Multi-index evaluation method and device for service team

Abstract

The invention provides a multi-index evaluation method and device for a business team, and relates to the technical field of big data, and the method comprises the steps: determining a to-be-evaluated business team; obtaining a numerical value corresponding to each evaluation index in a plurality of evaluation indexes of the to-be-evaluated business team; determining an objective weight corresponding to each evaluation index and a subjective weight corresponding to each evaluation index; based on the objective weight and the subjective weight, determining a combined weight corresponding to each evaluation index; based on the combined weight and a KL divergence optimization method, determining an integrated weight corresponding to each evaluation index; the integrated weight is a weight obtained by carrying out constraint optimization on the combined weight through a KL divergence optimization method; and based on the value corresponding to each evaluation index and the integration weight corresponding to each evaluation index, determining a comprehensive score of the to-be-evaluated business team. Therefore, the performance of the business team can be evaluated more accurately, and powerful support is provided for subsequent decision making.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of big data, and particularly relates to a multi-index evaluation method and device for a business team. BACKGROUND

[0002] At present, the operator grid team evaluation mainly adopts performance index evaluation, and a weighted average calculation is performed on product sales, assessment achievement rate and user satisfaction to obtain a conclusion to evaluate the overall level of a grid. The current scheme has the following problems: the team evaluation dimension is less considered, the setting of the weight lacks scientific basis through the weighted calculation, and the obtained conclusion has strong subjectivity.

[0003] Specifically, the existing operator grid team evaluation scheme has the following disadvantages:

[0004] 1. Simple and single rule: the existing technical scheme mainly depends on the weighted calculation of single indexes such as performance index ranking, regional user satisfaction and assessment achievement rate. Although this method is simple and easy to implement, the rule is too simple, and complex marketing and strategy behavior may not be fully captured.

[0005] 2. Limited information dimension: only single index calculation is used, and other possible important factors such as business opportunity collection dimension, learning dimension, team composition dimension and work order dimension are ignored, which limits the depth and accuracy of analysis.

[0006] 3. Low efficiency and prone to error: the data processing and analysis link mainly depends on manual work, which is not only low in efficiency, but also prone to human error, affecting the accuracy of the final result.

[0007] 4. Lack of depth in result display: the result display mainly uses tables and simple charts, which is intuitive but lacks in-depth analysis and insight, and is difficult to directly support business decision-making.

[0008] In summary, the existing operator grid team evaluation scheme has the problems of single evaluation rule, limited data dimension, low efficiency of manual processing and simple result display, which leads to high one-sidedness of comprehensive evaluation, insufficient accuracy and difficulty in effectively supporting business decision-making. SUMMARY

[0009] The embodiment of the present application provides a multi-index evaluation method and device for a business team to solve the technical problem that the existing operator grid team evaluation scheme has high one-sidedness of comprehensive evaluation, insufficient accuracy and difficulty in effectively supporting business decision-making.

[0010] In order to solve the above technical problems, the present application is implemented as follows:

[0011] In a first aspect, the embodiment of the present application provides a multi-index evaluation method for a business team, which comprises:

[0012] determining a business team to be evaluated;

[0013] obtaining a respective value of each evaluation index of a plurality of evaluation indexes of the business team to be evaluated;

[0014] determining a respective objective weight of each evaluation index and a respective subjective weight of each evaluation index;

[0015] determining a respective combined weight of each evaluation index based on the objective weight and the subjective weight;

[0016] determining a respective integrated weight of each evaluation index based on a KL divergence optimization method and the combined weight; the integrated weight is a weight of the combined weight after constraint optimization by the KL divergence optimization method;

[0017] determining a comprehensive score of the business team to be evaluated based on the respective value of each evaluation index and the respective integrated weight of each evaluation index.

[0018] Optionally, determining the respective objective weight of each evaluation index comprises:

[0019] determining the respective objective weight of each evaluation index based on an entropy weight method.

[0020] Optionally, determining the respective subjective weight of each evaluation index comprises:

[0021] determining the respective subjective weight of each evaluation index based on an analytic hierarchy process method.

