Evidential Reasoning Approach Research Articles

A location-allocation model for quality-based blood supply chain under IER uncertainty

Providing blood with high quality at the lowest cost and the shortest time is main challenge of blood supply chain management. This paper presents a new model for designing a dynamic and three level blood supply chain incorporating the quality issues. The proposed model intends to locate facilities, and to determine the best strategy for blood allocation by minimizing both cost and time and maximizing the customer satisfaction based on quality of blood delivery. In order to deal with consideration of real world, intricacies such as blood freshness, both separation and apheresis extraction methods, Cross match to Transfusion ratio (C/T) and equipment failure have been involved. Also, Interval Evidential Reasoning (IER) approach is applied to handle the uncertainty of blood product demand. Since the proposed model is NP-hard, MOPSO and NSGAII algorithms are utilized to solve it. Finally, to demonstrate the applicability of the problem some numerical examples are designed in different sizes and the most favorable algorithm is determined using TOPSIS method.

Belief-Rule-Base Inference Method Based on Gradient Descent With Momentum

The belief-rule-base (BRB) inference methodology using the evidential reasoning (ER) approach is widely used in different fields, such as fault diagnosis, system identification, and decision analysis. However, the calculation characteristic of the conventional rule activation weight makes the inference system have the rule zero activation problem. The difficulty of constructing partial derivatives restricts the optimization of parameters using the gradient method. Hence, this paper proposes a new belief rule structure and its gradient training method to solve the rule zero activation problem during the inference process and improve inference accuracy. The Gaussian function is applied to calculate the activation weight of the rule with the new structure. Its characteristics avoid the zero activation problem caused by the attribute reference value set in the original method. Based on the newly proposed method, the corresponding distance-sensitive parameter is set for each attribute, and the weight parameter of each rule is discarded. It simplifies the calculation of rule activation weights in the inference process and enables the partial derivatives of the parameters of the inference system to be easily constructed. In the parameter optimization, the momentum optimization gradient stochastic descent method is used to train the new BRB system, which improves the training speed and accuracy compared with the conventional methods. Experiments with nonlinear function fitting, oil pipeline leak detection, and classification of several public datasets are carried out to verify whether the new BRB system trained with momentum stochastic gradient descent (SGDM-BRB) has better performance than other conventional methods. The experimental results show that in the case of complete data, SGDM-BRB has higher accuracy and faster training speed than the conventional methods.

Collision risk assessment in Jiangsu section of the Yangtze River based on evidential reasoning

Collision between ships is one of the dominant types of accidents in the Jiangsu section of the Yangtze River, accounting for over 60% of the accidents. An evidential reasoning (ER) approach is introduced to perform a quantitative assessment of the safety of the whole waterway by dividing it into 17 sub-sections. The Risk Influencing Factors (RIFs) including channel condition, navigation environment and navigation aids conditions are considered and further decomposed into several sub-factors. The expert knowledge is used to quantify the relative importance of the RIFs to the collision risk. The historical data is used to make the Basic Probability Assignments (BPAs) of the belief structures. The hazard index (HI) is used as a measure of collision risk. The results indicate that Kouanzhi Waters and Jiaoshan Waters carry much higher collision risk than other waters, whereas Nanjing Waters and Fanjiafan Waters have the lowest collision risk. The results are useful for the maritime safety management in the Jiangsu section of the Yangtze River.

A multi-criteria decision model based on the evidential reasoning approach for the selection of fulcrum ports supporting Arctic shipping through the Northeast Passage

ABSTRACT As a result of the opening Arctic waters, Arctic Northeast Passage (NEP) can be used as a shortcut between Asian and European ports, which may bring considerable economic benefits. Due to the harsh environment and substantial costs, it is difficult to construct many fulcrum ports along the Russian Arctic coast, which provide support services for ships navigating the NEP. A multi-criteria decision model is hence constructed to select fulcrum ports among candidates to support Arctic shipping through the NEP. Based on the model, the evidential reasoning (ER) approach is used to aggregate quantitative data and qualitative expert judgments to generate distributed assessments for candidate ports, as well as modeling uncertainties such as incompleteness and ignorance in the assessment of candidate ports. Their suitability as fulcrum ports are prioritized according to the utilities. The results show that Pevek, Sabetta, and Murmansk are the ideal fulcrum ports for the eastern, central, and western section, respectively, of the Russian Arctic coast. The associated policy recommendations include prioritizing the improvement of selected fulcrum ports, attaching importance to promising candidate ports, updating outdated infrastructure, and implementing more related projects. A sensitivity analysis conducted in this study validates the feasibility and reasonability of the proposed model.

