Dempster-Shafer Research Articles

Combining improved DFMEA with knowledge graph for component risk analysis of complex products

With the development of information technology in manufacturing, product lifecycle iteration is accelerating. Efficiently analyzing and identifying component risks is important to ensure the reliability of products. This paper proposes a new method that combines improved design failure mode and effect analysis (DFMEA) with the knowledge graph for component risk analysis of complex products. A risk evaluation knowledge graph is built for the risk identification and knowledge provision in the process of DFMEA. Graph analysis is used to identify to-be-evaluated objectives based on historical maintenance data, which helps designers avoid the heavy task of identification. The graph matrix of the historical evaluations is used to calculate the comprehensive weights. A Dempster-Shafer (DS) evidence fusion method based on comprehensive weight evaluation information is proposed for the effective fusion of evaluation evidence. As an alternative to the exact risk ranking, we offer a risk trade-off VIKOR (VIseKriterijumska Optimizacija I Kompromisno Resenje) method for components classification that classifies components into different risk classes for different risk treatment measures. Finally, the proposed method is validated using a tunnel boring machine as an example, and comparison and sensitivity analyses are performed to demonstrate that the risk value shows a consistent and stable change with expert evaluation.

A fuzzy rule extraction method based on Dempster–Shafer theory

A fuzzy rule extraction method based on Dempster–Shafer theory

Multi-Biometric System Based On The Fusion Of Fingerprint And Finger-Vein

Biometrics is the process of measuring the unique biological traits of an individual for identification and verification purposes. Multiple features are used to enhance the security and robustness of the system. This study concentrates exclusively on the finger and employs two modalities - fingerprint and finger vein. The proposed system utilizes feature extraction for finger vein and two matching algorithms, namely ridge-based matching, and minutiae-based matching, to derive matching scores for both biometrics. The scores from the two modalities are combined using four fusion approaches: holistic fusion, non-linear fusion, sum rule-based fusion, and Dempster-Shafer theory. The ultimate decision is made by the performance metrics and the Receiver Operating Characteristics (ROC) curve of the fusion technique with the best results. The proposed technique is tested on images collected from the “Nanjing University Posts and Telecommunications- Fingerprint and Finger vein dataset (NUPT-FPV).” According to the results, which were obtained for 840 input images the proposed system accomplishes the Equal Error Rate (EER) of 0% while using Dempster Shafer-based fusion and 14% while using the other three fusion techniques. Also, the False Acceptance Rate (FAR) is very low at 0% for all the fusion techniques which are crucial for security and preventing unauthorized access.

Improving customer satisfaction in the hotel industry by fusing multi-source user-generated content: An integration method based on the heuristic-systematic model and evidence theory

Improving customer satisfaction in the hotel industry by fusing multi-source user-generated content: An integration method based on the heuristic-systematic model and evidence theory

Large-Scale Land Cover Mapping Framework Based on Prior Product Label Generation: A Case Study of Cambodia

Large-Scale land cover mapping (LLCM) based on deep learning models necessitates a substantial number of high-precision sample datasets. However, the limited availability of such datasets poses challenges in regularly updating land cover products. A commonly referenced method involves utilizing prior products (PPs) as labels to achieve up-to-date land cover mapping. Nonetheless, the accuracy of PPs at the regional level remains uncertain, and the Remote Sensing Image (RSI) corresponding to the product is not publicly accessible. Consequently, the sample dataset constructed through geographic location matching may lack precision. Errors in such datasets are not only due to inherent product discrepancies, and can also arise from temporal and scale disparities between the RSI and PPs. In order to solve the above problems, this paper proposes an LLCM framework for generating labels for use with PPs. The framework consists of three main parts. First, initial generation of labels, in which the collected PPs are integrated based on D-S evidence theory and initial labels are obtained using the generated trust map. Second, for dynamic label correction, a two-stage training method based on initial labels is adopted. The correction model is pretrained in the first stage, then the confidence probability (CP) correction module of the dynamic threshold value and NDVI correction module are introduced in the second stage. The initial labels are iteratively corrected while the model is trained using the joint correction loss, with the corrected labels obtained after training. Finally, the classification model is trained using the corrected labels. Using the proposed land cover mapping framework, this study used PPs to produce a 10 m spatial resolution land cover map of Cambodia in 2020. The overall accuracy of the land cover map was 91.68% and the Kappa value was 0.8808. Based on these results, the proposed mapping framework can effectively use PPs to update medium-resolution large-scale land cover datasets, and provides a powerful solution for label acquisition in LLCM projects.