[0022] Optionally, determining the respective combined weight of each evaluation index based on the objective weight and the subjective weight comprises:

[0023] determining the respective combined weight of each evaluation index based on a pre-set coefficient, the objective weight and the subjective weight, wherein the pre-set coefficient is used to represent a respective proportion of the objective weight and the subjective weight in the combination process, and the pre-set coefficient ∈ [0, 1].

[0024] Optionally, after determining the comprehensive score of the business team to be evaluated based on the respective value of each evaluation index and the respective integrated weight of each evaluation index, the method further comprises:

[0025] generating an analysis report based on each evaluation index, the respective integrated weight of each evaluation index, the respective combined weight of each evaluation index and the comprehensive score, wherein the analysis report is used to indicate strong points and weak points of the business team and provide corresponding suggestions based on the weak points.

[0026] Optionally, when the business team is an operator grid team, the plurality of evaluation indexes include at least two of the following: number of customer contacts, number of marketing activities, number of direct sales, order processing timeliness rate, number of orders per person, order completion rate, number of training sessions per person, examination completion rate, market share, and customer attrition rate.

[0027] In a second aspect, an embodiment of the present application provides a multi-index evaluation device for a business team, the device comprising:

[0028] a determination module configured to determine a business team to be evaluated;

[0029] an execution module configured to obtain a respective value of each evaluation index of a plurality of evaluation indexes of the business team to be evaluated;

[0030] determine a respective objective weight of each evaluation index and a respective subjective weight of each evaluation index;

[0031] determine a respective combined weight of each evaluation index based on the objective weight and the subjective weight;

[0032] determine a respective integrated weight of each evaluation index based on a KL divergence optimization method and the combined weight; the integrated weight is a weight obtained by constraint optimization of the combined weight through the KL divergence optimization method;

[0033] determine a comprehensive score of the business team to be evaluated based on the respective value of each evaluation index and the respective integrated weight of each evaluation index.

[0034] In a third aspect, an embodiment of the present application provides a network device, comprising a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the multi-index evaluation method for a business team according to the first aspect.

[0035] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program, when executed by a processor, implements the steps of the multi-index evaluation method for a business team according to the first aspect.

[0036] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising computer instructions, wherein the computer instructions, when executed by a processor, implement the steps of the multi-index evaluation method for a business team according to the first aspect.

[0037] In the embodiments of the present application, firstly, the basic data is formed by acquiring the specific values of the to-be-evaluated business team on each evaluation index. Then, the objective weight (calculated based on the characteristics of the data itself) and the subjective weight (set based on the experience or rules of experts) of each evaluation index are determined respectively, and the two are combined to generate a combined weight. Subsequently, the KL divergence optimization method is introduced to constrain the optimization of the combined weight, and a more reasonable integrated weight is obtained. This process ensures that the weight distribution conforms to both the objective data rule and the subjective judgment through mathematical optimization. Finally, the values of each evaluation index and its corresponding integrated weight are weighted and calculated to output the comprehensive score of the to-be-evaluated business team. Thus, this method adjusts the weight distribution by fusing the objective and subjective weights and using the optimization algorithm, so that the scoring result can reflect both the actual data characteristics and the field experience, and finally obtain a more accurate and balanced comprehensive evaluation effect. BRIEF DESCRIPTION OF DRAWINGS

[0038] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are for purposes of illustration only and are not intended to limit the present application. The same reference numbers in different drawings identify the same components throughout the text. In the drawings:

[0039] Figure 1 A flowchart of a multi-index evaluation method of a business team provided by an embodiment of the present application;

[0040] Figure 2 A flowchart of a multi-index evaluation method of a business team provided by an embodiment of the present application;

[0041] Figure 3 A schematic diagram of a team evaluation dimension of a business team provided by an embodiment of the present application;

[0042] Figure 4 A structural block diagram of a multi-index evaluation device of a business team provided by an embodiment of the present application;

[0043] Figure 5 A structural block diagram of a network device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0045] Figure 1A multi-index evaluation method of a business team is shown, and the method is provided by an embodiment of the application. Figure 1 As shown in the method comprises:

[0046] Step S101, determining a business team to be evaluated;

