Rapid tracking and positioning method for enterprise digital development maturity

Abstract

The invention belongs to the technical field of enterprise digitization, and particularly relates to a rapid tracking and positioning method for enterprise digitization development maturity. Comprising the following steps: S1, establishing a digital development maturity evaluation framework; s2, index system screening is carried out based on a fuzzy Delphi method; s3, establishing an improved K-means model based on density and weight; s4, carrying out enterprise digital development grade division; s5, establishing an enterprise digital transformation mature stage correction model; according to the method, a digital development maturity evaluation framework is constructed, index system screening based on a fuzzy Delphi method is utilized, an improved K-means model based on density and weight is combined, enterprise digital development grade division and an enterprise digital transformation maturity stage correction model are combined, enterprises are helped to quickly know the current situation of own digital development, and the evaluation accuracy of the digital development maturity is improved. Problems can be found in time and targeted measures can be taken so that a problem that rapid and accurate tracking and positioning of enterprise digital development maturity are lacked in the prior art can be solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of enterprise digitalization, and in particular relates to a method for quickly tracking and locating the maturity of enterprise digitalization development. Background Art

[0002] With the rapid development of new technologies such as the Internet, big data, cloud computing, and artificial intelligence, enterprises are experiencing a transition from the industrial economy era to the digital economy era. This transition has not only changed the operating model of enterprises, but also spawned new formats and business models. At present, the digital economy has entered a stage of rapid development, the digital transformation of traditional industries has been accelerating, the digital economy infrastructure has achieved leapfrog progress, and new formats and models have flourished.

[0003] At present, domestic and foreign scholars have made certain progress in their research on enterprise digital transformation, mainly focusing on the strategy of digital transformation, the relationship between digital transformation and market micro-subjects, the impact of enterprise digital transformation on enterprise innovation, enterprise value, production efficiency, etc., and the impact of digital economy supervision on enterprise digital transformation. Most of the existing research focuses on the economic consequences of digital transformation, which helps enterprises grasp the direction and rhythm of digital transformation to a certain extent, but lacks practical and operational rapid tracking and positioning methods. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a method for quickly tracking and locating the maturity of enterprise digital development, to provide enterprises with an efficient and accurate tool to help them quickly understand the current status of their own digital development, to promptly identify problems and take targeted measures to promote the smooth progress of the enterprise's digital transformation.

[0005] The object of the present invention is achieved by: a method for quickly tracking and locating the maturity of digital development of an enterprise, comprising the following steps:

[0006] S1. Establish a digital development maturity evaluation framework;

[0007] S2, screening of indicator system based on fuzzy Delphi method;

[0008] S3. Establish an improved K-means model based on density and weight;

[0009] S4. Classify the digital development levels of enterprises;

[0010] S5. Establish a correction model for the mature stage of enterprise digital transformation.

[0011] Furthermore, the step S1 of establishing a digital development maturity evaluation framework includes:

[0012] Firstly, based on the constructed elements of enterprise digital maturity, the expert survey method is used to obtain the relative importance scores of enterprise digital maturity evaluation indicators; secondly, the hierarchical analysis method is used to obtain the weights of each evaluation indicator; the fuzzy comprehensive evaluation method is introduced to evaluate the maturity of each enterprise digital indicator and analyze the evaluation results; finally, corresponding development suggestions are put forward for indicators with lower maturity levels.

[0013] Furthermore, the step S2 of screening the index system based on the fuzzy Delphi method includes:

[0014] (1) Determine the specific steps of the fuzzy Delphi model

[0015] 1) By consulting with selected experts, determine the upper and lower bounds of the importance interval of each indicator in the indicator system; the importance interval of each indicator is [0,10], and the larger the value, the more important the indicator; the smaller the lower bound represents the conservative value interval, and the larger the upper bound represents the optimistic value interval;

[0016] 2) Calculate the conservative triangular fuzzy function of each indicator and optimistic triangular fuzzy function in and Represents the minimum value of the conservative and optimistic values ​​of the experts’ judgment on indicator i; and Represents the maximum value of the conservative and optimistic values ​​of the experts’ judgment on indicator i; and Represents the geometric mean of the minimum and maximum values ​​of the optimistic and conservative values ​​of indicator i;

[0017] 3) Verify the consistency of expert judgment results and calculate the importance consistency value G of indicator i i ;

[0018] ①If This means that the judgment of indicator i is consistent, so G i The calculation is as follows:

[0019]

[0020] ②If Need to determine the grayscale range value and relationship;

[0021] If Z i ≤M i , then the consistent value G i It can be obtained by the following formula:

[0022]

[0023] If Zi >M i , indicating that there is a contradiction in the experts' judgments. At this time, the experts need to repeat steps 1) and 2) until the judgment results are consistent; the importance consistency value G i G not only refers to the consistency of experts' judgment on indicator i, i The size of G also indicates the importance of index i; i The larger the value is, the more important the indicator i is. In order to test the significance of the importance of each indicator, it is necessary to determine a reasonable threshold value, and take the geometric mean of the upper and lower bounds of all indicators as the threshold value for screening the final indicator.

