A method of multi-objective decision making based on evidence reasoning

Abstract

The present invention provides an evidence reasoning multi-objective decision-making method, comprising the following steps: S1: evidence collection and classification: various types of evidence in the implementation process are collected, and then the evidence is classified according to the operation type, the implementation plan and the impact degree; S2: evidence screening: according to the impact degree screening in step S1, the evidence reasoning multi-objective decision-making method provided by the present invention collects and classifies the evidence, and then removes the evidence with a smaller impact degree according to the impact degree, and then evaluates the credibility of the remaining evidence, so that the evidence with poor credibility can be removed again, thereby reducing the amount of evidence, thereby reducing the decision-making errors caused by the amount, and at the same time, through a preset linear graph, the pros and cons can be divided, and then the linear graph of the pros and cons can be combined, so that the decision can be judged intuitively, avoiding the problem that the best decision cannot be intuitively selected in the existing multi-objective decision-making.

Description

Technical Field

[0001] The present invention relates to the technical field of evidence reasoning, and in particular to a multi-objective decision-making method of evidence reasoning. Background Art

[0002] Multi-objective decision-making is a decision with more than two decision-making objectives, and multiple standards are needed to evaluate and optimize the plan. Most of them are the most important strategic decisions in corporate decision-making. For example, the decision of a major technological transformation project must consider economic benefits, social benefits, safe production and environmental protection and other objectives, and multiple standards are needed to evaluate and optimize the plan.

[0003] Existing multi-objective decision-making is based on reasoning from multiple pieces of evidence. However, when there is a large amount of evidence, it takes a lot of time to judge each piece of evidence. When the amount of evidence is large, the difficulty of decision-making increases, making it impossible to intuitively choose the best decision.

[0004] Therefore, it is necessary to provide an evidential reasoning multi-objective decision-making method to solve the above technical problems. Summary of the invention

[0005] The present invention provides an evidence reasoning multi-objective decision-making method, which solves the problem that the existing method alone cannot intuitively select the best decision.

[0006] In order to solve the above technical problems, the present invention provides an evidential reasoning multi-objective decision-making method, which includes the following steps:

[0007] S1: Evidence collection and classification: collect all kinds of evidence in the implementation process, and then classify the evidence according to the operation type, implementation plan and impact level;

[0008] S2: Evidence screening: Based on the impact screening in step S1, the evidence with an impact level of less than 35% is eliminated, and the impact level is graded from high to low;

[0009] S3: Evidence credibility assessment: Based on the remaining evidence in step S2, the credibility of the evidence is assessed based on the implementation plan of the evidence and the information provided by the person providing the evidence, and then some of the evidence is screened twice;

[0010] S4: Linear view analysis: Based on the results of the secondary screening in step S3, the multiple pieces of evidence after screening are divided into linear graphs according to the preset results, and then the multiple linear graphs are combined and displayed, and then the evidence is screened again according to the combined results;

[0011] S5: Decision determination: According to step S4, analyze the disadvantages of the linear graph of each piece of evidence, and then analyze the advantages and disadvantages in combination with the advantages in step S4, and then compare the analyzed data, and then judge the final decision.

[0012] Preferably, the classification types of evidence in step S1 may also be: similarity of the target action plan, feasibility of the target action plan, credibility of the target action plan, pros and cons of the target action plan, and implementation time of the action plan.

[0013] Preferably, in the step S2, the level of impact can be divided into top, high, medium and low levels, with the top level of impact accounting for 100%-85%, the high level of impact accounting for 84%-75%, the medium level of impact accounting for 74%-60%, and the low level of impact accounting for 59%-36%.

[0014] Preferably, the evaluation of the information of the person providing evidence in step S3 includes: an evaluation of the professional level of the person and an evaluation of the credibility of the person. The evaluation of the professional level of the person is an evaluation of the technology mastered by the person providing the information in the field in which the information is provided. If the person is not a person in the technical field, the credibility is divided into 40%-60%, and if the person is a person in the technical field, the credibility is divided into 60%-80%.

