Tai Yue Wang

Fuzzy support vector machine for multi-class text categorization

Document classification, with the blooming of the Internet information delivery, has become indispensable required and is expected to be disposed by an automatic text categorization. This paper presents a text categorization system to solve the multi-class categorization problem. The system consists of two modules: the processing module and the classifying module. In the first module, ICF and Uni are used as the indictors to extract the relevant terms. While the fuzzy set theory is incorporated into the OAA-SVM in the classifying module, we specifically propose an OAA-FSVM classifier to implement a multi-class classification system. The performances of OAA-SVM and OAA-FSVM are evaluated by macro-average performance index. Also the statistical significance test is examined by the McNemars test. The results from the empirical study show that the proposed OAA-FSVM method has out-performed OAA-SVM in the multi-class text categorization problem.

Machine selection in flexible manufacturing cell: A fuzzy multiple attribute decision-making approach

Flexible manufacturing cells (FMC) have been used as a tool to implement flexible manufacturing processes to increase the competitiveness of manufacturing systems. In implementing an FMC, decision-makers encounter the machine selection problem including attributes, e.g. machine type, cost, number of machines, floor space and planned expenditures. This paper proposes a fuzzy multiple attribute decision-making (FMADM) model to assist the decision-maker to deal with the machine selection problem for an FMC realistically and economically. In addition, the membership functions of weights for those attributes are determined in accordance with their distinguishability and robustness when the ranking is performed.

A simulated annealing algorithm for facility layout problems under variable demand in cellular manufacturing systems

Abstract In this paper, we formulate a model to solve the facility layout problem in Cellular Manufacturing Systems (CMS). This model assumes that the demand rate varies over the product life cycle. The objective is to minimize the total material handling cost and solves both inter- and intra-cell facility layout problems simultaneously. Owing to the problem’s complexity and intractability, we use the simulated annealing algorithm to solve the presented model. The computational results show that the simulated annealing can obtain satisfactory solutions within reasonable time.

Artificial neural networks to classify mean shifts from multivariate χ 2 chart signals

A traditional multivariate control chart is shown to be effective in monitoring a multivariate process to signal the out-of-control condition that arises when mean shifts occur. The immediate classification of the signals associated with mean vector shifts can greatly narrow down the set of possible assignable causes, facilitating rapid analysis and corrective action by a technician before numerous nonconforming units have been manufactured. A persistent problem presented by such multivariate control charts, however, concerns the analysis of signals and the provision of any shift-related information. This study develops an artificial neural network-based model to supplement the multivariate χ2 chart. The method not only identifies the characteristic or group of characteristics that cause the signal but also classifies the magnitude of the shifts when the χ2-statistic signals that mean shifts have occurred. The method is described from the perspectives of training and classification. An example of the application of the proposed method is provided. The results demonstrate that the proposed method provides an excellent rate of classification and the output generated by trained network is very strongly correlated with the corresponding actual target value for every quality characteristic. Additionally, general guidelines for the proper implementation of the proposed method are provided.

Mean shifts detection and classification in multivariate process: a neural-fuzzy approach

For monitoring multivariate quality control process, traditional multivariate control charts have been proposed to detect mean shifts. However, a persistent problem is that such charts are unable to provide any shift-related information when mean shifts occur in the process. In fact, the immediate classification of the magnitude of mean shifts can greatly narrow down the set of possible assignable causes, hence facilitating quick analysis and corrective action by the technician before many nonconforming units are manufactured. In this paper, we propose a neural-fuzzy model for detecting mean shifts and classifying their magnitude in multivariate process. This model is divided into training and classifying modules. In the training module, a neural network (NN) model is trained to detect various mean shifts for multivariate process. Then, in the classifying module, the outputs of NN are classified into various decision intervals by using a fuzzy classifier and an additional two-point-in-an-interval decision rule to determine shift status. An example is presented to illustrate the application of the proposed model. Simulation results show that it outperforms the multivariate T2control chart in terms of out-of-control average run length under fixed type I error. In addition, the correct classification percentages are also studied and the general guidelines are given for the proper use of the proposed model.

