Policy simulation system and method thereof with integrating multi-criteria decision making and system dynamics

Abstract

A policy simulation method includes: collecting relevant data; using the relevant data to initially simulate and establish at least one evaluation indicator; using the evaluation indicator to provide several preliminary expert questionnaires to obtain at least one preliminary expert questionnaire result; using the preliminary expert questionnaire result to establish a hierarchical architecture of a network procedure analysis method; establishing several ANP expert questionnaires to obtain at least one ANP expert questionnaire result; using the ANP expert questionnaire result to perform weight calculation to obtain at least one model weight value; using the ANP expert questionnaire result to establish an indicator of at least one system dynamics model; and using the ANP expert questionnaire result and the indicator of the system dynamics model to draw at least one causal feedback graph.

Description

[Technical Field]

[0001] This invention relates to a policy simulation system and method that integrates multi-criteria decision-making and system dynamics, employing an analytical network process (ANP) based on decision making trial and evaluation laboratory (DEMATEL) analysis; particularly, it relates to a policy simulation system and method that integrates multi-criteria decision-making and system dynamics, achieving overall simplification of the processing flow and reducing the amount of data processing. [Previous Technology]

[0002] Conventional machining strategy analysis systems employing DANP technology include, for example, the invention patent "Machining Strategy Analysis System for Vertical Cutting Center" published by the Republic of China Patent No. TW-I769798, which discloses a machining strategy analysis system for a vertical cutting center. This machining strategy analysis system for a vertical cutting center includes an acquisition module, a memory module, an information module, and a calculation module, and the acquisition module, memory module, and calculation module are coupled to the information module.

[0003] As above, the aforementioned computing module of No. TW-I769798 has a first computing program, a second computing program and a third computing program. The first computing program adopts the Fuzzy Delphi Method, the second computing program adopts the Decision-Making Laboratory Method, and the third computing program is the Network Hierarchical Analysis Method (DANP) based on the Decision-Making Laboratory Method.

[0004] Another conventionally used decision-making system for the development of medical wearable devices employing DANP technology is the new utility model patent No. TW-M605334 of the Republic of China, entitled "Decision-Making System for the Development of Medical Wearable Devices," which discloses a decision-making system for the development of medical wearable devices. This decision-making system includes a data processing device, which has an input module, a database, and a processing module.

[0005] Continuing from the above, the processing module of the aforementioned No. TW-M605334 has a key factor calculation program. This key factor calculation program processes an influence relationship matrix into a super matrix, and sums the row vectors of the super matrix to obtain a key factor ranking of the weights of each indicator. In addition, the key factor calculation program uses the Network Hierarchical Analysis (DANP) method based on decision laboratory analysis to remove the total influence value of each criterion from the influence relationship matrix to transpose it into a normalized influence relationship matrix, and calculates the super matrix from the normalized influence relationship matrix.

[0006] Another commonly used method for evaluating the level of agricultural modernization development using DANP technology is the invention patent application No. CN-113610444 entitled "An Evaluation Method for the Level of Agricultural Modernization Development Based on the Correlation of Indicators". It discloses that an evaluation method for the level of agricultural modernization development based on the correlation of indicators includes: Step 1: Based on an agricultural modernization development goal and current status, select an evaluation indicator for the level of agricultural modernization development and establish an evaluation indicator system for the level of agricultural modernization development.

[0007] As above, the evaluation method for the level of agricultural modernization development based on the correlation degree of indicators in the aforementioned CN-113610444 includes: Step 2, based on the correlation degree between the evaluation indicators, a grey relational analysis is used to calculate the grey relational degree between each indicator, and a direct influence matrix is ​​constructed. Furthermore, a fuzzy DANP method is introduced to construct a fuzzy GRA-DANP method in order to determine the weight of each evaluation indicator.

[0008] Continuing from the above, the evaluation method for the level of agricultural modernization development based on the correlation of indicators in the aforementioned CN-113610444 includes: Step 3, using the TOPSIS method to comprehensively evaluate the level of agricultural modernization development. Due to the mutual influence relationship between the evaluation indicators, the fuzzy GRA-DANP method is constructed to calculate the weights to ensure the reliability of the weights of each indicator, and the TOPSIS model is used to evaluate the level of agricultural modernization development so that the evaluation result is more comprehensive and accurate.

[0009] Another commonly used method and system for comprehensively evaluating the performance of demand response execution in load management using DANP technology is Chinese Patent Publication No. CN-115759711, entitled "Comprehensive Evaluation Method and System for Demand Response Execution Performance in Load Management," which discloses a comprehensive evaluation method for demand response execution performance in load management. This comprehensive evaluation method for demand response execution performance in load management includes: Step 1, constructing an evaluation index system suitable for the performance of demand response execution in load management.

[0010] As above, the aforementioned comprehensive evaluation method for the performance of demand response to load management in CN-115759711 includes: Step 2, obtaining a subjective evaluation weight vector based on the DANP method for the evaluation index system constructed in Step 1; Step 3, obtaining an objective evaluation weight vector based on the anti-entropy weight method for the evaluation index system constructed in Step 1.

[0011] As above, the aforementioned comprehensive evaluation method for the performance of demand response to load management in CN-115759711 includes: Step 4, using the subjective evaluation weight vector and the objective evaluation weight vector, a comprehensive evaluation weight is obtained by using the combined weighting method; Step 5, using an evaluation index and a comprehensive evaluation weight, a final evaluation level and a score are obtained by using a gray cloud model.