[0015] Preferably, the assessment of the personnel credibility is to judge the credibility of the information by whether the personnel who improve the information often improve the accuracy of the information, and can also formulate addition and subtraction items of the credibility according to each improvement of the accuracy of the information. When the accuracy is high, the credibility is increased, and when the accuracy is low, the credibility is reduced.

[0016] Preferably, in step S4, the linear graph is drawn by dividing the data according to the values ​​of the same type, and after the division, the divided linear graphs are compared.

[0017] Preferably, in the step S5, the division of advantages and disadvantages is based on zero as the boundary, the data below zero is the disadvantages, and the data above zero is the advantage. After the division is completed, the linear graph below zero and close to zero can be rotated while having a linear graph above zero and away from zero.

[0018] Compared with related technologies, the evidence reasoning multi-objective decision-making method provided by the present invention has the following beneficial effects:

[0019] The present invention provides an evidence reasoning multi-objective decision-making method, which collects and classifies evidence, removes evidence with a smaller degree of influence according to the degree of influence, and then evaluates the credibility of the remaining evidence, so that evidence with a lower degree of credibility can be removed again, thereby reducing the amount of evidence, thereby reducing decision-making errors caused by the amount, and at the same time, through a preset linear graph, the pros and cons can be divided, and then the linear graph of the pros and cons can be combined, so that the decision can be judged intuitively, avoiding the problem of being unable to intuitively select the best decision in existing multi-objective decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of a preferred embodiment of the evidential reasoning multi-objective decision-making method provided by the present invention. DETAILED DESCRIPTION

[0021] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0022] Please refer to Figure 1 ,in, Figure 1 A flowchart of a preferred embodiment of the evidence reasoning multi-objective decision-making method provided by the present invention. An evidence reasoning multi-objective decision-making method comprises the following steps:

[0023] S1: Evidence collection and classification: collect all kinds of evidence in the implementation process, and then classify the evidence according to the operation type, implementation plan and impact level;

[0024] S2: Evidence screening: Based on the impact screening in step S1, the evidence with an impact level of less than 35% is eliminated, and the impact level is graded from high to low;

[0025] S3: Evidence credibility assessment: Based on the remaining evidence in step S2, the credibility of the evidence is assessed based on the implementation plan of the evidence and the information provided by the person providing the evidence, and then some of the evidence is screened twice;

[0026] S4: Linear view analysis: Based on the results of the secondary screening in step S3, the multiple pieces of evidence after screening are divided into linear graphs according to the preset results, and then the multiple linear graphs are combined and displayed, and then the evidence is screened again according to the combined results;

[0027] S5: Decision determination: According to step S4, analyze the disadvantages of the linear graph of each piece of evidence, and then analyze the advantages and disadvantages in combination with the advantages in step S4, and then compare the analyzed data, and then judge the final decision.

[0028] The classification types of evidence in step S1 may also be: similarity of the target action plan, feasibility of the target action plan, credibility of the target action plan, pros and cons of the target action plan, and implementation time of the action plan.

[0029] The type of classification can also be adjusted specifically according to needs.

[0030] In the step S2, the impact level can be divided into top, high, medium and low levels, with the top impact level accounting for 100%-85%, the high impact level accounting for 84%-75%, the medium impact level accounting for 74%-60%, and the low impact level accounting for 59%-36%.

[0031] While taking the impact level as a reference, other target action plans also need to be referenced.

[0032] The evaluation of the information of the person providing evidence in step S3 includes: an evaluation of the professional level of the person and an evaluation of the credibility of the person. The evaluation of the professional level of the person is an evaluation of the technology mastered by the person providing the information in the field in which the information is provided. If the person is not a person in the technical field, the credibility is divided into 40%-60%, and if the person is a person in the technical field, the credibility is divided into 60%-80%.

[0033] Surveys or inspections can be conducted to determine whether personnel have mastered the technology.

[0034] The evaluation of the credibility of the personnel is to judge the credibility of the information by constantly improving the accuracy of the information provided by the personnel who improve the information, and can also formulate items of credibility addition and subtraction according to each improvement in the accuracy of the information. When the accuracy is high, the credibility is increased, and when the accuracy is low, the credibility is reduced.