Combination of dependent and partially reliable Gaussian random fuzzy numbers

Gaussian random fuzzy numbers are random fuzzy sets generalizing Gaussian random variables and possibility distributions. They define belief functions on the real line that can be conveniently combined by the product-intersection rule under the independence assumption. In this paper, we introduce various extensions of this rule to account for dependence and partial reliability of the pieces of evidence. We first provide formulas for the combination of an arbitrary number of Gaussian random fuzzy numbers whose dependence is described by a correlation matrix, and we introduce a minimum-conflict combination operation. To account for partially reliable evidence, we then introduce two discounting operations called possibilistic and evidential discounting, as well as several combination operators based on different assumptions, each one parameterized by a correlation matrix and a vector of discounting coefficients. We demonstrate the application of these operators to the combination of predictions with different sets of inputs in machine learning, and show that performance can be enhanced by optimizing the parameters of the combination operators.

Research on the Three-Level Integrated Environmental Evaluation Model for Multi-Greenhouse Potatoes

Aiming at the problems of large error and redundancy in the multi-node data acquisition of multi-greenhouse photo growth environmental information, a three-level fusion algorithm based on adaptive weighting, an LMBP network, and an improved D-S theory is proposed. The box-and-line graph method recognizes the original data and then replaces it based on the mean value method; the air temperature, humidity, and light intensity measurements are unbiased estimations of the true value to be estimated, so the first level of fusion chooses the adaptive weighted average algorithm to find the optimal weights of each sensor under the condition of minimizing the total mean-square error and obtains the optimal estimation of the weights of the homogeneous sensors of a greenhouse. The Levenberg–Marquardt algorithm was chosen for the second level of fusion to optimize the weight modification of the BP neural network, i.e., the LMBP network, and the three environmental factors corresponding to “suitable”, “uncertain” and “unsuitable” potato growth environments were trained for the three environmental factors in the reproductive periods. The output of the hidden layer was converted into probability by the Softmax function. The third level is based on the global fusion of evidence theory (also known as D-S theory), and the network output is used as evidence to obtain a consistent description of the multi-greenhouse potato cultivation environment and the overall scheduling of farming activities, which better solves the problem of the difficulty in obtaining basic probability assignments in the evidence theory; in the case of a conflict between the evidence, the BPA of the conflicting evidence is reallocated, i.e., the D-S theory is improved. Example validation shows that the total mean square error of the adaptive weighted fusion value is smaller than the variance of each sensor estimation, and sensors with lower variance are assigned lower weights, which makes the fusion result not have a large deviation due to the failure of individual sensors; when the fusion result of a greenhouse feature level is “unsuitable”, the fusion result of each data level is considered comprehensively, and the remote control agency makes a decision, which makes full use of the complementary nature of multi-sensor information resources and solves the problem of fusion of multi-source environmental information and the problem of combining conflicting environmental evaluation factors. Compared with the traditional D-S theory, the improved D-S theory reduces the probability of the “uncertainty” index in the fusion result again. The three-level fusion algorithm in this paper does not sacrifice data accuracy and greatly reduces the noise and redundancy of the original data, laying a foundation for big data analysis.

Evaluation of marine ranching socio-ecological system resilience in China

Marine ranching socio-ecological system (SES) resilience refers to how well a marine ranching system can withstand disturbances and bounce back while maintaining its levels of environmental restoration, resource conservation, and sustainable production. However, there is insufficient research on the evaluation of the resilience of marine ranching SESs. To fill this gap and determine how to effectively evaluate marine ranching SES resilience, this study developed an index-based evaluation system using the DPSIR framework. An evaluation model (abbreviated as DTED) was then constructed by combining four methods, namely DPSIR, triangular fuzzy numbers, entropy weighting, and the Dempster-Shafer theory of evidence. The developed DTED model was then applied to the data obtained from 34 marine ranches in China for the period from 2019 to 2022 to determine their corresponding SES resilience. The results showed that most of China's marine ranches have a medium level of SES resilience, with an overall upward trend. Nevertheless, there is still a possibility of risk, and measures should be taken to promote ecological diversification in marine ranching, mitigate impacts on social systems, and enhance the resilience of marine ranching SESs.