Measuring the national competitiveness of Southeast Asian countries

National competitiveness is a measure of the relative ability of a nation to create and maintain an environment in which enterprises can compete so that the level of prosperity can be improved. This paper proposes a methodology for measuring the national competitiveness and uses the 10 Southeast Asian countries for illustration. The basic idea is to deconstruct the complicated concept of national competitiveness to measurable criteria. The observations (data) on the criteria are then aggregated according to their importance to obtain an index of national competitiveness. For data collected from questionnaire surveys, a calibration technique has been devised to alleviate bias due to personal prejudice. In data aggregation, the importance is expressed by both a priori weights and a posteriori weights. These two types of weights consistently show that Singapore, Malaysia, and Thailand have the highest national competitiveness, while Myanmar, Cambodia, and Laos are the least competitive countries. The performance of each country in every criteria measured also provides directions for these countries to make improvements and for investors to allocate resources.

Improving forecasting performance by employing the Taguchi method

To satisfy the volatile nature of todays markets, businesses require a significant reduction in product development lead times. Consequently, the ability to develop precise product sales forecasts is of fundamental importance to decision-makers. Over the years, many forecasting techniques of varying capabilities have been introduced. The precise extent of their influences, and the interactions between them, has never been fully clarified, although various forecasting factors have been explored in previous studies. Accordingly, this study adopts the Taguchi method to calibrate the controllable factors of a forecasting model. An L9(3 4 ) inner orthogonal array is constructed for the controllable factors of data period, horizon length, and number of observations required. An experimental design is then performed to establish the appropriate levels for each factor. At the same time, an L4(2 3 ) outer orthogonal array is used to consider the inherited parameters of forecasting method as the noise factors of Taguchi method simultaneously. An illustrated example, employing data from a power company, serves to demonstrate the thesis. The results show that the proposed model permits the construction of a highly efficient forecasting model through the suggested data collection method. � 2005 Elsevier B.V. All rights reserved.

Productivity diagnosis via fuzzy clustering and classification: An application to machinery industry

Business units are always faced with intensifying pressure in a competitive economy. Increasing productivity is an effective solution for a firm to survive and prosper. The relative productivity in an industry has evolved into a significant determinant of the competitive position for a firm. This paper proposes a productivity diagnosis process for a firm on the basis of the productivity characters of an industry to gain an insight into the firms relative productivity and to find the shortcomings in its management of resources. Firstly, productivity structure is determined. Pattern recognition technologies, namely fuzzy clustering and fuzzy classification, are then employed. After fuzzily clustering a training set according to three feature spaces, the productivity characters of the industry can be determined. A business unit can be diagnosed through fuzzily classifying its productivity features in a particular feature space and productivity indications can be furnished based on the associated productivity characters. As an illustration, data from 23 machinery firms in Taiwan are collected as a training set to analyze the productivity characters in each space, and two hypothetical firms are diagnosed.

An improved simulated annealing for facility layout problems in cellular manufacturing systems

Abstract In this paper, we formulate a model solving both inter-cell and intra-cell facility layout problems for cellular manufacturing systems. This model minimizes the total material handling distance on the shop floor. Due to the complexity of the problem, we propose an improved simulated annealing algorithm to solve this problem. This algorithm modifies the generation mechanism of neighborhood configurations. This new generation mechanism can always generate a neighborhood configuration that satisfies all of the zoning constraints. Then, the comparison between the improved simulated annealing algorithm and Kouvelis’s is conducted. The results show that the improved simulated annealing algorithm produces the same solution quality while requiring less computation time as the problem size is increased.

Productivity improvement: Efficiency approach vs effectiveness approach

Traditionally, raising the level of technology is considered the most effective way to improve productivity. Nevertheless, without the support of sound management systems, the contribution of technology to productivity is limited. In a sample of fifteen machinery firms, this paper calculates three indices for automation technology, production management, and productivity, respectively, to represent their levels of achievement. By using technology and management as the explanatory variables, a piecewise linear productivity frontier is constructed by applying a data envelopment analysis approach. At a given combination of the levels of technology and management, a firm may not be able to achieve the expected maximum productivity due to inefficient utilization of the input factors. One approach, the efficiency approach, for improving productivity which does not require the consumption of extra resources is to efficiently utilize the input factors. Another approach, the effectiveness approach, is to adjust the levels of technology and management toward the best combination to accomplish the highest productivity. Based on the productivity frontier constructed from the surveyed firms, the two approaches for improving productivity are discussed.