Performance assessment framework based on competitive priorities for sustainable freight transportation systems

Stringent environmental norms and growing demands for energy-efficient and cost-effective freight shipping have put organizations under immense pressure to consider sustainability aspects while assessing their performance. Moreover, amid growing awareness about competitive priorities, it has become prudent to understand how an organization’s sustainability performance is balanced against competitive priorities. Therefore, this study attempts to develop an integrated performance assessment framework (PAF) on the basis of competitive priorities for sustainable freight transportation (SFT) systems. A unified approach consisting of fuzzy group decision-making, fuzzy evidential reasoning approach, and expected utility concept is utilized to assess critical success factors of SFT based on four competitive priorities. The applicability of the developed model is exhibited through a case example, and sensitivity analysis is also performed to examine its robustness. This competitive performance score based PAF would facilitate logistics professionals in selecting the most competitive SFT system.

Bid or No-Bid Decision Model for International Construction Projects: Evidential Reasoning Approach

The decision to bid or not bid on international construction project tenders is highly complex, due to the many uncertain factors associated with such projects. This study proposes a multiattribute bid or no-bid decision model based on the evidential reasoning (ER) approach, which effectively deals with the uncertainty arising from qualitative criteria and incomplete information. Evaluation of the bid or no-bid decision is divided into six stages according to the objectives set in the decision framework, as proposed in an earlier study. The ER approach plays a triple role in the model, improving the method for ascertaining the weights of criteria, aggregating the qualitative and incomplete information assessments of experts, and fusing the value of lower attributes to an upper attribute. The decision model is validated as feasible through the successful application of the model in a case project and verified as practical by an experimental case study. Although this decision model has certain limitations, it offers an innovative way of solving the bid or no-bid decision problem within the international construction setting. This study contributes to the body of knowledge in international construction management by providing a validated bid or no-bid decision model tailored to the specific considerations required of overseas projects and enriches the knowledge and application of the ER approach.

New uncertainty modelling for cargo stowage plans of general cargo ships

The current approach to the cargo stowage plans (CSP) of general cargo ships (GCS) is safety-driven, which means that any CSP satisfying minimum safety requirements can be used in practice. Such an approach taking into account no economic and environmental concerns cannot help sustain GCS growth in today’s competitive freight transportation market. This paper introduces a revised evidential reasoning (ER) approach to cope with the complex decision-making problem associated with the CSP of GCS. The complexity mainly results from the dynamic interdependency between the decision criteria and alternatives. The revised ER can determine the functions of the safety-related criteria in the decision making process by considering the extent to which each decision alternative meets the minimum safety requirements. The model is tested in multiple forms by an empirical study using a national GSC loading laboratory, and a real-life application by a shipping company in practice. The results reveals that the new model can aid general ship owners to make sustainable CSPs from a multiple-dimensional perspective and select an optimal CSP based on specific voyage scenarios.

Emergency material reserve location selection by a time-series-based evidential reasoning approach under bounded rationality

Location selection of emergency facilities is a vital issue in the disaster management, which can be considered as a multi-criteria decision making problem with uncertainty. In this regard, the evidential reasoning (ER) approach can be used for such a multiple criteria analysis with uncertain information. However, the original ER approach neither took into account the bounded rationality of experts, nor examined the information at different time nodes, which may not conform to the actual situation of location selection. To solve this problem, we propose a time-series-based ER approach under the bounded rationality of experts to make full use of all evidence. Firstly, we propose a prospect-theory-based modification function to adjust the evidence given by experts and reduce the corresponding subjectivity. Considering the information at different time, we then propose a time-series-based ER approach to generate the comprehensive evidence of each alternative over the whole period. Besides, a VIKOR-based ranking method is proposed to improve the way to rank alternatives in the original ER approach. Finally, the proposed method is applied in the location selection of the emergency material reserve and the corresponding comparative analyses are given to show the rationality of the proposed method.