[0012] Another conventionally used data transmission method, apparatus, device, and dual-connection PRP node employing DANP technology is disclosed in Chinese Patent Publication No. CN-115914138, entitled "Data Transmission Method, Apparatus, Device, and Dual-Connection PRP Node," which discloses a dual-connection Parallel Redundancy Protocol (PRP) node. This dual-connection parallel redundancy protocol node includes a decision logic submodule, a two-in-one submodule, and a DANP submodule.

[0013] As above, the judgment logic sub-module of the aforementioned CN-115914138 is provided with a sending unit and a receiving unit. The sending unit is connected to the receiving unit and the two-in-one sub-module respectively, and determines, according to a destination IP address, whether to send a data to be sent via a communication link formed by the receiving unit or via a communication link formed by the two-in-one sub-module.

[0014] As above, the receiving unit of the aforementioned CN-115914138 is connected to the transmitting unit and the DANP sub-module respectively, and determines, based on a source IP address, whether to receive data to be received via a communication link formed by the transmitting unit or via a communication link formed by the DANP sub-module.

[0015] Continuing from the above, the aforementioned dual-mode sub-module of CN-115914138 is connected to the DANP sub-module and merges the data to be transmitted into one data stream, or copies the data received by the DANP sub-module into two data streams. The DANP sub-module is used to copy the data stream and upload it to an optical network and a power grid respectively, or to merge the data from the optical network and the power grid to obtain a received data stream.

[0016] However, the aforementioned Republic of China Patent Publications No. TW-I769798, No. TW-M605334, Chinese Patent Publications No. CN-113610444, No. CN-115759711 and No. CN-115914138 only disclose various simple DANP technologies. Therefore, they do not provide any improvements to the DANP technology. Thus, there must be a need for further improvements to meet the requirements for enhancing the technology.

[0017] Obviously, the aforementioned Republic of China Patent Publications No. TW-I769798, No. TW-M605334, Chinese Patent Publications No. CN-113610444, No. CN-115759711 and No. CN-115914138 are only for reference and explanation of the technical background of this invention and the current state of technological development, and are not intended to limit the scope of this invention.

[0018] In view of the above, in order to meet the above-mentioned technical problems and needs, the present invention provides a policy simulation system and method that integrates multi-criteria decision-making and system dynamics. It collects relevant literature or data, and uses the relevant literature or data to initially simulate and establish at least one evaluation index. It also uses the evaluation index to provide several preliminary expert questionnaires to obtain at least one preliminary expert questionnaire result. Using the preliminary expert questionnaire result, it establishes a hierarchical architecture of network procedure analysis (DPA) and establishes several DPA expert questionnaires to obtain at least one DPA expert questionnaire result. Then, it uses the DPA expert questionnaire result to perform weight calculation to obtain at least one model weight value. It also uses the expert questionnaire result to establish at least one system dynamics model index, and uses the DPA expert questionnaire result and the system dynamics model index to draw at least one causal feedback graph. This improves upon the technical problems of conventional methods that simply employ DANP technology without any improvements to DANP technology. [Summary of the Invention]

[0019] The main objective of this invention is to provide a policy simulation system and method that integrates multi-criteria decision-making and system dynamics. This system collects relevant literature or data, and uses this literature or data to initially simulate and establish at least one evaluation index. It also uses this evaluation index to provide several preliminary expert questionnaires to obtain at least one preliminary expert questionnaire result. Using the preliminary expert questionnaire result, it establishes a hierarchical architecture for network procedure analysis (LPA) and establishes several LPA expert questionnaires to obtain at least one LPA expert questionnaire result. Then, it uses the LPA expert questionnaire result to perform weight calculations to obtain at least one model weight value. Furthermore, it uses the expert questionnaire result to establish at least one system dynamics model index, and uses the LPA expert questionnaire result and the system dynamics model index to draw at least one causal feedback graph, thereby achieving the goal and effect of simplifying the overall processing flow and reducing the amount of data processed.

[0020] To achieve the above objectives, the preferred embodiment of the policy simulation system integrating multi-criteria decision-making and system dynamics of the present invention includes:

[0020] At least one data collection unit is used to collect a relevant document or relevant data;

[0020] At least one simulated evaluation indicator unit, which uses the relevant data to initially simulate and establish at least one evaluation indicator;

[0020] At least one preliminary expert questionnaire unit, which uses the evaluation index to provide several preliminary expert questionnaires in order to obtain at least one preliminary expert questionnaire result;

[0020] At least one hierarchical architecture for network program analysis, which uses the preliminary expert questionnaire results to establish the hierarchical architecture for network program analysis;

[0020] Several ANP expert questionnaire units, used to establish several ANP expert questionnaires in order to obtain expert questionnaire results for at least one network procedure analysis method, and to use the expert questionnaire results of the network procedure analysis method to calculate weights to obtain at least one model weight value; and

[0020] A computing unit having a causal feedback module, and using the expert questionnaire results to establish an index of at least one system dynamics model, and using the expert questionnaire results of the network program analysis method and the index of the system dynamics model to draw at least one causal feedback diagram.

[0021] In a preferred embodiment of the present invention, the influence relationship of several criteria is adopted, and the influence relationship of several criteria is defined using several ANP expert questionnaires.