[0024] (2) Screening the index system: selecting experts from relevant fields and, based on the experts’ judgment results, using the fuzzy Delphi method to screen out key indicators to form an evaluation index system;

[0025] First, the experts give their judgment results on the importance of each initial indicator, and then obtain the conservative value and optimistic value of each indicator; secondly, the conservative triangular fuzzy function of each indicator is calculated based on this. and optimistic triangular fuzzy function Then, judge and The size relationship of G can be calculated using formula (1) or (2). i To test the consistency of expert judgment.

[0026] Furthermore, the step S3 of establishing an improved K-means model based on density and weight includes:

[0027] The data set M = {x1, x2, x3, ..., x n Cluster into k clusters. First, calculate the variance of the sample point data by the following formula (6), find the sample point with the largest density as the initial cluster center, take the average value of the weighted distance of the sample data points as the radius, and the sample data points in this area constitute the set M1; select the sample point with the smallest variance in M-M1, take the average value of the weighted distance of all the remaining sample data points as the radius, and the sample data points in this area constitute the set M2; repeat the above steps until the data set M to be clustered is divided into K sets located in different areas; then take the mean of each set M1, M2, ..., MK as the new cluster center of the set; finally, calculate the sum of squared errors of the clusters by formula (8);

[0028] (1) Determine the initial cluster center

[0029] 1) Calculate sample variance

[0030] When calculating the distance, the weight calculation formula for the data of different dimensions of the sample points is:

[0031]

[0032] In the formula, x id Represents the value of the dth component of the ith sample; Represents the average value of the dth component in the sample data set;

[0033] Sample point x i ,x j The Euclidean distance between is,

[0034]

[0035] Sample point x i The average distance to all sample points is,

[0036]

[0037] Sample point x i The variance of is,

[0038]

[0039] The average Euclidean distance of sample data points is,

[0040]

[0041] The square of clustering error is,

[0042]

[0043] 2) Find the sample point with the smallest variance As the initial cluster center; get the first cluster set

[0044] 3) Let M = M-M1, and find the sample point with the smallest variance in the new data set As the cluster center, with the average distance of sample points as the radius, we get the second cluster set M2 = {d(x j ,x i 2 )<cmean,j=1,2,…,n};

[0045] 4) Repeat the above steps until K mutually disjoint sets are found;

[0046] (2) Construct initialization points

[0047] 1) According to formula (4), the Euclidean distance from each sample point to the selected K initial cluster centers is obtained, and then the sample points are divided into the nearest class to form the initial division;

[0048] 2) Calculate the mean of each class of the initial division and use it as the new center of the class;

[0049] 3) Calculate the sum of square errors of clustering results by formula (8);

[0050] (3) Iterative update clustering

[0051] 1) Based on the last clustering, a new cluster center is obtained, and the Euclidean distance from the sample data point to the new center is calculated by formula (4) and assigned to the nearest class;

[0052] 2) Calculate the mean of each class and use it as the new center of the class;

[0053] 3) Calculate the sum of square errors of clustering results by formula (8);

[0054] 4) Compare this result with the sum of squared errors of the previous clustering. If E'-E<10 -10 , the cluster center does not change, the iteration is terminated, and the clustering result is output; otherwise, continue the above steps until the clustering result converges.

[0055] Furthermore, the step S4 of classifying the digital development level of the enterprise includes:

[0056] The digital transformation of enterprises follows the idea of ​​"from easy to difficult, from point to surface, long-term iteration, and multi-party collaboration". In this way, the maturity of enterprise digital transformation can be divided into 5 categories: levels 1-5; among them, enterprises at the active exploration level have the awareness of digital transformation, begin to explore and start related preparations; enterprises at the unit practice level begin to use a single business field as an entry point, use digital technology to transform and standardize corresponding activities, and achieve single-point efficiency improvement; enterprises at the field innovation level begin to integrate multiple businesses, reduce manual intervention in the business, and achieve efficient collaboration of applications and data; enterprises at the comprehensive transformation level are characterized by a high degree of autonomy in all fields and links, opening up all factors and all-system processes, and realizing effective docking of supply and demand, coordinated resource scheduling, and coordinated quality control; enterprises at the innovation leading level are in the advanced stage of transformation. They will continuously iterate and optimize their own businesses, explore new business models, open up the ecological value network, and achieve sustainable and resilient development of the enterprise.