[0047] Step S102, obtaining a respective value of each evaluation index of a plurality of evaluation indexes of the business team to be evaluated;

[0048] Step S103, determining a respective objective weight of each evaluation index and a respective subjective weight of each evaluation index;

[0049] Step S104, determining a respective combined weight of each evaluation index based on the objective weight and the subjective weight;

[0050] Step S105, determining a respective integrated weight of each evaluation index based on a KL divergence optimization method and the combined weight;

[0051] The integrated weight is a weight of the combined weight after constraint optimization by the KL divergence optimization method;

[0052] Step S106, determining a comprehensive score of the business team to be evaluated based on the respective value of each evaluation index and the respective integrated weight of each evaluation index.

[0053] It should be noted that first, the specific values of the business team to be evaluated on each evaluation index are obtained to form basic data. Then, the objective weight (calculated based on the characteristics of the data itself) and the subjective weight (set based on expert experience or rules) of each evaluation index are determined respectively, and the two are combined to generate the combined weight. Subsequently, the KL divergence optimization method is introduced to constrain and optimize the combined weight, and a more reasonable integrated weight is obtained. This process ensures that the weight distribution conforms to both objective data rules and subjective judgments through mathematical optimization. Finally, the values of each evaluation index and its corresponding integrated weight are weighted and calculated to output the comprehensive score of the business team to be evaluated. Thus, the method combines objective and subjective weights and uses an optimization algorithm to adjust the weight distribution, so that the scoring result can reflect both actual data characteristics and field experience, and ultimately obtain a more accurate and balanced comprehensive evaluation effect.

[0054] In a possible implementation, when the business team is an operator grid team, the plurality of evaluation indexes include at least two of the following: the number of customers contacted, the number of marketing activities, the number of direct sales, the on-time rate of work order processing, the number of work orders per person, the work order completion rate, the number of training sessions per person, the examination completion rate, the market share, and the stock customer attrition rate.

[0055] In a possible implementation, determining the respective objective weight corresponding to each evaluation index comprises: determining the respective objective weight corresponding to each evaluation index based on an entropy weight method.

[0056] It should be noted that, first, based on the actual values of each index of the business team to be evaluated, the mathematical principle of the entropy weight method is used for calculation. The core of the entropy weight method is to quantify the information amount of the index data by analyzing the dispersion degree of the index data. The higher the dispersion degree of the index, the greater the amount of information it carries, and the higher the corresponding objective weight. In specific implementation, the original data needs to be standardized (for example, the range method is used to eliminate the dimensional difference), and the values of each index are converted to the range of 0-1; then the information entropy value of each index is calculated. The smaller the information entropy value, the higher the dispersion degree of the index data, and the objective weight is determined by the normalized entropy value calculation result. This method completely depends on the distribution characteristics of the data itself, avoids human intervention, and can objectively reflect the importance difference of the index at the data level. The effect is to automatically allocate weights by quantifying data volatility, which is particularly suitable for scenarios where the differences between indexes are significant and key volatile indexes need to be highlighted.

[0057] In a possible implementation, determining the respective subjective weight corresponding to each evaluation index comprises: determining the respective subjective weight corresponding to each evaluation index based on an analytic hierarchy process.

[0058] It should be noted that the core of this possible implementation is to compare and quantify the importance between indexes by expert experience or business rules, and to build a judgment matrix based on this, and to calculate the subjective weight reflecting the business focus by mathematical method. This process ensures that the subjective weight reflects both business experience and mathematical logic self-consistency. The effect is to convert the cognition of business experts on the importance of indexes into quantifiable weights, effectively making up for the lack of business background knowledge of data-driven objective weights.

[0059] In a possible implementation, determining the respective combined weight corresponding to each evaluation index based on the objective weight and the subjective weight comprises: determining the respective combined weight corresponding to each evaluation index based on a pre-set coefficient, the objective weight and the subjective weight, wherein the pre-set coefficient is used to represent the respective proportion of the objective weight and the subjective weight in the combination process, and the pre-set coefficient is in [0, 1].