[0022] In a preferred embodiment of the present invention, several mutual influence relationship values ​​are used, and the several mutual influence relationship values ​​are obtained from the influence relationship of several criteria.

[0023] In a preferred embodiment of the present invention, the calculation unit further includes at least one direct influence matrix, and the direct influence matrix is ​​established using several of the mutual influence relationship values.

[0024] In a preferred embodiment of the present invention, the calculation unit further includes a normalization module, and uses the direct influence matrix to establish at least one normalized influence matrix, and uses the normalized influence matrix to perform calculations to obtain at least one total influence matrix, and performs normalization processing on the total influence matrix to obtain a normalized total influence matrix, and uses the normalized total influence matrix to perform limit calculations to obtain at least one limited super matrix, so as to obtain the weight values ​​of several criteria.

[0025] To achieve the above objectives, the preferred embodiment of the policy simulation method integrating multi-criteria decision-making and system dynamics of the present invention includes:

[0025] Collect a relevant document or relevant information;

[0025] At least one evaluation indicator is preliminarily established using relevant literature or materials;

[0025] Several preliminary expert questionnaires are provided using this evaluation indicator in order to obtain at least one preliminary expert questionnaire result;

[0025] A hierarchical architecture for network program analysis was established using the preliminary expert questionnaire results;

[0025] Establish several expert questionnaires for network program analysis in order to obtain expert questionnaire results for at least one network program analysis method;

[0025] The weights are calculated using the expert questionnaire results of this network program analysis method to obtain at least one model weight value;

[0025] Using the expert questionnaire results of this network program analysis method, at least one index for the system dynamics model is established; and

[0025] At least one causal feedback diagram is drawn using the expert questionnaire results and the indicators of the system dynamics model of the network program analysis method.

[0026] In a preferred embodiment of the present invention, several expert questionnaires of the network program analysis method are used to define the influence relationship of several criteria in order to obtain several mutual influence relationship values.

[0027] In a preferred embodiment of the present invention, at least one direct influence matrix is ​​established using several of the mutual influence relationship values, and at least one normalized influence matrix is ​​established using the direct influence matrix.

[0028] In a preferred embodiment of the present invention, the normalized influence matrix is ​​used to perform calculations to obtain at least one total influence matrix, and the total influence matrix is ​​normalized to obtain a normalized total influence matrix.

[0029] In a preferred embodiment of the present invention, the normalized total influence matrix is ​​used to perform limit calculations to obtain at least one limited super matrix in order to obtain the weight values ​​of several criteria. [Simplified Explanation of the Diagram]

[0030] Figure 1: Schematic diagram of a policy simulation system integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention.

[0030] Figure 1A: Schematic diagram of the architecture of the preferred embodiment of the present invention, which integrates multi-criteria decision-making and system dynamics for policy simulation and applies it to the reduction of marine debris or waste.

[0030] Figure 2: Flowchart of the policy simulation method integrating multi-criteria decision-making and system dynamics according to the first preferred embodiment of the present invention.

[0030] Figure 2A: A schematic diagram of the architecture of the preferred embodiment of the present invention, which integrates multi-criteria decision-making and system dynamics for policy simulation and obtains a corrected reduction in marine debris or waste.

[0030] Figure 3: A schematic diagram of the causal feedback graph obtained by the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention.

[0030] Figure 4: A schematic diagram of the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention, which uses the network program analysis method (DANP) based on decision laboratory analysis to perform expert integration calculations.

[0030] Figure 4A: Schematic diagram of the architecture of the policy simulation system and method integrating multi-criteria decision-making and system dynamics of the present invention using expert questionnaire processing.

[0030] Figure 5: A flowchart illustrating the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to the second preferred embodiment of the present invention, which employs the network procedure analysis method (DANP) based on decision laboratory analysis for expert integration calculation.

[0030] Figure 6: A flowchart illustrating the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to the third preferred embodiment of the present invention, which employs the network procedure analysis method (DANP) based on decision laboratory analysis for expert integration calculation.

Implementation Method

[0031] In order to fully understand the present invention, preferred embodiments will be described in detail below with reference to the accompanying drawings, and these are not intended to limit the present invention.

[0032] The preferred embodiment of the present invention, which integrates multi-criteria decision-making and system dynamics for policy simulation, its method and operation method are applicable to various human questionnaires and related application technology fields, such as various public opinion surveys and their analysis, various market surveys and their analysis, various industry surveys and their analysis, various environmental surveys and their analysis, various ecological surveys and their analysis, various policy surveys and their analysis, various cultural studies or surveys and their analysis, or other surveys and their analysis, but it is not intended to limit the scope of the present invention.

[0033] Generally speaking, traditional policy simulation systems and methods that integrate multi-criteria decision-making and system dynamics may fail to fully express or reflect the true mental patterns of questionnaire respondents due to biases caused by human questionnaire completion. Therefore, the preferred embodiment of the present invention, which uses the Network Process Analysis (DANP) method based on decision laboratory analysis, can improve the calculation process and solve the problem of biases caused by human questionnaire completion.

[0034] Figure 1 illustrates a schematic diagram of a policy simulation system integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention. Referring to Figure 1, for example, the policy simulation system integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention includes an input unit 1, a data collection unit 1a, a preliminary expert questionnaire unit 10a, an ANP expert questionnaire unit 10b, a calculation unit 2, a simulation evaluation index unit 2a, an ANP hierarchical structure 2b, a system dynamic model index 2c, and an output unit 3.