[0057] Furthermore, the step S5 of establishing a correction model for the mature stage of enterprise digital transformation includes:

[0058] The fuzzy comprehensive evaluation method considers multiple factors, makes a comprehensive judgment on a matter, and can handle fuzzy information that other methods cannot handle. It is a decision-making process that combines qualitative and quantitative methods. The fuzzy comprehensive evaluation method is used to construct an enterprise digital transformation maturity evaluation model. The specific steps are as follows;

[0059] (1) Establish the evaluation factor set U of fuzzy comprehensive evaluation

[0060] U is a set of factors that affect the evaluation object, U = {u1,u2,...,u m}, where u i (i=1,2,3,…,m) is the influencing factor of the evaluation object;

[0061] (2) Establish a comprehensive evaluation review set V

[0062] The comment set is a set of various possible evaluation results made by the evaluator on the evaluation object, V = {v1, v2, ..., v n}, and divide it into five levels, v = {very good, good, average, poor, very poor}, corresponding to the proposed enterprise digital transformation maturity level, where v j (j=1,2,3,…,n) is the jth evaluation result, and each level is equivalent to a fuzzy subset;

[0063] (3) Establish the fuzzy comprehensive evaluation matrix R

[0064] First determine the membership of the evaluation object to the evaluation set; suppose the evaluation object is based on the i-th factor u in the factor set U i Evaluate the jth factor v in the evaluation set V j The membership degree is r ij , then the evaluation result of Ui can be expressed by fuzzy sets,

[0065] R i =(r i1 ,r i2 ,…r im ) (9)

[0066] After the single factor fuzzy evaluation, all factors are evaluated separately using the same principle to construct a total evaluation matrix, that is, each evaluated indicator determines the fuzzy relationship R from U to V, and the fuzzy comprehensive evaluation matrix can be obtained.

[0067]

[0068] R is called the fuzzy comprehensive evaluation matrix, which represents a fuzzy relationship between the factor set U and the evaluation set V; ij (i=1,2,…,m;j=1,2,…n) indicates the possibility of making the jth comment on the evaluation object based on the i-th factor;

[0069] (4) Determine the weight vector of evaluation factors

[0070] Assign a corresponding weight w to each factor U i (i=1,2,…,m) represents the weight of the i-th factor, wi Satisfy w i ≥0,∑w i =1, and then the weights are combined into a weight set W, and the weight coefficient is determined by the hierarchical analysis method;

[0071] (5) Establishing a fuzzy comprehensive evaluation model

[0072] Using appropriate synthesis operators to synthesize W and the fuzzy relationship matrix R, we can get the fuzzy evaluation result vector C of each evaluated object. The model of fuzzy comprehensive evaluation is:

[0073]

[0074] where c j (j=1,2,…n) is obtained by calculating the jth column of W and R, indicating the overall evaluation of the rated object on v j The degree of membership of the hierarchical fuzzy subset; according to the basic principle of maximum membership, the comment item with the largest proportion is the final evaluation conclusion, that is, the enterprise digital transformation maturity stage corresponding to each indicator;

[0075] The first-level indicators with digital maturity at levels IV and V are in good development status and should continue according to their current plans; while the first-level indicators at levels I, II, and III are less mature, and corresponding suggestions are put forward to promote their development to the mature stage.

[0076] Beneficial effects of the present invention: A method for quickly tracking and locating the maturity of digital development of an enterprise of the present invention, through the method of quickly tracking and locating, an enterprise can quickly understand its own digital maturity, and compared with the traditional evaluation method, the evaluation cycle is greatly shortened. The fuzzy Delphi method and the improved K-means model are used to improve the accuracy of the evaluation and reduce the interference of human factors. Through the evaluation framework, enterprises can timely discover the weak links in digital development, take targeted measures, and use big data and algorithms to assist in decision-making, turning from "experience decision-making" to "data decision-making", and improving the comprehensiveness and clarity of decision-making. Maturity standardization helps guide enterprises to fully understand and scientifically master the laws of digital transformation, enhance the consensus of all employees on transformation, and enterprises can systematically plan the digital transformation development roadmap by accurately locating their own digital transformation development stage, and gradually improve the level and ability of digital transformation. This technical solution adds scientific evaluation and correction. Through real-time monitoring and rapid positioning of digital development, enterprises can adjust their strategies in a timely manner, reduce risks in the process of digital transformation, and ensure the smooth progress of transformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0078] Figure 1 This is a schematic diagram of the digital development maturity evaluation framework structure;

[0079] Figure 2 A schematic diagram of recommendations for improving enterprise digital maturity. DETAILED DESCRIPTION

[0080] The present invention will be further described below in conjunction with the accompanying drawings.

[0081] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0082] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0083] like Figure 1-2 As shown, a method for quickly tracking and locating the digital development maturity of an enterprise of the present invention comprises the following steps:

[0084] S1. Establish a digital development maturity evaluation framework;

[0085] S2, screening of indicator system based on fuzzy Delphi method;

[0086] S3. Establish an improved K-means model based on density and weight;

[0087] S4. Classify the digital development levels of enterprises;

[0088] S5. Establish a correction model for the mature stage of enterprise digital transformation.

