Journal of Intelligent & Fuzzy Systems - Volume 44, issue 6

Authors: Nurhidayat, Irfan | Pimpunchat, Busayamas | Klomsungcharoen, Wiriyabhorn

Article Type: Research Article

Abstract: This study aims to present the modified SVM polynomial method in order to evaluate insurance data. The research methodology discusses classical and modified SVM polynomial methods by R programming, and uses performance profiles to create the most preferable methods. It offers a new algorithm called an accurate evaluating algorithm as the way to construct the modified SVM polynomial method. The classical SVM polynomial method is also represented as the main idea in finding the modified polynomial SVM method. Model Performance Evaluation (MPE), Receiver Operating Characteristics (ROCs) Curve, Area Under Curve (AUC), partial AUC (pAUC), smoothing, confidence intervals, and thresholds are …further named an accurate evaluating algorithm, employed to build the modified SVM polynomial method. The research paper also presents the best performance profiles based on the computing time and the number of iterations of both classical and modified SVM polynomial methods. Performance profiles show numerical comparisons based on both methods involving insurance data also displayed in this paper. It can be concluded that applying an accurate evaluating algorithm on the modified SVM polynomial method will improve the data accuracy up to 86% via computing time and iterations compared to the classical SVM polynomial method, which is only 79%. This accurate evaluating algorithm can be applied to various large-sized data by utilizing R programming with changing any suitable kernels for that data. This vital discovery will offer solutions for faster and more accurate data analysis that can benefit researchers, the private sector, or governments struggling with data. Show more

Keywords: Modified SVM polynomial method, classical SVM polynomial method, accurate evaluating algorithm, insurance data, simulation

DOI: 10.3233/JIFS-222879

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9129-9141, 2023

Authors: Feng, Xiangqian | Zibibula, Minawaer | Wei, Cuiping

Article Type: Research Article

Abstract: With the rapid development of science and technology, high-tech enterprises need to constantly carry out technological innovation to adapt to the changes in the external environment, and maintain their competitive advantages. However, the current research on technological innovation of high-tech enterprises is carried out from a static perspective, which is difficult to understand the dynamic evolution process of continuous technological innovation of high-tech enterprises in a turbulent environment. Therefore, this paper studies high-tech enterprises’ dynamic technological innovation ability from a dynamic perspective, through literature reading and the investigation of the technological innovation status of high-tech enterprises, the evaluation index system …of 12 indicators under three dimensions is constructed. The multi-objective optimization by ratio analysis plus full multiplicative form (MULTIMOORA) –Level-based weight assessment (LBWA) comprehensive evaluation model based on Pythagorean fuzzy number (PFN) is proposed to evaluate the dynamic technological innovation ability of high-tech enterprises. Finally, the accuracy and reliability of the model are verified by case analysis. The result of this study shows that the ability to identify new technological knowledge and information outside the enterprise, the ability to obtain technological innovation resources, and the ability to strengthen the input of innovation resources are important factors for the dynamic technological innovation capability of enterprises, so enterprises should pay more attention from these aspects. This study provides a new comprehensive evaluation model and evaluation results can help the decision-makers find their strengths and weaknesses in time and improve them, to promote the sustainable development of high-tech enterprises. Show more

Keywords: Dynamic capability, Pythagorean fuzzy set, LBWA, MULTIMOORA, high-tech enterprises innovation

DOI: 10.3233/JIFS-222965

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9143-9165, 2023

Authors: Jin, Feifei | Li, Danning | Guo, Shuyan | Zhou, Ligang | Chen, Yi | Zhu, Jiaming

Article Type: Research Article

Abstract: Under the Pythagorean fuzzy environment, this paper presents a multi-attribute decision-making (MADM) model based on exponential entropy measure and exponential similarity measure to evaluate new energy battery supplier’s performance. In this method, the notion of Pythagorean fuzzy linguistic sets (PFLSs) is first introduced by combining the linguistic fuzzy sets (LFSs) and the Pythagorean fuzzy sets (PFSs). Then, the axiomatic definitions of Pythagorean fuzzy entropy and Pythagorean fuzzy similarity measure are developed to measure the degree of uncertainty and similarity between two Pythagorean fuzzy linguistic values (PFLVs). The PFLVs can be expressed by the linguistic membership degree (LMD) and linguistic non-membership …degree (LNMD). In addition, we construct two new information measure formulas based on exponential function. Through a series of proofs, we verify that they satisfy the axiomatic conditions of entropy and similarity measure of Pythagorean fuzzy language respectively. On this basis, we research the relationship between the two information measures. Finally, we present a novel Pythagorean fuzzy linguistic MADM model. An example for evaluating performance of new energy battery supplier is given to explain the effectiveness of the newly-developed approach. The stability and validity of the newly-developed approach is performed by sensitivity analysis and comparative analysis. Show more

Keywords: Pythagorean fuzzy linguistic sets, information entropy, similarity measure, new energy battery supplier evaluation

