Shenghan Zhou

Approaches to comprehensive evaluation with 2-tuple linguistic information

The paper investigates the comprehensive evaluation problems with 2-tuple linguistic information. The study develops 2-tuple linguistic Maclaurin symmetric mean (2TLMSM) operator by improving Maclaurin symmetric mean (MSM) operator. The paper studies its properties and discusses its special cases. For the situations where the input arguments have different importance, the 2-tuple linguistic weighted Maclaurin symmetric mean (2TLWMSM) operator is defined based on the procedure which is developed for comprehensive evaluation problems under the 2-tuple linguistic environments. Finally, a practical example for evaluating the financial management performance is given to verify the developed approach and to demonstrate its practicality and effectiveness.

An improved MCDM model with cloud TOPSIS method

The paper aims to develop an improved MCDM model with cloud TOPSIS method. In the past literature, many methods were used to deal with the complexity and uncertainty of the world in Multiple-criteria decision-making process, such as the linguistic variable, fuzzy set and so on. However, linguistic concept, as an effective tool to describe human cognition, has both fuzziness and randomness, which cannot be dealt with very well by traditional methods. Cloud Model, the very method to handle both fuzziness and randomness of the linguistic concept, is specially imported into the TOPSIS to solve the fuzziness and randomness in decision-making. In order to achieve the Cloud TOPSIS, the difference of Cloud is proposed; meanwhile PIC (Positive Ideal Cloud) and NIC (Negative Ideal Cloud) are defined. Finally, the method of Cloud TOPSIS is demonstrated applicable and effective, compared with the TOPSIS method based on interval data. The result suggests that the improved model has better distinction degree.

Research on supply chain risk assessment with intuitionistic fuzzy information

The paper suggests an intuitionistic fuzzy operator to assess supply chain risk. With the global economic integration trend, more and more enterprises adopt the modern management model of supply chain in order to have access to the advantage of responding quickly to the market demand. Through the implementation of supply chain management, their management efficiency has been significantly improved which led to brought good economic benefits. However, at the same time risk is becoming an unavoidable problem for the enterprises in supply chain, because of the uncertainty of the environment of the supply chain and the ever-increasing complexity of its own. The management of supply chain risk assessment has attracted increasing attention of theoretical and business research. The study investigates the supply chain risk assessment with intuitionistic fuzzy information, and then proposes a dependent intuitionistic fuzzy Hamacher weighted geometric (DIFHWG) operator. This operator is used to design an algorithm for supply chain risk assessment with intuitionistic fuzzy numbers. To demonstrate the effectiveness of this approach, several experiments are conducted to verify the developed method.

A Novel Bearing Multi-Fault Diagnosis Approach Based on Weighted Permutation Entropy and an Improved SVM Ensemble Classifier

Timely and accurate state detection and fault diagnosis of rolling element bearings are very critical to ensuring the reliability of rotating machinery. This paper proposes a novel method of rolling bearing fault diagnosis based on a combination of ensemble empirical mode decomposition (EEMD), weighted permutation entropy (WPE) and an improved support vector machine (SVM) ensemble classifier. A hybrid voting (HV) strategy that combines SVM-based classifiers and cloud similarity measurement (CSM) was employed to improve the classification accuracy. First, the WPE value of the bearing vibration signal was calculated to detect the fault. Secondly, if a bearing fault occurred, the vibration signal was decomposed into a set of intrinsic mode functions (IMFs) by EEMD. The WPE values of the first several IMFs were calculated to form the fault feature vectors. Then, the SVM ensemble classifier was composed of binary SVM and the HV strategy to identify the bearing multi-fault types. Finally, the proposed model was fully evaluated by experiments and comparative studies. The results demonstrate that the proposed method can effectively detect bearing faults and maintain a high accuracy rate of fault recognition when a small number of training samples are available.

Using the Capture-Recapture Method to Estimate the Incidence of Amyotrophic Lateral Sclerosis in Beijing, China

Background: To assess the total, gender-related and ageing process-related incidence rates of amyotrophic lateral sclerosis (ALS) in Beijing, China. Determine whether the decreased male to female ratio among postmenopausal age groups. Methods: We used the 2-source capture-recapture method to estimate the incidence of ALS in Beijing. The primary and secondary data sources were from diagnostic hospitals and assisted care institutions in the same area from 2010 to 2015. Results: A total of 562 cases and 283 cases were extracted from 2 data sources, and a total of 962 patients diagnosed with ALS within the 6-year period were estimated (95% CI 883–1041). The average yearly incidence was 0.77/100,000 persons (95% CI 0.71–0.83). The female to male ratio was 1.63. The incidence was associated with age and peaked in the 55–64 year age group. There was no obvious decline in the male:female ratio among postmenopausal age groups. Conclusions: The total incidence of ALS in Beijing is similar to international reports. The onset of ALS is not merely the result of ageing.