[0089] Furthermore, the step S1 of establishing a digital development maturity evaluation framework includes:

[0090] Firstly, according to the constructed enterprise digital maturity components, the expert survey method is used to obtain the relative importance scores of enterprise digital maturity evaluation indicators; secondly, the analytic hierarchy process is used to obtain the weights of each evaluation indicator; the fuzzy comprehensive evaluation method is introduced to evaluate the maturity of each enterprise digitalization indicator and analyze the evaluation results; finally, corresponding development suggestions are put forward for indicators with lower maturity levels; the specific structure is as follows: Figure 1 shown.

[0091] The Delphi method is essentially a feedback anonymous inquiry method. It is an effective and reliable method for collecting expert judgments and opinions, and is suitable for research such as decision-making and forecasting. However, after obtaining the opinions of experts, the traditional Delphi method needs to organize, summarize, and count them, and then anonymously feedback them to each expert, and repeat this process until a consensus is reached. Generally, this process needs to be repeated four times, which is undoubtedly time-consuming and inefficient, and it is difficult to obtain a unified opinion through consultation feedback.

[0092] The fuzzy Delphi method can effectively solve the above problems. The fuzzy Delphi method combines fuzzy theory and traditional Delphi method. In the fuzzy Delphi method, the invited experts need to use triangular fuzzy numbers to make three-point evaluations of indicators. The membership function is used to verify the expert's judgment results, so there is no need to verify the expert's judgment round by round. The membership function can make the most of the information given by the expert's judgment. Due to the superiority of the fuzzy Delphi method, many studies use it to construct an evaluation index system.

[0093] Furthermore, the step S2 of screening the index system based on the fuzzy Delphi method includes:

[0094] (1) Determine the specific steps of the fuzzy Delphi model

[0095] 1) By consulting with selected experts, determine the upper and lower bounds of the importance interval of each indicator in the indicator system; the importance interval of each indicator is [0,10], and the larger the value, the more important the indicator; the smaller the lower bound represents the conservative value interval, and the larger the upper bound represents the optimistic value interval;

[0096] 2) Calculate the conservative triangular fuzzy function of each indicator and optimistic triangular fuzzy function in and Represents the minimum value of the conservative and optimistic values ​​of the experts’ judgment on indicator i; and Represents the maximum value of the conservative and optimistic values ​​of the experts’ judgment on indicator i; and Represents the geometric mean of the minimum and maximum values ​​of the optimistic and conservative values ​​of indicator i;

[0097] 3) Verify the consistency of expert judgment results and calculate the importance consistency value G of indicator i i ;

[0098] ①If This means that the judgment of indicator i is consistent, so G i The calculation is as follows:

[0099]

[0100] ②If Need to determine the grayscale range value and relationship;

[0101] If Z i ≤M i , then the consistent value G i It can be obtained by the following formula:

[0102]

[0103] If Z i >M i , indicating that there is a contradiction in the experts' judgments. At this time, the experts need to repeat steps 1) and 2) until the judgment results are consistent; the importance consistency value G i G not only refers to the consistency of experts' judgment on indicator i, i The size of G also indicates the importance of index i; i The larger the value is, the more important the indicator i is. In order to test the significance of the importance of each indicator, it is necessary to determine a reasonable threshold value, and take the geometric mean of the upper and lower bounds of all indicators as the threshold value for screening the final indicator.

[0104] (2) Screening the index system: selecting experts from relevant fields and, based on the experts’ judgment results, using the fuzzy Delphi method to screen out key indicators to form an evaluation index system;

[0105] First, the experts give the results of their judgment on the importance of each initial indicator, and then obtain the conservative value and optimistic value of each indicator; secondly, the conservative triangular fuzzy function of each indicator is calculated based on this. and optimistic triangular fuzzy function Then, judge and The size relationship of G can be calculated using formula (1) or (2). i To test the consistency of expert judgment.

[0106] K-means clustering algorithm is a widely used clustering algorithm based on partitioning. The traditional K-means clustering algorithm uses the sum of squared errors as the criterion function for judging the clustering effect. Its basic idea is: first, randomly select K samples from a set containing R samples as the initial clustering centers, and assign the samples to the most similar clustering centers according to the Euclidean distance of each sample to the K centers, so as to obtain K non-intersecting clusters; recalculate the new centers of the K clusters, and then assign the R samples to the most similar clusters according to the Euclidean distance principle. Repeat this process until the centers of the K clusters no longer change, so as to obtain K non-intersecting stable clusters of the original set.

[0107] The traditional K-means clustering algorithm has randomness in determining the initial cluster center, and may also select isolated points or noise points, which may cause the clustering results to be inconsistent with the actual distribution of the sample data set and fail to obtain the correct clustering results. The K-means clustering algorithm based on sample density and weight improvement takes the highest compactness of the sample space as the principle, and then uses the average distance after introducing weights as the radius, and selects K initial cluster centers located in different areas, which can avoid randomness and reduce the impact of outliers on clustering results.