[0035] If a person makes continuous correct or incorrect statements, the speed of adding or subtracting credit can be adaptively increased or decreased.

[0036] In the step S4, the linear graph is drawn by dividing the data according to the values ​​of the same type, and after the division, the divided linear graphs are compared.

[0037] In the step S5, the division of advantages and disadvantages is based on zero as the boundary, the data below zero is the disadvantages, and the data above zero is the advantages. After the division is completed, the linear graph below zero and close to zero can be rotated while having a linear graph above zero and away from zero.

[0038] Compared with related technologies, the evidence reasoning multi-objective decision-making method provided by the present invention has the following beneficial effects:

[0039] By aggregating and classifying the evidence, and then removing the evidence with lesser impact according to the degree of influence, and then evaluating the credibility of the remaining evidence, the evidence with less credibility can be removed again, thereby reducing the amount of evidence, thereby reducing decision-making errors caused by quantity. At the same time, the pros and cons can be divided through a preset linear graph, and then the linear graph of pros and cons can be combined to make intuitive judgments on decisions, avoiding the problem of not being able to intuitively select the best decision in existing multi-objective decision-making.

[0040] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for multi-objective decision-making based on evidence reasoning, It is characterized in that The following steps are involved: S1: Evidence collection and classification: collect all kinds of evidence in the implementation process, and then classify the evidence according to the operation type, implementation plan and impact level; S2: Evidence screening: Based on the impact screening in step S1, the evidence with an impact level of less than 35% is eliminated, and the impact level is graded from high to low; S3: Evidence credibility assessment: Based on the remaining evidence in step S2, the credibility of the evidence is assessed based on the implementation plan of the evidence and the information provided by the person providing the evidence, and then some of the evidence is screened twice; S4: Linear view analysis: Based on the results of the secondary screening in step S3, the multiple pieces of evidence after screening are divided into linear graphs according to the preset results, and then the multiple linear graphs are combined and displayed, and then the evidence is screened again according to the combined results; S5: Decision determination: According to step S4, analyze the disadvantages of the linear graph of each piece of evidence, and then analyze the advantages and disadvantages in combination with the advantages in step S4, and then compare the analyzed data, and then judge the final decision; In the step S4, the linear graph is drawn by dividing the data according to the values ​​of the same type, and after the division, the divided linear graphs are compared; In the step S5, the division of advantages and disadvantages is based on zero as the boundary, the data below zero is the disadvantages, and the data above zero is the advantages. After the division is completed, the linear graph below zero and close to zero can be rotated while having a linear graph above zero and away from zero.

2. According to the evidence reasoning multi-objective decision-making method of claim 1, It is characterized in that The classification types of evidence in step S1 may also be: similarity of the target action plan, feasibility of the target action plan, credibility of the target action plan, pros and cons of the target action plan, and implementation time of the action plan.

3. According to the evidence reasoning multi-objective decision-making method of claim 1, It is characterized in that In the step S2, the impact level can be divided into top, high, medium and low levels, with the top impact level accounting for 100%-85%, the high impact level accounting for 84%-75%, the medium impact level accounting for 74%-60%, and the low impact level accounting for 59%-36%.

4. According to the multi-objective decision-making method of evidence reasoning according to claim 1, It is characterized in that The evaluation of the information of the person providing evidence in step S3 includes: an evaluation of the professional level of the person and an evaluation of the credibility of the person. The professional level of the person is an evaluation of the technology mastered by the person providing the information in the field in which the information is provided. If the person is not a person in the technical field, the credibility is divided into 40%-60%, and if the person is a person in the technical field, the credibility is divided into 60%-80%.

5. According to the evidence reasoning multi-objective decision-making method of claim 4, It is characterized in that The evaluation of the credibility of the personnel is to judge the credibility of the information by constantly improving the accuracy of the information provided by the personnel who improve the information, and can also formulate items of credibility addition and subtraction according to each improvement in the accuracy of the information. When the accuracy is high, the credibility is increased, and when the accuracy is low, the credibility is reduced.

Images