A Prospect Theory-Based Evidential Reasoning Approach for Multi-expert Multi-criteria Decision-Making with Uncertainty Considering the Psychological Cognition of Experts

To model and deal with complex decision-making problems, many methods have been proposed in the past decades. However, how to deal with probabilistic uncertainty is still an open question. In this regard, this paper introduces a prospect theory-based evidential reasoning (ER) approach to process such uncertainty in multi-expert multi-criteria decision-making. Firstly, we introduce a novel way to model the uncertainty by assigning the uncertain belief to the envelope of evidence, which makes the expression consistent with experts’ perceptions. The ER approach is used to aggregate individual decision-making matrices with probabilistic uncertainty to obtain a collective matrix. Then, the ER approach is applied again on the obtained collective matrix to aggregate the evaluations under different criteria to generate the comprehensive evaluation of each alternative. Given that the decision-maker may have different risk preferences to gain and loss from those of the experts, we introduce a prospect function of evaluation grades to rank alternatives. In this way, we can obtain the ranking of alternatives which considers both the evaluations of experts and the risk preference of the decision-maker. The advantages of this method are highlighted by a case study concerning the evaluation of sustainable development ability of enterprises.

Performance evaluation of complex systems using evidential reasoning approach with uncertain parameters

The composition of the modern aerospace system becomes more and more complex. The performance degradation of any device in the system may cause it difficult for the whole system to keep normal working states. Therefore, it is essential to evaluate the performance of complex aerospace systems. In this paper, the performance evaluation of complex aerospace systems is regarded as a Multi-Attribute Decision Analysis (MADA) problem. Based on the structure and working principle of the system, a new Evidential Reasoning (ER) based approach with uncertain parameters is proposed to construct a nonlinear optimization model to evaluate the system performance. In the model, the interval form is used to express the uncertainty, such as error in testing data and inaccuracy in expert knowledge. In order to analyze the subsystems that have a great impact on the performance of the system, the sensitivity analysis of the evaluation result is carried out, and the corresponding maintenance strategy is proposed. For a type of Inertial Measurement Unit (IMU) used in a rocket, the proposed method is employed to evaluate its performance. Then, the parameter sensitivity of the evaluation result is analyzed, and the main factors affecting the performance of IMU are obtained. Finally, the comparative study shows the effectiveness of the proposed method.

Performance Assessment of R&D-Intensive Manufacturing Companies on Dynamic Capabilities

In today's business landscape, improving competitive advantage of manufacturing companies depends on their continuous performance improvement. This necessitates a generic and multi-dimensional view that organisational and managerial processes should be assessed by the underlying micro-foundation of dynamic capabilities (DC) in conjunction with enhanced new product development (NPD) projects. This study aims to propose an operationalised model of the conceptual DC framework including sensing, seizing, and reconfiguration capacities. The advantage of the two aforementioned models, which are based on a multi-criteria decision analysis (MCDA) framework, is that they can assist managers in the automotive industry to identify improvement plans and goals for sound and robust decision making. For this purpose, the evidential reasoning (ER) approach, which is realised in the intelligent decision system (IDS) software tool, is employed to perform performance self-assessment for the selected manufacturing companies on DC. This study provides managers with a useful tool to assess their company's strengths and weaknesses in regard to the DC components.

A novel model used for assessing supply chain sustainability integrating the ANP and ER approaches and its application in marine ranching

The evolution of competition has set greater demands for supply chain sustainability. As the existing models mostly consider the issue only by employing the triple bottom line (TBL), this paper aims to establish a novel supply chain sustainability assessment model. Firstly, this paper establishes the indicator system from dimensions of TBL and supply chain capability, and integrates innovation and value co-creation. Innovation can help enterprises improve sustainability and benefit from it. Value co-creation has also become a new source of sustainable competitive edge. Compared with the previous qualitative research, the combination of quantitative and qualitative indicators used in this paper can more objectively reflect the subject. Secondly, this paper employs the Analytic Network Process (ANP) method to determine the weight of indicator. This method breaks the hypothesis of independence and considers the interactions among indicators. Thirdly, the Evidential Reasoning (ER) method is used to process expert evaluation information of qualitative indicators by membership function. This method incorporates the idea of fuzziness, allows experts to make no evaluation on unfamiliar indicators. Finally, this paper validates the practicability through a case study of marine ranching. The contribution lies in that this paper provides a novel theoretical model for supply chain sustainability assessment.