[0035] Please refer to Figure 1 again. For example, the input unit 1 can be selected from a computer data input unit or a unit with similar data input function. The input unit 1 can be connected to the data collection unit 1a by appropriate technology (e.g., wired or wireless communication). The data collection unit 1a can be selected from a computer data storage unit or a unit with similar data storage function.

[0036] Referring again to Figure 1, for example, in another preferred embodiment of the present invention, the computing unit 2 may be connected to various database units by appropriate technical means. The computing unit 2 may be selected from a workstation computer, a desktop computer, a notebook computer, a tablet personal computer, a mobile communication device, a smartphone, or other devices with computer functions, but this is not intended to limit the scope of the present invention.

[0037] Please refer to Figure 1 again. For example, the output unit 3 can be selected from a computer data output unit or a unit with similar data output function, and the input unit 1, data collection unit 1a and output unit 3 can be connected to the computing unit 2 by appropriate technology (e.g., wired or wireless communication).

[0038] Please refer to Figure 1 again. For example, the output unit 3 is connected to the computing unit 2 by appropriate technical means, and the output unit is further connected to a terminal device, a display device, a projection device, a storage device, a server device, an information management center or any combination thereof by appropriate technical means.

[0039] Please refer to Figure 1. For example, the computing unit 2 has the simulation evaluation index unit 2a, the hierarchical architecture of ANP 2b and the index 2c of the system dynamics mode, and the simulation evaluation index unit 2a, the hierarchical architecture of ANP 2b and the index 2c of the system dynamics mode can be appropriately configured in the computing unit 2.

[0040] Figure 1A illustrates the architecture of a preferred embodiment of the present invention, which integrates multi-criteria decision-making and system dynamics for policy simulation and applies the system and method thereof to the reduction of marine debris or waste. Referring to Figure 1A, for example, the architecture for reducing marine debris or waste includes a first layer (top layer), a second layer (middle layer), and a third layer (bottom layer).

[0041] Please refer to Figure 1A again. For example, the first, second and third layers are arranged from top to bottom. The first layer can be selected as a target layer, the second layer can be selected as a surface layer, and the third layer can be selected as a criterion layer (e.g., an index of system dynamics). Three experts from academia, government, and NGOs can be selected.

[0042] Please refer again to Figure 1. For example, the execution mode of the policy simulation system and method integrating multi-criteria decision-making and system dynamics of the preferred embodiment of the present invention is a computer-executable process step, which can be executed on various computer devices, such as workstation computers, desktop computers, notebook computers, tablet computers, mobile communication devices, smartphones or other devices with computer functions, but it is not intended to limit the scope of the present invention.

[0043] Figure 2 illustrates a schematic diagram of the policy simulation method integrating multi-criteria decision-making and system dynamics according to a first preferred embodiment of the present invention. Referring to Figures 1 and 2, the policy simulation method integrating multi-criteria decision-making and system dynamics according to a first preferred embodiment of the present invention includes step S1: First, for example, relevant literature or relevant data is collected in the data collection unit 1a or other units with similar data collection functions using appropriate technical means (e.g., automated, semi-automated, or manual methods).

[0044] Please refer again to Figures 1 and 2. The policy simulation method integrating multi-criteria decision-making and system dynamics in the first preferred embodiment of the present invention includes step S2: Next, for example, by means of appropriate technical means (e.g., automated, semi-automated or manual), the relevant literature or relevant data may be used to conduct a preliminary simulation to establish at least one evaluation index, or it may be applied to the computing unit 2 or other units with similar computing functions.

[0045] Referring again to Figures 1 and 2, the policy simulation method integrating multi-criteria decision-making and system dynamics in the first preferred embodiment of the present invention includes step S3: Next, for example, using appropriate technical means (e.g., automated, semi-automated or manual), several preliminary expert questionnaires (e.g., Delphi method expert questionnaires) are provided through the preliminary expert questionnaire unit 10a using the evaluation index, so as to obtain at least one preliminary expert questionnaire result, or it can be selected to be applied to the calculation unit 2 or other units with similar calculation functions.

[0046] Referring again to Figures 1 and 2, the policy simulation method integrating multi-criteria decision-making and system dynamics in the first preferred embodiment of the present invention includes step S4: Next, for example, using appropriate technical means (e.g., automated, semi-automated or manual), a hierarchical architecture of network program analysis (ANP) is established using the preliminary expert questionnaire results, or it may be applied to the computing unit 2 or other units with similar computing functions.

[0047] Figure 2A illustrates a schematic diagram of the architecture of a policy simulation system and method integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention, which obtains a corrected reduction in marine debris or waste. Referring again to Figures 1, 2, and 2A, the policy simulation method integrating multi-criteria decision-making and system dynamics according to a first preferred embodiment of the present invention includes step S5: Next, for example, several expert questionnaires for network procedure analysis are established through the ANP expert questionnaire unit 10b using appropriate technical means (e.g., automated, semi-automated, or manual methods) to obtain at least one expert questionnaire result for network procedure analysis, or it can be selectively applied to the computing unit 2 or other units with similar computing functions.