A hybrid method for asynchronous detection of motor imagery electroencephalogram fusing alpha rhythm and movement-related cortical potential

The development of brain–computer interfaces (BCIs) enables direct human–computer interaction by real-time monitoring and translation of brain signals. Motor imagery electroencephalography (MI-EEG) systems, known for their non-invasiveness and user-friendliness, are particularly promising. Asynchronous systems, offering enhanced flexibility, represent the future of practical BCI applications. However, existing asynchronous detection methods in MI-EEG systems have yet to achieve satisfactory accuracy and latency. This paper proposes a hybrid asynchronous detection method that combines alpha rhythm changes and movement-related cortical potential (MRCP) features based on weighted Dempster–Shafer theory (AMAD-DS). The AMAD-DS method employs a hybrid architecture and two multi-domain joint analysis algorithms to process EEG signals from different areas: detecting alpha rhythm features in the occipital area and MRCP features in the sensorimotor area. The method fuses the results of these detections at the decision level using weighted D–S theory to produce the final output. Experiments conducted on a MI-EEG-based BCI system demonstrate that AMAD-DS outperformed methods using only MRCP or alpha rhythm features, improving the true positive rate by 12.6%, reducing the false positive rate by 0.5 FPs/min, and ensuring that the detection time of motor imagery onset is less than 500 ms. Online experiments further validate the method’s effectiveness, achieving a true positive rate of 91.1% and a false positive rate of 0.16 FPs/min.

Hydrogen leakage risk assessment of HECS based on dynamic bayesian network

As a new energy system, the hydrogen-electric coupling system（HECS） involves complex conditions with multiple hazards such as electricity, hydrogen, heat, and high pressure, and there are complex safety risk issues. This paper proposes a hydrogen leakage risk assessment method of HECS based on dynamic Bayesian networks (DBN). Firstly, the bow-tie model is mapped to the DBN model, and the prior probability of the DBN model is obtained using intuitionistic fuzzy theory, Dempster-Shafer evidence theory, and an improved similarity aggregation algorithm. Then, leakage noise or gate models are introduced to improve the conditional probability table of the DBN model, and Weibull distributions are used to construct the probability transfer process between different time periods. Finally, through practical case studies, it is shown that the proposed method can effectively predict the dynamic probability and potential consequences of failure events through time series information. Sensitivity analysis proves the practicality and robustness of the proposed method. The research results can provide theoretical support for risk assessment and daily maintenance of HECS.

New belief divergence measure based on cosine function in evidence theory and application to multisource information fusion

Multisource information fusion technology significantly benefits from using information across various sources for decision-making, particularly by leveraging evidence theory to manage uncertain information efficiently. Nonetheless, dealing with highly conflicting evidence presents a considerable challenge. To tackle this issue, this paper introduces a new belief divergence measure within the framework of evidence theory. The proposed measure, which incorporates the cosine function and pignistic probability transformation, is adept at quantifying the disparity between the evidences while maintaining key properties, such as boundedness, non-degeneracy and symmetry. Moreover, building upon the concepts of proposed belief divergence and belief entropy, this paper further proposes a new fusion method that employs a weighted evidence average prior to the application of Dempster’s rule. The performance of the proposed method is validated on several applications, and the results demonstrate its superior ability to absorb highly conflicting evidence compared with existing methods.

An integrated knowledge and data model for adaptive diagnosis of lubricant conditions

Lubricant condition diagnosis often encounters conflicting conclusions due to the reverse degradation of indicators and coupling failures. To address this issue, a knowledge-guided adaptive diagnosis (KG-ADLC) method is developed with expert knowledge. Based on lubricant indicators in three attributes, a confidence-driven failure inference mechanism is developed by integrating rule-based reasoning with evidence theory. Furthermore, the KG-ADLC model is established that employs the inference mechanism to guide the synchronous evaluation of lubricant failures. For verification, the developed model is tested with the samples from the simulation experiment and authentic aero-engines. Experimental results reveal that the constructed model can accurately identify six types of lubricant conditions. Notably, the average accuracy has been improved from 74.8 % to 93.7 % when compared to existing methods.

