and the modeling premise is that combining the allocated cost with other input elements averagely. Firstly, the decision making units' super CCR effciency evaluation model considering allocated cost is given. Then the input-output variation and allocated fixed cost oriented decision making units' relative benefit recognition model is built. Based on the Nash bargaining cooperative game theory, the cost allocation model considering relative effciency and benefit is proposed. The approach is illustrated by a numerical example, which figures that the approach is available and acceptable.


Yang Y.,Hefei University of Technology | Yang Y.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | Yang S.-L.,Hefei University of Technology | Yang S.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | And 2 more authors.
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

Focusing on characteristics of new product development projects, this paper investigates project portfolio selection considering consistency between projects and enterprise strategy, and interdependency among projects. Firstly, the concept of strategic fit was proposed to describe the consistency degree and formulated; then the strategic fit was integrated into project portfolio selection model. Considering the effect of the strategic fit on combinational revenues, this paper improves traditional methods that only took interdependencies into account. A model was proposed to optimize project selection based on strategic fit, and an improved simulated annealing algorithm was designed from the variable neighborhood construction to solve the model. Lastly, an example of an automobile company was adopted to illustrate its application, which verified reasonability of the proposed methods.


Zhou K.-L.,Hefei University of Technology | Zhou K.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | Yang S.-L.,Hefei University of Technology | Yang S.-L.,The MOE Key Laboratory of Process Optimization and Intelligent Decision Making | And 4 more authors.
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | Year: 2014

Clustering is an unsupervised learning process, hence it is difficult to find the optimal cluster number. Cluster validation is a process in which a cluster validity index (CVI) is constructed to evaluate the quality of clustering results and determine the optimal cluster number. Firstly, the mathematical description of clustering and the classification of CVIs are introduced. Then, 12 CVIs only considering the geometry information of the data set, 6 CVIs only considering the degree of membership, and 9 CVIs considering both the geometry information of the data set and the degree of membership are reviewed respectively based on different components in the indices. And the status quo of each type of CVIs is analyzed. Afterwards, the studies of other types of CVIs, such as external and stability-based indices, are briefly summarized. Finally, we point out the main challenges and research directions in the area of cluster validation.

Loading The MOE Key Laboratory of Process Optimization and Intelligent Decision making collaborators
Loading The MOE Key Laboratory of Process Optimization and Intelligent Decision making collaborators