[0048] Referring again to Figures 1 and 2, the policy simulation method integrating multi-criteria decision-making and system dynamics in the first preferred embodiment of the present invention includes step S6: Next, for example, by appropriate technical means (e.g., automated, semi-automated or manual), the weight calculation can be performed on the computing unit 2 or other units with similar computing functions using the expert questionnaire results of the network program analysis method to obtain at least one model weight value (as shown in Table 2), or it can be applied to the computing unit 2 or other units with similar computing functions.

[0048]

[0049] Referring again to Figures 1 and 2, the policy simulation method integrating multi-criteria decision-making and system dynamics in the first preferred embodiment of the present invention includes step S7: Next, for example, by appropriate technical means (e.g., automated, semi-automated or manual), at least one indicator of the system dynamics model (such as the third layer shown in Figure 2A, i.e., the criterion layer as the indicator of system dynamics) can be established in the computing unit 2 or other units with similar computing functions using the expert questionnaire results of the network program analysis method, or it can be applied to the computing unit 2 or other units with similar computing functions.

[0050] Figure 3 illustrates a schematic diagram of the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention, showing the causal feedback graph obtained. Referring again to Figures 1, 2, and 3, the policy simulation method integrating multi-criteria decision-making and system dynamics according to a first preferred embodiment of the present invention includes step S8: Next, for example, using appropriate technical means (e.g., automated, semi-automated, or manual methods), at least one causal feedback graph 30 can be drawn in the computing unit 2 or other units with similar computing functions using the expert questionnaire results of the network program analysis method and the indicators of the system dynamics model, as shown in Figures 1 and 3.

[0051] Figure 4 illustrates a schematic diagram of the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to a preferred embodiment of the present invention, which employs the Decision Laboratory Analysis-Based Network Program Analysis (DANP) method for expert integration calculation. Referring to Figure 4, for example, the Decision Laboratory Analysis-Based Network Program Analysis (DANP) method for expert integration calculation according to a preferred embodiment of the present invention includes an input unit 1, an expert questionnaire unit 10, a calculation unit 2, and an output unit 3.

[0052] Please refer to Figure 4. For example, the computing unit 2 has at least one normalization module 20. Generally speaking, there are two main normalization methods. The first method uses the maximum value of the sum of each column vector as the normalization benchmark, and the second method uses the maximum value of the sum of each row vector and column vector as the normalization benchmark.

[0053] Figure 4A shows a schematic diagram of the architecture of the policy simulation system and method integrating multi-criteria decision-making and system dynamics of the present invention, which adopts expert questionnaire processing. It corresponds to the expert integration calculation system based on the network program analysis method (DANP) of the decision laboratory analysis in Figure 4.

[0054] Referring to Figures 4 and 4A, for example, the expert questionnaire unit 10 contains several expert questionnaires 11, and these expert questionnaires 11 are prepared and processed by several experts (e.g., completion or other tasks). Generally, the system has various criteria, and these criteria are appropriately compared by executing the expert questionnaires 11.

[0055] Figure 5 illustrates a schematic diagram of the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to the second preferred embodiment of the present invention, which employs the Network Process Analysis Method (DANP) based on decision laboratory analysis for expert integration calculation. Referring to Figures 4, 4A and 5, the expert integration calculation method of the Network Process Analysis Method (DANP) based on decision laboratory analysis according to the second preferred embodiment of the present invention includes step S10: First, for example, using appropriate technical means (e.g., automated, semi-automated or manual methods), several expert questionnaires 11 are used to define the influence relationships of several criteria 21 in order to obtain several mutually influent relation values ​​22.

[0056] Please refer again to Figures 4, 4A and 5. For example, in a preferred embodiment of the present invention, the influence relationship 21 of several criteria is compared pairwise using several expert questionnaires 11. The mutual influence relationship value 22 includes no influence, low influence, moderate influence, high influence and extremely high influence. The mutual influence relationship values ​​of no influence, low influence, moderate influence, high influence and extremely high influence have a predetermined scale, and the predetermined scale includes influence degree value 0, influence degree value 1, influence degree value 2, influence degree value 3 and influence degree value 4.

[0057] Please refer again to Figures 4, 4A and 5. The expert integration calculation method based on Decision Laboratory Analysis (DANP) of the second preferred embodiment of the present invention includes step S20: Next, for example, at least one initial-influence matrix X is established using several of the mutual influence relationship values ​​22 by appropriate technical means (e.g., automated, semi-automated or manual).

[0058] In a preferred embodiment of the present invention, after defining the magnitude of the influence, the direct influence matrix X can be established. If there are n evaluation criteria, the criteria are compared pairwise according to their influence, presenting an n*n direct influence matrix X=[Xij], where Xij is the degree to which criterion i influences criterion j, and the diagonal of the matrix is ​​the degree of influence of each criterion on itself. Since it is considered to have no influence, it is set to 0.

[0059] The direct influence matrix X of the preferred embodiment of the present invention is as follows:

[0059]

[0059] where Xij represents the degree to which criterion i influences criterion j.

[0060] Please refer again to Figures 4, 4A and 5. The expert integration calculation method based on the Decision Laboratory Analysis Network Program Analysis (DANP) of the second preferred embodiment of the present invention includes step S30: Next, for example, at least one normalized direct-influence matrix D is established using the direct-influence matrix X by appropriate technical means (e.g., automated, semi-automated or manual methods).