Research on detection methods based on Doc2vec abnormal comments

Abstract The purpose of this paper is to explore a method for detecting abnormal comments. With the growth of e-commerce sites and reviews sites, user reviews of messages begin to affect merchant’s revenue and impact consumer’s choice. It is meaningful to help users get real comment messages by using natural language processing means. In this paper, the doc2vec model, the clustering algorithm and emotion analysis are applied to identify abnormal comments.

A β-accurate linearization method of Euclidean distance for the facility layout problem with heterogeneous distance metrics

Abstract Most existing research on facility layout problems (FLPs) considers a single distance metric, mainly Rectilinear distance, in the calculation of the material handling cost between departments. However, there are many industrial cases in which heterogeneous distance metrics may need to be used simultaneously to cater for different styles of material handling, such as the Euclidean distance metric for conveyor belts and the Tchebychev distance metric for overhead cranes. In this paper, we study the unequal area facility layout problem with heterogeneous distance metrics (UA-FLP-HDM), considering a hybrid use of three metrics, i.e., Rectilinear, Euclidean, and Tchebychev, as distance measures of different styles of material handling in the production system. We propose a β -accurate linearization method that uses a set of tangent planes to convert the non-linear Euclidean distance constraint into a set of linear constraints that guarantee the approximation error within a given percentage β , e.g., as small as −0.01% in our experiments, and develop linear constraints for the Tchebychev distance metric as well. Based on these contributions, we present a mixed-integer linear programming (MILP) model for the UA-FLP-HDM. Computational experiments are carried out to test the performance of the MILP model with five benchmark problems in the literature and compare the layout designs using different distance metrics. Numerical results indicate that different distance metrics may lead to significantly different solutions and a hybrid use of heterogeneous distance metrics fits better for real industrial applications.

An improved aircraft hard landing prediction model based on panel data clustering

This paper proposes a hard landing prediction method based on panel data clustering with flight data. The hard landing is a hazard that is critical to flight during the landing phase. It may cause damage to the aircraft structure, resulting in direct or indirect economic losses, damaging to human comfort and other adverse consequences. Firstly, based on the panel data in economics, the flight panel data is established; secondly, extracts the characteristic information of several key flight variables that affect the hard landing in each landing. The feature information includes mean, standard deviation, median, maximum, kurtosis, skewness and trend, and constitutes the eigenvectors describing the landings; then the k-means method is used to cluster the feature information. Finally, the empirical study is carried out on the 22 landing data of fixed wing unmanned aerial vehicles (UAVs). The results show that the clustering of flight panel data can be applied to hard landing prediction, and the prediction effect is obvious.

A prediction model of hard landing based on RBF neural network with K-means clustering algorithm

This paper proposes a prediction model for forecasting the hard landing problem. The landing phase has been demonstrated the most dangerous phase in flight cycle for fatal accidents. The landing safety problem has become one of the hot research problems in engineering management field. The study concentrates more on the prediction and advanced warning of hard landing. Firstly, flight data is preprocessed with data slicing method based on flight height and dimension reduction. Subsequently, the radial basis function (RBF) neural network model is established to predict the hard landing. Then, the structure parameters of the model are determined by the K-means clustering algorithm. In the end, compared with Support Vector Machine and BP neural network, the RBF neural network based on K-means clustering algorithm model is adopted and the prediction accuracy of hard landing is better than traditional ways.

Quality evaluation on diesel engine with improved TOPSIS based on information entropy

This paper aims to evaluate the quality of diesel engine using an improved TOPSIS method. In the past research, the quality evaluation of the diesel engine is mainly based on fuzzy evaluation. However, the data generated in the run-in process, which is the test run of the whole diesel engine, have not been fully utilized. In this paper, evaluating the quality of the diesel engine with improved TOPSIS will use the run-in data. By calculating the correlation coefficient matrix of the data, there is a correlation between the indexes, which will have an impact on the evaluation results. Thus, Mahalanobis distance is adopted. Meanwhile, the information entropy is used to determine the weights. Then, a processing method for the appropriate index is put forward. Finally, the improved TOPSIS is demonstrated effective in evaluating the quality of the diesel engine compared with the TOPSIS method based on the subjective weight. The result suggests that the improved method has better distinction degree.