A New Fuzzy Bayesian Inference Approach for Risk Assessments

Bayesian network (BN) inference is an important statistical tool with additive symmetry. However, BN inference cannot deal with the uncertain, fuzzy, random, and conflicting information from experts’ knowledge in the process of conducting a risk assessment. To tackle this issue, a new fuzzy BN inference approach for risk assessments was proposed based on cloud model (CM), interval type-2 fuzzy set (IT2 FS), interval type-2 fuzzy logic system (IT2 FLS), modified Dempster–Shafer (D-S) evidence theory (ET), and Latin hypercube sampling (LHS) methods along the following lines. Firstly, CM was integrated into IT2 FS, and CM-based IT2 FS (CM-IT2 FS) was defined in the IT2 FLS. Secondly, modified D-S ET was utilized to determine the CM-IT2 FS-based a priori probabilities, and the CM-IT2 FS-based BN model was established. Thirdly, the CM-IT2 FS-based a priori probabilities were reduced to the CM-IT1 FS-based ones using a type reducer in the IT2 FLS, LHS was applied to propose a new fuzzy BN inference algorithm, and then, the new algorithm was used in a typical case to perform the fuzzy BN positive inference for risk prediction and the fuzzy BN reverse inference for risk sensitivity analysis. Finally, the BN inference results were analyzed using the proposed algorithm and the two common BN inference algorithms, and the effectiveness of the proposed approach was validated. It can be concluded that the proposed approach was both accurate and promising.

Evidence representation of uncertain information on a frame of discernment with semantic association

Belief functions as a powerful model to represent and deal with uncertain information are widely used in information fusion. However, semantic association within a frame of discernment is not well defined in traditional framework of belief function theory. To solve the problem, in this work models and methods for evidence representation of uncertain information on a frame of discernment with semantic association are studied. These contributions are made in the study. At first, a formal definition for the concept of semantic association is proposed via an axiomatic manner. Second, new evidence representation models including belief, plausibility and commonality measures, are designed with the consideration of semantic association. Third, a novel evidence discounting operation, called associative discounting, is proposed to amend original evidence in terms of more refined meta-knowledge regarding the reliability of information sources. The effectiveness of proposed models and methods is verified through practical applications on UCI (University of California, Irvine) data sets for the combination of multiple classifiers. This work, on the one hand, has successfully imported the concept of semantic association into the framework of belief function theory; on the other hand, it provides a new scheme to correct original information in a more refined manner.

An evidence theory-based uncertainty quantification method of building thermal parameters for accurate reliability assessment of building air-conditioning design loads

Probabilistic methods are frequently employed in the reliability assessment of building air-conditioning design loads to quantify the uncertainty associated with building thermal parameters. However, its implementation requires some assumptions because of inadequate data in the early stages of design. These assumptions cause the uncertainty quantification results of the building thermal parameters to differ from reality, which may lead to a significant overestimation or underestimate of reliability. Therefore, an evidence theory-based method was proposed to quantify the uncertainty of the building thermal parameters. In this method, evidence theory was employed instead of probability theory because the former could make full use of small sample information to achieve objective uncertainty quantification. Taking an office building in Tianjin, China, as an example, the new method was tested, and its effectiveness was verified. The results showed that the uncertainty quantification results of the building thermal parameters based on the new method included those of the benchmark method. Under small sample conditions, the reliability assessment results based on the new method were closer to those of the benchmark compared with the traditional method.