[0061] The normalized influence matrix D of the preferred embodiment of the present invention is as follows:

[0061] D=kX

[0061]

[0061] The normalized influence matrix is ​​D, and the normalized baseline value is k.

[0062] Referring again to Figures 4, 4A and 5, the expert integration calculation method based on Decision Laboratory Analysis (DANP) of the second preferred embodiment of the present invention includes step S40: Next, for example, the normalized influence matrix D is calculated using appropriate technical means (e.g., automated, semi-automated or manual) to obtain at least one total-influence matrix T, i.e., the total influence relationship matrix.

[0063] Please refer again to Figures 4, 4A and 5. For example, in the second preferred embodiment of the present invention, the total influence matrix T (i.e., the total influence relationship matrix) can be formed by appropriately combining the direct influence matrix X (i.e., the direct influence relationship matrix) with an indirect influence matrix ID (i.e., the indirect influence relationship matrix).

[0064] In a preferred embodiment of the present invention, after obtaining the normalized influence matrix D (i.e., the normalized influence relation matrix), the total influence matrix T (i.e., the total influence relation matrix) is equivalent to the direct influence matrix D (i.e., the direct influence relation matrix) plus the indirect influence matrix ID (i.e., the indirect influence relation matrix).

[0065] Continuing above, in a preferred embodiment of the present invention, computationally, a sub-stochastic matrix can be obtained by similarly deleting rows or columns that result in an absorbing state from an absorbing Markov chain, thereby obtaining the following:

[0065] lim s→∞ D s=0 lim s→∞(I+D+D2+…+Ds)=(ID)-1

[0065] where 0 is the zero matrix and I is the identity matrix.

[0066] The total influence matrix T of the preferred embodiment of the present invention is as follows:

[0066] T=lim s→∞(D+D2+D3+…+Ds)=D(ID)-1

[0067] Referring again to Figures 4, 4A and 5, the expert integration calculation method based on Decision Laboratory Analysis (DANP) of the second preferred embodiment of the present invention includes step S50: Next, for example, the total influence matrix T is normalized by appropriate technical means (e.g., automated, semi-automated or manual) to obtain a normalized total-influence matrix TC.

[0068] In the second preferred embodiment of the present invention, the total influence matrix of the criteria calculated in the previous step is listed in matrix TC, and its formula is:

[0068]

[0068] where tij is the influence value of criterion i on criterion j, and n is the total number of criteria.

[0069] In the second preferred embodiment of the present invention, the normalization benchmark of the total influence matrix T is calculated, and fi is the normalization benchmark. The formula is as follows:

[0069]

[0070] In the second preferred embodiment of the present invention, the "influence value" criterion is used to normalize the "influenced value" of the total influence matrix T. The formula for normalizing the total influence matrix TC* is as follows:

[0070]

[0071] Please refer again to Figures 4, 4A and 5. In the second preferred embodiment of the present invention, each of the criteria and its definition is determined using the Delphi method. When defining the influence relationship of several criteria in several expert questionnaires, a surface influence is omitted. When normalizing the total influence matrix, a uniform normalization operation is used.

[0072] Referring again to Figures 4, 4A and 5, the expert integration calculation method of the second preferred embodiment of the present invention based on the Decision Laboratory Analysis-based Network Program Analysis (DANP) includes step S60: Next, for example, using appropriate technical means (e.g., automated, semi-automated or manual), the normalized total influence matrix TC is used to perform limit calculation to obtain at least one limited super matrix L in order to obtain the weight of criteria 23.

[0073] In the second preferred embodiment of the present invention, since the influence of the criteria can be calculated on the surface, it is not necessary to perform weighted calculations on the super matrix of the criteria. Since the sum of the row vectors of the weighted super matrix of the criteria is 1, the value of each column of the matrix tends to be stable when performing limit calculations, and this characteristic is used to calculate the weight value of each criterion.

[0074] In the second preferred embodiment of the present invention, the extreme supermatrix L is obtained by continuously multiplying a weighted supermatrix by itself until the weighted supermatrix reaches a stable state.

[0075] The formula for the extreme supermatrix L in the second preferred embodiment of the present invention is as follows:

[0075]

[0076] In the second preferred embodiment of the present invention, the weighted super matrix is ​​continuously self-multiplied until the weighted super matrix reaches a stable state, where m is the number of self-multiplications performed when the stable state is reached, and the weight values ​​of each criterion under each surface are added together to obtain the weight value of each surface.

[0077] Figure 6 illustrates a schematic diagram of the policy simulation system and method integrating multi-criteria decision-making and system dynamics according to the third preferred embodiment of the present invention, which employs the Network Processing Analysis (DANP) method based on decision laboratory analysis for expert integration calculation. This corresponds to the expert integration calculation method based on the Network Processing Analysis (DANP) method in Figure 5. Referring to Figures 5 and 6, compared to the second preferred embodiment, the expert integration calculation method based on the Network Processing Analysis (DANP) method in the third preferred embodiment of the present invention has a more complex overall processing flow and increases the amount of data processed.

[0078] Please refer to Figures 4, 4A and 6. The expert integration calculation method based on Decision Laboratory Analysis (DANP) in the third preferred embodiment of the present invention includes step S10a: First, for example, the influence relationship 21 of several criteria is defined using several expert questionnaires 11 by appropriate technical means (e.g., automated, semi-automated or manual) in order to obtain several mutual influence relationship values ​​22 (as shown in step S10 of Figure 5).