Generalized combination rule for evidential reasoning approach and Dempster–Shafer theory of evidence

The Dempster–Shafer (DS) theory of evidence can combine evidence with one parameter. The evidential reasoning (ER) approach is an extension of DS theory that can combine evidence with two parameters (weights and reliabilities). However, it has three infeasible aspects: reliability dependence, unreliability effectiveness, and intergeneration inconsistency. This study aimed to establish a generalized combination (GC) rule with both weight and reliability, where ER and DS can be viewed as two particular cases, and the problems of infeasibility of the parameters can be solved. In this paper, the infeasibilities of ER are analyzed, and a generalized discounting method is introduced to reasonably discount the belief distributions of the evidence using both the weight and the reliability. A GC rule is then constructed to combine evidence by means of the orthogonal sum operation, and the corresponding theorems and corollaries are provided. Finally, the superiority of the GC rule is shown through numerical comparisons and discussion, and an illustrative example is provided to demonstrate its applicability.

Realising advanced risk assessment of vessel traffic flows near offshore wind farms

Offshore wind farms (OWFs) are relatively new installations at sea. Accident records related to vessel collisions with OWFs are insufficient to support a full quantitative risk analysis using traditional probabilistic approaches. This paper aims to develop a semi-qualitative risk model to assess the vessel-turbine collision risks by incorporating Bayesian networks (BN) with evidential reasoning (ER) approaches. First, a BN is trained based on Automatic Identification Systems (AIS) data to characterise real vessel traffic flows, including the detailed information and relationships between traffic flow parameters. Secondly, through synthesising expert judgements by ER, five risk factors influencing the probability and consequence of vessel-turbine collisions are identified (incl. the associated conditional probabilities) in the established BN. Finally, the updated BN with ER input is tested through ten real scenarios and validated by processing a validity framework. This paper pioneers the use of multi-data-driven BNs to characterise traffic flows and assess vessel-turbine collision risk for navigational safety assurance near OWFs. The research findings provide empirical evidence of using ER to supplement BN subjective data to advance its applications in risk analysis.

A Probabilistic Linguistic Group Decision-Making Method From a Reliability Perspective Based on Evidential Reasoning

A probabilistic linguistic multiattribute group decision-making (PLMAGDM) problem is studied from a reliability perspective based on an evidential reasoning approach and linguistic granulation optimization. As decision-making reliability, indicating the validity and accuracy, greatly depends on the reliabilities of both group similarity and the degree of familiarity, it is helpful to study a decision-making problem from the perspective of information reliability. To this end, a PLMAGDM method which considers reliability is put forward in this paper. First, two sets of information (i.e., probabilistic linguistic terms on alternative grades, and linguistic terms on familiarity) are given by experts as measures of expert reliability and attribute reliability. A granulation optimization model is then constructed to generate the utility of grades while maximizing group similarity, and a model that redistributes the probabilities of grades is established for the situation of incomplete information. We further provide another granulation optimization model to generate the degree of familiarity. Additionally, using the evidential reasoning rule, the aggregated weights of experts are obtained employing an optimization method that is used to retrieve attribute information. Lastly, through the method comparison in two cases as well as the sensitivity analysis, the validity and applicability of the method are verified.

An evidential reasoning approach based on risk attitude and criterion reliability

In the evidential reasoning (ER) approach, in addition to the weight of a criterion, reliability is another important concept in connection with the criterion although it attracts little attention. Additionally, a decision maker’s risk attitude is of particular importance in decision-making. To simultaneously consider these two important factors, this paper proposes a new ER approach. In the approach, with the consideration of a decision maker’s risk attitude, the combinational reliability of each criterion for each alternative is constructed from the original reliability of the criterion. By following the regression idea to learn the original reliability of each criterion, a unified optimization model is constructed, in which the maximum difference between combinational reliabilities and their estimations is minimized. Within the post-optimal solution space of criterion reliabilities found by solving the unified model, another optimization model is constructed, from which the minimum and maximum expected utilities of each alternative are determined. Solutions to multi-criteria decision-making problems are then generated from the expected utilities. The proposed approach is used to analyze a material supplier selection problem for a company located in Tongling, Anhui, China. The analysis of the problem demonstrates the applicability of the proposed approach.