DOI: 10.3233/JIFS-223088

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9167-9182, 2023

Authors: Choudhary, Ashutosh Kumar | Rahamatkar, Surendra

Article Type: Research Article

Abstract: DoS, GH, Sybil, Masquerading, Spoofing, Man in the Middle, etc. constantly attack IoT networks. Internal or external attacks reduce end-to-end delay, throughput, energy use, and other metrics. To counter these attacks, researchers have proposed a number of security & privacy mechanisms with varying computational complexity and security levels. Immutability, traceability, transparency, and distributed nature make blockchain-based models secure. QoS depends on blockchain length, so these models aren’t scalable. Researchers say sidechaining improves QoS while remaining secure. Splitting or merging complex sidechains requires machine learning. Low-power IoT networks can’t use models. This text suggests a lightweight MGWO Model that helps establish …initial routes by choosing high-trust nodes, reducing sidechaining power consumption, and incorporating fault-aware trust establishment. MGWO Model determines blockchain piece count for high QoS. MGWO Model uses Q-Learning to detect network faults. Fault identification is controlled by a stochastically modelled and activated Intrinsic Genetic Algorithm (IGA). Q-Learning, MGWO, and IGA can mitigate Sybil, Masquerading, Grey Hole, DDoS, and MITM attacks. Even when attacked, the proposed model maintains high QoS, improving real-time deployment efficiency. The proposed model improves energy efficiency by 15.9%, throughput by 10.6%, communication speed by 8.3%, and packet delivery by 0.8% for different network scenarios. Show more

Keywords: Trust, wireless, IoT, blockchain, sidechain, MITM, MGWO, Q learning, IGA, DDoS, Sybil, QoS

DOI: 10.3233/JIFS-223316

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9183-9201, 2023

Authors: Xiao, Yanjun | Zhao, Yue | Li, Zeyu | Wan, Feng

Article Type: Research Article

Abstract: Fault diagnosis of rapier loom is an inevitable requirement to meet the demand of intelligent manufacturing. Facing the strong noise interference caused by complex working environment, accurate and reliable vibration signal detection of blade loom spindle is the key to realize the rapier loom fault diagnosis. This paper proposes a method to extract the spindle vibration signal of the rapier loom by Adaptive Piecewise Hybrid Stochastic Resonance (APHSR) after the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN). Firstly, ICEEMDAN is used to pre-process the weak vibration signal containing noise, decompose the signal into multiple IMF components and …display the high and low frequency signal characteristics of the original signal. Then, the energy density method and the correlation coefficient method are used to remove high and low noise, respectively, to filter the optimal IMF components, and then the signal containing valid information is reconstructed. Finally, the reconstructed signal is input to APHSR for noise-assisted enhancement after scale transformation to restore the faint vibration signal feature frequencies and achieve effective feature extraction. Through the simulation experiment and the engineering fault experiment analysis, comparing ICEEMDAN-APHSR with CEEMDAN-SR, ICEEMDAN-SR, CEEMDAN-APHSR methods. The difference between the spectrum amplitude, the spectrum amplitude and the maximum noise and the maximum signal to noise ratio (SNR) of the fault feature frequency of the rapier loom spindle bearing increased by 3.3668 dB,1.7205 dB,2.3952 dB, respectively. The results show that ICEEMDAN-APHSR method can accurately extract the fault feature frequency of the spindle bearing of rapier loom, and effectively solves the problem of extracting the weak vibration signal feature of rapier loom in the background of strong noise. This method is beneficial to the future research of rapier loom fault diagnosis, and is of great significance to promote the maintenance of loom equipment and production safety and quality. Show more

Keywords: Weak signal detection, ICEEMDAN, APHSR, feature extraction, rapier loom, fault diagnosis

DOI: 10.3233/JIFS-223664

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9203-9230, 2023

Authors: Ma, Haishu | Ma, Zongzheng

Article Type: Research Article

Abstract: Unexpected failure of production equipment may lead to fatal accidents and economic losses of the enterprise. It is important to find out the cause and reason as soon as possible and take appropriate maintenance measures. Condition monitoring is often applied to predict equipment failures based on certain parameters. Moreover, when the parts of the rotating machinery fail, the vibration signals collected by the sensors are often mixed with a large amount of noise, which will cause difficulties for the accuracy and generalization of traditional fault diagnosis models. How to extract more effective feature information from complex vibration signals is of …indescribable importance for optimizing fault diagnosis models. In order to improve the accuracy of fault diagnosis in manufacturing system, a deep neural network model was proposed, which was validated on a blower. First, the vibration signal was collected using the sensors mounted on the blower. Then, wavelet packet decomposition and fast fourier transform were applied for feature extraction. Deep learning model was built using keras to diagnose the blower. The stacked Autoencoder is adopted in the DNN for dimension reduction. The extracted features are fed into the Multilayer Perceptron for fault diagnosis. Experimental results show that the proposed deep neural network model is able to predict the degradation of the mechanical equipment with high accuracy. Show more

Keywords: Deep neural network, wavelet packet decomposition, Fourier transform, feature extraction, fault diagnosis