[0108] Furthermore, the step S3 of establishing an improved K-means model based on density and weight includes:

[0109] The data set M = {x1, x2, x3, ..., x n Cluster into k clusters. First, calculate the variance of the sample point data by the following formula (6), find the sample point with the largest density as the initial cluster center, take the average value of the weighted distance of the sample data points as the radius, and the sample data points in this area constitute the set M1; select the sample point with the smallest variance in M-M1, take the average value of the weighted distance of all the remaining sample data points as the radius, and the sample data points in this area constitute the set M2; repeat the above steps until the data set M to be clustered is divided into K sets located in different areas; then take the mean of each set M1, M2, ..., MK as the new cluster center of the set; finally, calculate the sum of squared errors of the clusters by formula (8);

[0110] (1) Determine the initial cluster center

[0111] 1) Calculate sample variance

[0112] When calculating the distance, the weight calculation formula for the data of different dimensions of the sample points is:

[0113]

[0114] In the formula, x idRepresents the value of the dth component of the ith sample; Represents the average value of the dth component in the sample data set;

[0115] Sample point x i ,x j The Euclidean distance between is,

[0116]

[0117] Sample point x i The average distance to all sample points is,

[0118]

[0119] Sample point x i The variance of is,

[0120]

[0121] The average Euclidean distance of sample data points is,

[0122]

[0123] The square of clustering error is,

[0124]

[0125] 2) Find the sample point with the smallest variance As the initial cluster center; get the first cluster set

[0126] 3) Let M = M-M1, and find the sample point with the smallest variance in the new data set As the cluster center, with the average distance of sample points as the radius, we get the second cluster set M2 = {d(x j ,x i 2 )<cmean,j=1,2,…,n};

[0127] 4) Repeat the above steps until K mutually disjoint sets are found;

[0128] (2) Construct initialization points

[0129] 1) According to formula (4), the Euclidean distance from each sample point to the selected K initial cluster centers is obtained, and then the sample points are divided into the nearest class to form the initial division;

[0130] 2) Calculate the mean of each class of the initial division and use it as the new center of the class;

[0131] 3) Calculate the sum of square errors of clustering results by formula (8);

[0132] (3) Iterative update clustering

[0133] 1) Based on the last clustering, a new cluster center is obtained, and the Euclidean distance from the sample data point to the new center is calculated by formula (4) and assigned to the nearest class;

[0134] 2) Calculate the mean of each class and use it as the new center of the class;

[0135] 3) Calculate the sum of square errors of clustering results by formula (8);

[0136] 4) Compare this result with the sum of squared errors of the previous clustering. If E'-E<10 -10 , the cluster center does not change, the iteration is terminated, and the clustering result is output; otherwise, continue the above steps until the clustering result converges.

[0137] Furthermore, the step S4 of classifying the digital development level of the enterprise includes:

[0138] The digital transformation of enterprises follows the idea of ​​"from easy to difficult, from point to surface, long-term iteration, and multi-party collaboration". In this way, the maturity of enterprise digital transformation can be divided into 5 categories: levels 1-5; among them, enterprises at the active exploration level have the awareness of digital transformation, begin to explore and start related preparations; enterprises at the unit practice level begin to use a single business field as an entry point, use digital technology to transform and standardize corresponding activities, and achieve single-point efficiency improvement; enterprises at the field innovation level begin to integrate multiple businesses, reduce manual intervention in the business, and achieve efficient collaboration of applications and data; enterprises at the comprehensive transformation level are characterized by a high degree of autonomy in all fields and links, opening up all factors and all-system processes, and realizing effective docking of supply and demand, coordinated resource scheduling, and coordinated quality control; enterprises at the innovation leading level are in the advanced stage of transformation. They will continuously iterate and optimize their own businesses, explore new business models, open up the ecological value network, and achieve sustainable and resilient development of the enterprise.

[0139] As a multi-criteria decision-making tool, the analytic hierarchy process (AHP) decomposes complex decision-making problems into multiple criteria and sub-criteria by constructing a hierarchical model in the evaluation of enterprise digital transformation, and then uses a pairwise comparison method to determine the relative importance of each criterion. In the process of digital transformation, enterprises need to make choices among many alternatives, such as technology selection, resource allocation, risk assessment, etc. These decisions often involve uncertainty and complexity. The AHP method quantifies the subjective judgment of decision makers, helps enterprises comprehensively evaluate the priorities of alternatives from multiple dimensions, and finally derives an overall priority, providing scientific decision support for decision makers. This method not only conforms to the behavioral patterns of decision makers, but also can effectively deal with the risks and uncertainties in the process of digital transformation, and promote enterprises to make more informed decisions.