A Multi-Criteria Assessment Model for Cooperative Technology Transfer Projects from Universities to Industries

Cooperative Technology Transfer (CTT) is a technology transfer model where universities and enterprises jointly participate throughout the entire process of technology transfer activities. Most discussions focus on its mechanisms and influencing factors, yet a framework to guide CTT projects in practice is still lacking. This study proposes an assessment model based on the life-cycle of CTT projects, covering the initial cooperation relationship, project management during the mid-term, and technological achievements at the end. The model was evaluated by 14 experts first and then validated through two CTT projects in China. Gray Relation Analysis was employed to calculate the weights of different factors based on their relative importance, while the Dempster–Shafer theory was utilized to combine evidence from various sources and address the uncertainty in the assessment. The results of the case analysis indicate that the attitudes of universities and enterprises are considered critical in influencing the success of CTT projects, while management issues that arise during the projects can pose potential risks. This research serves as an applied exploration and has three functions. Firstly, the model can be used as a feasibility study before the project commences. Secondly, it can be utilized to analyze and improve potential issues during the project. Finally, it can be used for a post-project experience summary.

Design of network security monitoring system based on CNN and exponential weighted D-S evidence theory

ABSTRACT To address the poor network security protection and low attack traffic identification, the study designs a network security monitoring system based on convolutional neural network and exponential weighted Dempster Shafer evidence theory. The validation of the UNSW-NB15 data set showed that the output of multi-source fusion after exponential weighted Dempster Shafer evidence theory was higher than the output of the feature fusion by 3.92%. The accuracy of the attack recognition was as high as 93.72%, which was 1.85% higher than feature fusion. The accuracy of the proposed network security monitoring system increased by 3.70% on average over other methods. The results indicate that the proposed network security monitoring system can effectively improve the efficiency of network attack identification, monitor network security in real-time, and effectively protect network operation. The system is feasible and reasonable in terms of network security situational awareness, which can provide effective situational analysis for network administrators.

Frequentist belief update under ambiguous evidence in social networks

In this paper, we study a frequentist approach to belief updating in the framework of Dempster-Shafer Theory (DST). We propose several mechanisms that allow the gathering of possibly ambiguous pieces of evidence over time to obtain a belief mass assignment. We then use our approach to study the impact of ambiguous evidence on the belief distribution of agents in social networks. We illustrate our approach by taking three representative situations. In the first one, we suppose that there is an unknown state of nature, and agents form belief in the set of possible states. Nature constantly sends a signal which reflects the true state with some probability but which can also be ambiguous. In the second situation, there is no ground truth, and agents are against or in favor of some ethical or societal issues. In the third situation, there is no ground state either, but agents have opinions on left, center, and right political parties. We show that our approach can model various phenomena often observed in social networks, such as polarization or bounded confidence effects.

Mixed uncertainty analysis on pumping by peristaltic hearts using Dempster-Shafer theory.

In this paper, we introduce the numerical strategy for mixed uncertainty propagation based on probability and Dempster-Shafer theories, and apply it to the computational model of peristalsis in a heart-pumping system. Specifically, the stochastic uncertainty in the system is represented with random variables while epistemic uncertainty is represented using non-probabilistic uncertain variables with belief functions. The mixed uncertainty is propagated through the system, resulting in the uncertainty in the chosen quantities of interest (QoI, such as flow volume, cost of transport and work). With the introduced numerical method, the uncertainty in the statistics of QoIs will be represented using belief functions. With three representative probability distributions consistent with the belief structure, global sensitivity analysis has also been implemented to identify important uncertain factors and the results have been compared between different peristalsis models. To reduce the computational cost, physics constrained generalized polynomial chaos method is adopted to construct cheaper surrogates as approximations for the full simulation.

A novel multiple linearization method for reliability analysis based on evidence theory

Evidence theory has been applied in structural reliability assessment to address epistemic uncertainties, given its capability to deal with imprecise, incomplete and even contradictory information. An important challenge in structural reliability assessment based on the framework of evidence theory is its expensive computational cost in multidimensional problems. In this work, a modified multiple linearization method is proposed to identify a set of hyperplane approximations to the limit state function. It provides a clear guidance for the hyperplanes by evaluating curvature magnitude of the original limit state at the most probable point along each axis in the rotated standard normal space, and shows good stability and satisfactory accuracy in nonlinear reliability analysis based on evidence theory. The belief measure and plausibility measure of the structure are assessed by the hyperplanes through the intersection or union of the failure domains. In the meantime, the classification and division of focal elements in failure set, safety set and uncertain set will be made through the segment edge of the failure and safety domain to reduce the computation burden. Finally, four mathematical examples and engineering cases are explored to illustrate the methodology and show its benefits.