DOI: 10.3233/JIFS-224077

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9231-9239, 2023

Authors: Shao, Yubo | Zhang, Bangcheng | Yin, Xiaojing | Gao, Zhi | Li, Jing

Article Type: Research Article

Abstract: The anomaly detection research of drive end bearings (DEBs) is of great significance to the safe and reliable operation of hoist. This paper proposes an anomaly detection method of DEBs based on the linear weighted sum combines with the belief rule base. First, in order to improve the accuracy of anomaly detection, the time-domain features and frequency-domain features are integrated by linear weighted sum (LWS) respectively. Then, belief rule base (BRB) method is provided for anomaly detection using fused features. Meanwhile, the covariance matrix adaption evolution strategy (CMA-ES) is utilized to optimize the parameters of belief rule base model. Finally, …the validity of the proposed method is verified by the vibration data, which are acquired from the condition monitoring system of hoist in body-in-white (BIW) welding production line. The proposed method achieves a high detection accuracy. It is proved that the proposed method is suitable for anomaly detection of DEBs in the actual BIW welding production line. Show more

Keywords: Anomaly detection, linear weighted sum, belief rule base, drive end bearing

DOI: 10.3233/JIFS-224102

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9241-9255, 2023

Authors: Liu, Peng | Geng, Xiaonan

Article Type: Research Article

Abstract: Coal is a vital basic energy source for any economy in the world, and our country is no exception. Our coal resources are abundant, with high production and demand, not comparable to oil and natural gas. The coal supply chain plays an equally important role in economic production, but unfortunately, the current coal supply chain is not focused on greening while creating profits. Unfortunately, the current coal supply chain does not focus on green production and energy conservation and emission reduction while creating profits, which has caused irreversible harm and loss to resources and environment. This has caused irreversible damage …and loss to resources and the environment. The green supplier selection for coal enterprises is affirmed as multiple attribute decision making (MADM). In such paper, motivated by the idea of cosine similarity measure (CSM), the CSMs are extended to DVNSs and four CSMs are created under DVNSs. Then, two weighted CSMs are built for MADM under DVNSs. Finally, a numerical example for Green supplier selection for coal enterprises is affirmed and some comparative algorithms are produced to affirm the built method. Show more

Keywords: Multiple attribute decision making (MADM), double-valued neutrosophic sets (DVNSs), cosine similarity measure (CSM), green supplier selection

DOI: 10.3233/JIFS-224123

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9257-9265, 2023

Authors: Chang, Kuei-Hu

Article Type: Research Article

Abstract: Risk prediction, assessment, and control are key parts of the successful operation and sustainable development of any enterprise. During the process of product failure risk assessment, evaluated risk factors belong to the group of multiple-criteria decision-making (MCDM) problems, including severity, occurrence, and detection when failure occurs. However, the traditional risk ranking method does not consider the subjective and objective weights of the assessment factors, and during risk prediction, assessment, and control, some unknown information in many practical situations is included. These reasons may cause the risk assessment results to be biased. In order to effectively deal with the problem of …risk assessment, this paper proposes a D numbers risk ranking method by considering subjective and objective weights between assessment factors under incomplete linguistic information. An illustrative example of screening unit failure risk assessment is used to explain and prove the rationality and correctness of the proposed method. Some risk ranking methods are compared with the proposed D numbers risk ranking method, and the simulation results present that the proposed ranking method handles the issue of incomplete information and provides more reasonable risk ranking results. Show more

Keywords: D numbers, risk ranking method, subjective weights and objective weights, multiple-criteria decision-making

DOI: 10.3233/JIFS-224139

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9267-9280, 2023

Authors: Abdul Lathif, Syed Ismail | Cruz Antony, J. | Noel Jeygar Robert, V. | Aishwarya, D.

Article Type: Research Article

Abstract: A failure risk assessment must be carried out and potential drilling equipment failure risks must be promptly addressed in order to prevent drilling fluid pollution during offshore oil drilling. The qualitative, comprehensive, and quantitative failure risks for Drilling Permanent Magnetic Synchronous Motors (DPMSM) are examined in this article using a hybrid methodology. First, the Drilling PMSM using Failure Mode Analysis (FMA) method is combined with the Risk Matrix (RM) approach to analyse the risk levels of risk factors individually. Next, the Borda number is introduced to compare the risk levels exactly. To execute a Fuzzy Comprehensive Evaluation (FCE) of the …system failure risk, a fuzzy relation matrix of risk factors is generated, and the weight of each risk component is calculated using importance analysis. The failure rate is then determined using fuzzy inference, and the Fault Tree (FT) is then built based on the risk variables. Fault tree analysis is used to compute the system failure rate, and the significance of the bottom event is evaluated. The Bayesian network (BN) is used to depict the Fuzzy Fault Tree (FFT) analysis. By utilizing Bayesian forward causal inference and reverse diagnostic inference to calculate the leaf node failure rate and root node posterior probability, the system’s weak points and potential failure causes are determined. Show more

Keywords: Risk matrix, fuzzy comprehensive evaluation, fault tree, bayesian network, failure mode analysis

DOI: 10.3233/JIFS-224462

Citation: Journal of Intelligent & Fuzzy Systems, vol. 44, no. 6, pp. 9281-9295, 2023