[0140] The fuzzy comprehensive evaluation method comprehensively evaluates relevant factors based on fuzzy transformation theory and the principle of maximum membership. It is an effective method for evaluating objects affected by multiple factors. For objects affected by several factors, a one-layer model can be used. If the object is more complex and has more factors, a two-layer or multi-layer model can be used. Fuzzy comprehensive evaluation uses the comprehensive evaluation model of fuzzy mathematics to evaluate the digital maturity index and draw conclusions. The key to fuzzy comprehensive evaluation is to determine the factor set and reasonably construct the fuzzy evaluation matrix.

[0141] The AHP-fuzzy comprehensive evaluation method is used to comprehensively evaluate the digital maturity of an enterprise. The hierarchical analysis method can clarify the hierarchical relationship between these influencing factors and determine the weight of each indicator, which is helpful to give specific suggestions for improving the digital maturity level of the enterprise. The fuzzy comprehensive evaluation method can obtain the evaluation results of the digital maturity of the enterprise, thereby scientifically and rationally evaluating the digital maturity of the enterprise and giving corresponding suggestions.

[0142] Furthermore, the step S5 of establishing a correction model for the mature stage of enterprise digital transformation includes:

[0143] The fuzzy comprehensive evaluation method considers multiple factors, makes a comprehensive judgment on a matter, and can handle fuzzy information that other methods cannot handle. It is a decision-making process that combines qualitative and quantitative methods. The fuzzy comprehensive evaluation method is used to construct an enterprise digital transformation maturity evaluation model. The specific steps are as follows;

[0144] (1) Establish the evaluation factor set U of fuzzy comprehensive evaluation

[0145] U is a set of factors that affect the evaluation object, U = {u1,u2,...,u m}, where u i (i=1,2,3,…,m) is the influencing factor of the evaluation object;

[0146] (2) Establish a comprehensive evaluation review set V

[0147] The comment set is a set of various possible evaluation results made by the evaluator on the evaluation object, V = {v1, v2, ..., v n}, and divide it into five levels, v = {very good, good, average, poor, very poor}, corresponding to the proposed enterprise digital transformation maturity level, where v j (j=1,2,3,…,n) is the jth evaluation result, and each level is equivalent to a fuzzy subset;

[0148] (3) Establish the fuzzy comprehensive evaluation matrix R

[0149] First determine the membership of the evaluation object to the evaluation set; suppose the evaluation object is based on the i-th factor u in the factor set U i Evaluate the jth factor v in the evaluation set V j The membership degree is r ij , then the evaluation result of Ui can be expressed by fuzzy sets,

[0150] R i =(r i1 ,r i2 ,…r im ) (9)

[0151] After the single factor fuzzy evaluation, all factors are evaluated separately using the same principle to construct a total evaluation matrix, that is, each evaluated indicator determines the fuzzy relationship R from U to V, and the fuzzy comprehensive evaluation matrix can be obtained.

[0152]

[0153] R is called the fuzzy comprehensive evaluation matrix, which represents a fuzzy relationship between the factor set U and the evaluation set V; ij (i=1,2,…,m;j=1,2,…n) indicates the possibility of making the jth comment on the evaluation object based on the i-th factor;

[0154] (4) Determine the weight vector of evaluation factors

[0155] In order to reflect the importance of each factor, each factor U is assigned a corresponding weight w. i (i=1,2,…,m) represents the weight of the i-th factor, w i Satisfy w i ≥0,∑w i =1, and then the weights are combined into a weight set W, and the weight coefficient is determined by the hierarchical analysis method;

[0156] (5) Establishing a fuzzy comprehensive evaluation model

[0157] Using appropriate synthesis operators to synthesize W and the fuzzy relationship matrix R, we can get the fuzzy evaluation result vector C of each evaluated object. The model of fuzzy comprehensive evaluation is:

[0158]

[0159] where c j (j=1,2,…n) is obtained by calculating the jth column of W and R, indicating the overall evaluation of the rated object on v j The degree of membership of the hierarchical fuzzy subset; according to the basic principle of maximum membership, the comment item with the largest proportion is the final evaluation conclusion, that is, the enterprise digital transformation maturity stage corresponding to each indicator;

[0160] The first-level indicators with digital maturity at levels IV and V are in good development and should continue according to their current plans; while the first-level indicators at levels I, II, and III are less mature, and corresponding suggestions are put forward to promote their development to the mature stage, such as Figure 2 shown.

[0161] In summary, a method for fast tracking and positioning of the maturity of digital development of an enterprise of the present invention can enable an enterprise to quickly understand its digital maturity through the fast tracking and positioning method, which greatly shortens the evaluation cycle compared with the traditional evaluation method. The fuzzy Delphi method and the improved K-means model are used to improve the accuracy of the evaluation and reduce the interference of human factors. Through the evaluation framework, enterprises can timely discover the weak links in digital development, take targeted measures, and use big data and algorithms to assist in decision-making, turning from "experience decision-making" to "data decision-making" to improve the comprehensiveness and clarity of decision-making. Maturity standardization helps guide enterprises to fully understand and scientifically master the laws of digital transformation, enhance the consensus of all employees on transformation, and enterprises can systematically plan the development roadmap of digital transformation by accurately positioning their own digital transformation development stage, and gradually improve the level and ability of digital transformation. This technical solution adds scientific evaluation and correction. Through real-time monitoring and rapid positioning of digital development, enterprises can adjust their strategies in time, reduce the risks in the process of digital transformation, and ensure the smooth progress of transformation.