[0079] Referring again to Figures 4, 4A and 6, the expert integration calculation method of the third preferred embodiment of the present invention based on the Decision Laboratory Analysis Network Program Analysis (DANP) includes step S20a: Next, for example, at least one direct influence matrix X is established using appropriate technical means (e.g., automated, semi-automated or manual) using several of the mutual influence relationship values ​​(as shown in step S20 of Figure 5).

[0080] Referring again to Figures 4, 4A and 6, the expert integration calculation method based on the Decision Laboratory Analysis Network Program Analysis (DANP) of the third preferred embodiment of the present invention includes step S30a: Next, for example, at least one normalized influence matrix D is established using the direct influence matrix X using appropriate technical means (e.g., automated, semi-automated or manual methods) (as shown in step S30 of Figure 5).

[0081] Referring again to Figures 4, 4A and 6, the expert integration calculation method based on Decision Laboratory Analysis (DANP) of the third preferred embodiment of the present invention includes step S40a: Next, for example, the normalized influence matrix is ​​used to perform calculations using appropriate technical means (e.g., automated, semi-automated or manual) to obtain at least one total influence matrix T (as shown in step S40 of Figure 5).

[0082] Referring again to Figures 4, 4A and 6, the expert integration calculation method based on Decision Laboratory Analysis (DANP) in the third preferred embodiment of the present invention includes step S50a: Next, for example, the total influence matrix T is normalized and transposed by appropriate technical means (e.g., automated, semi-automated or manual) to obtain an unweighted super-matrix W.

[0083] The total influence matrix TC of the calculation criteria in the third preferred embodiment of the present invention is as follows:

[0083]

[0083] where tij is the influence value of criterion i on criterion j, and n is the total number of criteria.

[0084] In the third preferred embodiment of the present invention, when calculating the normalization benchmark of the total influence matrix T, the normalization benchmark is set to fi, and the calculation method is as follows:

[0084]

[0084] The fi values ​​must be calculated separately for different planes. For example, criteria 1 to 5 belong to the first plane, and criteria 6 to 13 belong to the second plane. The normalization criterion for the first 5 rows of values ​​contained in the first plane is calculated by the sum of the column vectors of the first 5 rows in matrix TC, which can yield n fi values. The normalization criterion for the values ​​of the 6th to 13th rows contained in the second plane is calculated by the sum of the column vectors of the 6th to 13th rows in matrix TC, which can yield n fi values.

[0085] In the third preferred embodiment of the present invention, the total influence matrix T is normalized as follows:

[0085]

[0085] The normalization calculation must be performed according to the surface. Following the assumption of the previous step, the normalization process of the first 5 rows of values ​​contained in the first surface of the normalized total influence matrix is ​​to divide the values ​​of the first 5 rows by the normalization reference value of each column, and perform n calculations for each row; the normalization calculation of the values ​​of the 6th to 13th rows contained in the second surface is to divide the values ​​of the 6th to 13th rows by the normalization reference value of each column, and perform n calculations for each row.

[0086] In the third preferred embodiment of the present invention, the unweighted supermatrix W is obtained after normalizing and transposing the total influence matrix T as follows:

[0086]

[0087] Please refer again to Figures 4, 4A and 6. The expert integration calculation method based on Decision Laboratory Analysis (DANP) in the third preferred embodiment of the present invention includes step S60a: Next, for example, weighted calculation is performed using the unweighted supermatrix W by appropriate technical means (e.g., automated, semi-automated or manual) to obtain at least one weighted supermatrix.

[0088] In the third preferred embodiment of the present invention, the total influence matrix of the surface obtained by DEMATEL can be selected as the basis for weighted calculation of the unweighted super matrix W, and the total influence matrix of the surface calculated in the previous step can be normalized to form TD, which has the following form:

[0088]

[0088] where tD ij is the total influence value of surface i on surface j, and n is the total number of surfaces.

[0089] In the third preferred embodiment of the present invention, when calculating the normalization datum of the surface, the normalization datum is set to vi, and the total influence matrix of the normalized surface is set to TD*, the formula of which is as follows:

[0089]

[0090] In the third preferred embodiment of the present invention, after transposing the normalized total influence matrix of the surface, the weighted supermatrix S is obtained by multiplying it by a weighted value according to the relative relationship of each surface in the unweighted supermatrix. The formula is as follows:

[0090]

[0091] In the third preferred embodiment of the present invention, after normalizing the total influence matrix of the surface, it must be transposed, and then weighted according to the relative positions after transposition. Otherwise, the property that the sum of the row vectors of the weighted super matrix is ​​1 cannot be satisfied, which will lead to the inability to perform the operation of the limit super matrix. In addition, when performing weighted calculation, attention must also be paid to the relative positions of the criteria contained in the surface, otherwise calculation errors are easy to occur.

[0092] Referring again to Figures 4, 4A and 6, the expert integration calculation method of the third preferred embodiment of the present invention based on the Decision Laboratory Analysis-based Network Program Analysis (DANP) includes step S70a: Next, for example, the weighted supermatrix is ​​used to perform limit calculations by appropriate technical means (e.g., automated, semi-automated or manual) to obtain at least one limited supermatrix L in order to obtain the weight values ​​23 of several criteria (as shown in step S60 of Figure 5).