A method of automatically generating initial parameters for large-scale belief rule base

The traditional rule-base inference methodology using evidential reasoning approach (RIMER) needs to traverse the reference values of all antecedent attributes when constructing belief rule base (BRB). Therefore, when many attributes have multiple reference levels, the scale of BRB will face the problem of combination explosion. Thus, it is inoperable to require experts to give all the parameters of each rule of BRB model, because of the limitation of expert knowledge in complex problems. If the initial BRB parameters are set unreasonably, the optimization speed and accuracy of the model will be affected. To solve this problem, a method of automatically generating initial parameters for large-scale BRB by using part of the standard rules and cloud model is proposed in this paper. Experts determine the number and parameters of standard rules through prior knowledge based on specific practical problems, and use cloud models to convert qualitative knowledge and quantitative information to automatically generate the remaining rules. A case study is established with Intel Berkeley Research lab data set to verify the effectiveness of the proposed method.

Improving Driver Assistance in Intelligent Transportation Systems: An Agent-Based Evidential Reasoning Approach

Providing accurate real-time traffic information is an inherent problem for intelligent transportation systems (ITS). In order to improve the knowledge base of advanced driver assistance systems (ADAS), ITS are strongly concerned with data fusion techniques of all kinds of sensors deployed over the traffic network. Driver assistance is devoid of a comprehensive evidential reasoning system on contextual information, more specifically when a combination involves inside and outside sensory information of the driving environment. In this paper, we propose a novel agent-based evidential reasoning system using contextual information. Based on a series of information handling techniques, specifically, the belief functions theory and heuristic inference operations to achieve a consensus about daily driving activity in automatically inferring. That is quite different from other existing proposals, as it deals jointly with the driving behavior and the driving environment conditions. A case study including various scenarios of experiments is introduced to estimate behavioral information based on synthetic data for prediction, prescription, and policy analysis. Our experiments show promising, thought-provoking results encouraging further research.

Regression model for estimation of attribute reliability

In this paper, a regression model is developed to estimate attribute reliability in the evidential reasoning (ER) context. By analysing the difference between attribute weight and attribute reliability, a general qualitative definition of attribute reliability is provided. The reliability of an attribute is quantitatively measured in consistence with the qualitative definition in the context of the ER approach. A regression model is then constructed to generate attribute reliabilities by minimising the maximum differences between the real value of attribute reliability and its estimation. Within the post-optimal solution space of attribute reliabilities, an optimisation model is constructed to determine the expected utilities of each alternative in order to generate solutions to multiple attribute decision analysis problems. A sale place selection problem in Qingyang County of Chizhou in Anhui province of China is analysed using the proposed regression model to demonstrate its detailed implementation process, validity and applicability.

Evaluation Information Fusion of Scientific Research Project Based on Evidential Reasoning Approach Under Two-Dimensional Frames of Discernment

Owing to the complexity and uncertainty of scientific research projects, government-funded institutions often invite multiple peer experts to evaluate the projects. How to aggregate the evaluation opinions of multiple peer experts is an important issue. With consideration of the experts’ knowledge background and historical evaluation performance, this paper proposes an evidential reasoning approach under two-dimensional frames of discernment to aggregate the evaluation opinions of multiple experts. Firstly, two-dimensional frames of discernment are constructed. The I-dimensional frame of discernment describes experts’ evaluation opinions, and the II-dimensional frame of discernment describes experts’ knowledge background and historical evaluation performance. Experts’ evaluation opinions and characteristic information are transformed into pieces of evidence, which could be expressed by their respective belief distributions. Secondly, the II-dimensional evidence information is used to generate the evidence correction factor to reflect experts’ reliabilities. Finally, the experts’ evaluation opinions are corrected using the evidence correction factors. Thus, the corrected experts’ evaluation opinions contain the experts’ characteristic information. In addition, the evidential reasoning approach is used to aggregate the evaluation opinions of multiple experts to complete the selection of the scientific research projects. An illustrative example on aggregating the evaluation opinions of the National Natural Science Foundation of China is provided to demonstrate the applicability and validity of the proposed method. Experimental results show that the proposed method can enhance the description of uncertainty in experts’ evaluation opinions, reflect the qualities of the experts’ evaluation opinions, and make the aggregation results of multiple experts’ evaluation opinions more reasonable and accurate.