[0162] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0163] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and 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, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A method for quickly tracking and locating the maturity of digital development of an enterprise, characterized in that: The following steps are involved: S1. Establish a digital development maturity evaluation framework; S2, screening of indicator system based on fuzzy Delphi method; S3, establish an improved K-means model based on density and weight; S4. Classify the digital development levels of enterprises; S5. Establish a correction model for the mature stage of enterprise digital transformation.

2. A method for quickly tracking and locating the digital development maturity of an enterprise as claimed in claim 1, characterized in that: The step S1 of establishing a digital development maturity evaluation framework includes: Firstly, based on the constructed elements of enterprise digital maturity, the expert survey method is used to obtain the relative importance scores of enterprise digital maturity evaluation indicators; secondly, the hierarchical analysis method is used to obtain the weights of each evaluation indicator; the fuzzy comprehensive evaluation method is introduced to evaluate the maturity of each enterprise digital indicator and analyze the evaluation results; finally, corresponding development suggestions are put forward for indicators with lower maturity levels.

3. A method for quickly tracking and locating the digital development maturity of an enterprise as claimed in claim 1, characterized in that: The step S2 of screening the index system based on the fuzzy Delphi method includes: (1) Determine the specific steps of the fuzzy Delphi model 1) By consulting with selected experts, determine the upper and lower bounds of the importance interval of each indicator in the indicator system; the importance interval of each indicator is [0,10], and the larger the value, the more important the indicator; the smaller the lower bound represents the conservative value interval, and the larger the upper bound represents the optimistic value interval; 2) Calculate the conservative triangular fuzzy function of each indicator and optimistic triangular fuzzy function in and Represents the minimum value of the conservative and optimistic values ​​of the experts’ judgment on indicator i; and Represents the maximum value of the conservative and optimistic values ​​of the experts’ judgment on indicator i; and Represents the geometric mean of the minimum and maximum values ​​of the optimistic and conservative values ​​of indicator i; 3) Verify the consistency of expert judgment results and calculate the importance consistency value G of indicator i i ; ①If This means that the judgment of indicator i is consistent, so G i The calculation is as follows: ②If Need to determine the grayscale range value and relationship; If Z i ≤M i , then the consistent value G i It can be obtained by the following formula: If Z i >M i , indicating that there is a contradiction in the experts' judgments. In this case, the experts need to repeat steps 1) and 2) until the judgment results are consistent; (2) Screening the index system: selecting experts from relevant fields and, based on the experts’ judgment results, using the fuzzy Delphi method to screen out key indicators to form an evaluation index system; First, the experts give their judgment results on the importance of each initial indicator, and then obtain the conservative value and optimistic value of each indicator; secondly, the conservative triangular fuzzy function of each indicator is calculated based on this. and optimistic triangular fuzzy function Then, judge and The size relationship of G can be calculated using formula (1) or (2). i To test the consistency of expert judgment.

4. A method for quickly tracking and locating the digital development maturity of an enterprise as claimed in claim 1, characterized in that: The step S3 of establishing an improved K-means model based on density and weight includes: The data set M = {x1, x2, x3, ..., x n Cluster into k clusters. First, calculate the variance of the sample point data by the following formula (6), find the sample point with the largest density as the initial cluster center, take the average value of the weighted distance of the sample data points as the radius, and the sample data points in this area constitute the set M1; select the sample point with the smallest variance in M-M1, take the average value of the weighted distance of all the remaining sample data points as the radius, and the sample data points in this area constitute the set M2; repeat the above steps until the data set M to be clustered is divided into K sets located in different areas; then take the mean of each set M1, M2, ..., MK as the new cluster center of the set; finally, calculate the sum of squared errors of the clusters by formula (8); (1) Determine the initial cluster center 1) Calculate sample variance When calculating the distance, the weight calculation formula for the data of different dimensions of the sample points is: In the formula, x id Represents the value of the dth component of the ith sample; Represents the average value of the dth component in the sample data set; Sample point x i ,x j The Euclidean distance between is, Sample point x i The average distance to all sample points is, Sample point x i The variance of is, The average Euclidean distance of sample data points is, The square of clustering error is, 2) Find the sample point with the smallest variance As the initial cluster center; get the first cluster set 3) Let M = M-M1, and find the sample point with the smallest variance in the new data set As the cluster center, with the average distance of sample points as the radius, the second cluster set is obtained 4) Repeat the above steps until K mutually disjoint sets are found; (2) Construct initialization points 1) According to formula (4), the Euclidean distance from each sample point to the selected K initial cluster centers is obtained, and then the sample points are divided into the nearest class to form the initial division; 2) Calculate the mean of each class of the initial division and use it as the new center of the class; 3) Calculate the sum of square errors of clustering results by formula (8); (3) Iterative update clustering 1) Based on the last clustering, a new cluster center is obtained, and the Euclidean distance from the sample data point to the new center is calculated by formula (4) and assigned to the nearest class; 2) Calculate the mean of each class and use it as the new center of the class; 3) Calculate the sum of square errors of clustering results by formula (8); 4) Compare this result with the sum of squared errors of the previous clustering. If E'-E<10 -10 , the cluster center does not change, the iteration is terminated, and the clustering result is output; otherwise, continue the above steps until the clustering result converges.