[0093] In the third preferred embodiment of the present invention, since the sum of the row vectors of the weighted super matrix S is 1, the value of each column of the matrix tends to stabilize during limit operations. Therefore, this characteristic is used to calculate the weight values ​​of each criterion. The limited super matrix L of the third preferred embodiment of the present invention is as follows:

[0093]

[0094] In the third preferred embodiment of the present invention, the weighted supermatrix is ​​continuously multiplied by itself until the weighted supermatrix reaches a stable state, where m is the number of self-multiplications performed when the stable state is reached.

[0095] Please refer again to Figures 5 and 6. Compared with the second preferred embodiment (steps S10 to S60 as shown in Figure 5), the expert integration calculation method of the third preferred embodiment of the present invention based on the Decision Laboratory Analysis-based Network Program Analysis (DANP) has a more complex overall processing flow and increased data processing volume (steps S10a to S70a as shown in Figure 6).

[0096] The foregoing preferred embodiments are merely illustrative of the present invention and its technical features. The technology of these embodiments can still be implemented with various substantially equivalent modifications and / or substitutions. Therefore, the scope of the present invention shall be determined by the scope defined in the appended claims. The copyright of this case is limited to the use of the patent application in the Republic of China.

Claims

1. A policy simulation system integrating multi-criteria decision-making and system dynamics, comprising: at least one data collection unit for collecting relevant policy data; at least one simulation evaluation indicator unit for initially simulating and establishing at least one evaluation indicator using the relevant policy data; at least one preliminary expert questionnaire unit for providing several preliminary expert questionnaires using the evaluation indicator to obtain at least one preliminary expert questionnaire result; and at least one hierarchical architecture of network procedure analysis, which establishes the hierarchical architecture of the network procedure analysis using the preliminary expert questionnaire result, wherein the hierarchical architecture of the network procedure analysis includes several criteria... Then; several ANP expert questionnaire units, which are used to establish several ANP expert questionnaires in order to obtain at least one network procedure analysis expert questionnaire result, and use the network procedure analysis expert questionnaire result to calculate weights to obtain at least one model weight value; and a calculation unit, which has a causal feedback module, and uses the expert questionnaire result to establish an indicator of at least one system dynamics model, and uses the network procedure analysis expert questionnaire result and the indicator of the system dynamics model to draw at least one causal feedback graph, so as to make a policy decision by integrating multi-criteria decision-making and system dynamics.

2. A policy simulation system integrating multi-criteria decision-making and system dynamics according to claim 1, wherein the influence relationships of several criteria are adopted, and the influence relationships of the several criteria are defined using several ANP expert questionnaires.

3. A policy simulation system integrating multi-criteria decision-making and system dynamics according to claim 2, wherein several mutual influence relationship values ​​are used, and the several mutual influence relationship values ​​are obtained from the influence relationship of several criteria.

4. A policy simulation system integrating multi-criteria decision-making and system dynamics according to claim 3, wherein the calculation unit further includes at least one direct influence matrix, and the direct influence matrix is ​​established using several of the mutual influence relationship values.

5. A policy simulation system integrating multi-criteria decision-making and system dynamics according to claim 4, wherein the calculation unit further includes a normalization module, and uses the direct influence matrix to establish at least one normalized influence matrix, and uses the normalized influence matrix to perform calculations to obtain at least one total influence matrix, and performs normalization processing on the total influence matrix to obtain a normalized total influence matrix, and uses the normalized total influence matrix to perform limit calculations to obtain at least one limited super matrix, so as to obtain the weight values ​​of several criteria.

6. A method for policy simulation integrating multi-criteria decision-making and system dynamics, comprising: collecting relevant policy data; using the relevant policy data to initially simulate and establish at least one evaluation indicator; using the evaluation indicator to provide several preliminary expert questionnaires to obtain at least one preliminary expert questionnaire result; using the preliminary expert questionnaire result to establish a hierarchical architecture of network procedure analysis, wherein the hierarchical architecture of network procedure analysis includes several criteria; establishing several expert questionnaires for network procedure analysis to obtain at least one expert questionnaire result for network procedure analysis; using the expert questionnaire result for network procedure analysis to perform weight calculation to obtain at least one model weight value; using the expert questionnaire result to establish an indicator of at least one system dynamics model; and using the expert questionnaire result for network procedure analysis and the indicator of the system dynamics model to draw at least one causal feedback graph, so as to conduct a policy decision in an integrated multi-criteria decision-making and system dynamics manner.

7. A policy simulation method integrating multi-criteria decision-making and system dynamics according to claim 6, wherein several expert questionnaires of the network procedure analysis method are used to define the influence relationship of several criteria in order to obtain several mutual influence relationship values.

8. The method for integrating multi-criteria decision-making and system dynamics for policy simulation according to claim 7, wherein at least one direct influence matrix is ​​established using several of the mutual influence relationship values, and at least one normalized influence matrix is ​​established using the direct influence matrix.

9. A policy simulation method integrating multi-criteria decision-making and system dynamics according to claim 8, wherein the normalized influence matrix is ​​used to calculate at least one total influence matrix, and the total influence matrix is ​​normalized to obtain a normalized total influence matrix.

10. A policy simulation method integrating multi-criteria decision-making and system dynamics according to claim 9, wherein the normalized total influence matrix is ​​used to perform limit calculations to obtain at least one limited super matrix in order to obtain the weight values ​​of several criteria.

Images