5. A method for quickly tracking and locating the digital development maturity of an enterprise as claimed in claim 1, characterized in that: The step S4 of classifying the digital development level of the enterprise includes: The maturity of enterprise digital transformation is divided into five categories: levels 1-5; among them, enterprises at the active exploration level have the awareness of digital transformation, begin to explore and initiate relevant preparatory work; enterprises at the unit practice level begin to use a single business field as the entry point, use digital technology to transform and standardize corresponding activities, and achieve single-point efficiency improvement; enterprises at the field innovation level begin to integrate multiple businesses, reduce manual intervention in the business, and achieve efficient collaboration of applications and data; enterprises at the comprehensive transformation level are characterized by a high degree of autonomy in all links in each field, opening up all factors and all-system processes, and realizing effective docking of supply and demand, coordinated resource scheduling, and coordinated quality control; enterprises at the innovation leading level are in the advanced stage of transformation. They will continuously iterate and optimize their own businesses, explore new business models, open up the ecological value network, and achieve sustainable and resilient development of the enterprise.

6. A method for quickly tracking and locating the digital development maturity of an enterprise as claimed in claim 1, characterized in that: The step S5 of establishing a correction model for the mature stage of enterprise digital transformation includes: The fuzzy comprehensive evaluation method is used to construct an enterprise digital transformation maturity evaluation model. The specific steps are as follows; (1) Establish the evaluation factor set U of fuzzy comprehensive evaluation U is a set of factors that affect the evaluation object, U = {u1,u2,...,u m }, where u i (i=1,2,3,…,m) is the influencing factor of the evaluation object; (2) Establish a comprehensive evaluation review set V The comment set is a set of various possible evaluation results made by the evaluator on the evaluation object, V = {v1, v2, ..., v n }, and divide it into five levels, v = {very good, good, average, poor, very poor}, corresponding to the proposed enterprise digital transformation maturity level, where v j (j=1,2,3,…,n) is the jth evaluation result, and each level is equivalent to a fuzzy subset; (3) Establish the fuzzy comprehensive evaluation matrix R First determine the membership of the evaluation object to the evaluation set; suppose the evaluation object is based on the i-th factor u in the factor set U i Evaluate the jth factor v in the evaluation set V j The membership degree is r ij , then the evaluation result of Ui can be expressed by fuzzy sets, R i ＝(r i1 ,r i2 ,…r im ) (9) After the single factor fuzzy evaluation, all factors are evaluated separately using the same principle to construct a total evaluation matrix, that is, each evaluated indicator determines the fuzzy relationship R from U to V, and the fuzzy comprehensive evaluation matrix can be obtained. R is called the fuzzy comprehensive evaluation matrix, which represents a fuzzy relationship between the factor set U and the evaluation set V; ij (i=1,2,…,m;j=1,2,…n) indicates the possibility of making the jth comment on the evaluation object based on the i-th factor; (4) Determine the weight vector of evaluation factors Assign a corresponding weight w to each factor U i (i=1,2,…,m) represents the weight of the i-th factor, w i Satisfy w i ≥0,∑w i =1, and then the weights are combined into a weight set W, and the weight coefficient is determined by the hierarchical analysis method; (5) Establishing a fuzzy comprehensive evaluation model Using appropriate synthesis operators to synthesize W and the fuzzy relationship matrix R, we can get the fuzzy evaluation result vector C of each evaluated object. The model of fuzzy comprehensive evaluation is: where c j (j=1,2,…n) is obtained by calculating the jth column of W and R, indicating the overall evaluation of the rated object on v j The degree of membership of the hierarchical fuzzy subset; according to the basic principle of maximum membership, the comment item with the largest proportion is the final evaluation conclusion, that is, the enterprise digital transformation maturity stage corresponding to each indicator; The first-level indicators with digital maturity at levels IV and V are in good development status and should continue according to their current plans; while the first-level indicators at levels I, II, and III are less mature, and corresponding suggestions are put forward to promote their development to the mature stage.

Images