[0060] It should be noted that the possible implementation describes a specific method of how to fuse the objective weight and the subjective weight by the preset coefficient to determine the combined weight. The core is to set a coefficient with a value range of [0, 1] for controlling the contribution ratio of the objective weight and the subjective weight in the combination process. In specific implementation, firstly, independent weight values of each index are obtained through the entropy weight method (objective weight) and the analytic hierarchy process (subjective weight) respectively; then, linear weighting calculation is performed on each index according to the formula “combined weight = preset coefficient x objective weight + (1-preset coefficient) x subjective weight”, and finally the combined weight of the integrated subjective and objective information is generated. For example, if the preset coefficient is 0.7, the objective weight accounts for 70% of the combined weight, and the subjective weight accounts for 30%. This method can flexibly adapt to the needs of different business scenarios (such as increasing the coefficient for data-driven scenarios and reducing the coefficient for experience-dominated scenarios) by adjusting the preset coefficient, retaining the data characteristics of the objective weight and integrating the business cognition of the subjective weight. The specific value of the preset coefficient depends on artificial experience or historical verification, and its effect lies in balancing data rules and business logic through the controllable weight fusion mechanism, providing a basis for subsequent optimization of integrated weights, and the rationality of the combined weight highly depends on the scientific setting of the preset coefficient. A reasonable coefficient can be selected according to experience in specific implementation.

[0061] In a possible implementation, after determining the comprehensive score of the business team to be evaluated based on the respective numerical values of each evaluation index and the respective integrated weights of each evaluation index, the method further includes:

[0062] Based on each evaluation index, the respective integrated weight of each evaluation index, the respective combined weight of each evaluation index, and the comprehensive score, an analysis report is generated, wherein the analysis report is used to indicate the strong points and weak points of the business team, and provide corresponding suggestions based on the weak points.

[0063] It should be noted that the possible implementation describes the complete process of generating a business team analysis report after deriving the comprehensive score. The core is to construct a multi-dimensional analysis framework based on the numerical value of each evaluation index, integrated weight, combined weight and final comprehensive score, and to identify business performance characteristics by comparing the matching degree of index value and weight allocation. For example, in specific implementation, first, the index values are weighted according to the integrated weight to obtain the comprehensive score; then, the order of the index values in the team is analyzed horizontally, and the historical data or industry benchmark value is compared vertically, combined with the size of the integrated weight (reflecting the global importance after optimization) to determine the strong points (index with high value and high weight) and weak points (index with low value but high weight, or mismatched value and weight); at the same time, by comparing the difference between the combined weight and the integrated weight (such as the weight deviation caused by KL divergence optimization), the deviation between business expectations (combined weight) and actual optimization results (integrated weight) is identified, and the causes of weak points are further explained; finally, based on the business scenarios of weak point indicators (such as opportunity collection dimension, learning dimension, etc.), targeted suggestions (such as optimizing work order processing flow, strengthening team composition training, etc.) are put forward. The effect is to convert the score results into actionable business insights, helping the team to clarify the improvement direction.

[0064] In summary, the method shown in the embodiments of the present application can perform automatic data processing and analysis: through automatic means for data processing and analysis, human intervention is reduced, processing efficiency is improved, and the risk of human error is reduced; a KL divergence group evaluation model can be constructed: by constructing a group evaluation model, the grid team can be more comprehensively evaluated, and the evaluation dimensions including satisfaction, output, discipline, market service, work order processing, team composition, etc. can be fully evaluated; targeted guidance can be effectively provided according to team scores: not only the comprehensive ability of the operator grid team can be evaluated according to the algorithm, but also targeted suggestions can be put forward according to the scores, for example, when the team satisfaction ratio is low in the weight matrix, the grid can be suggested to strengthen the team service quality.

[0065] In summary, the method shown in the embodiments of the present application aims to more comprehensively score the grid team by constructing a group evaluation model, improving the richness and accuracy of feature extraction, automatic data processing and analysis, and deepening the results display and application, etc., to provide more comprehensive and in-depth insights and support for the management evaluation of business teams (such as telecom operator grassroots teams).

[0066] Now, the multi-index evaluation method of the business team shown in the embodiments of the present application will be described from the perspective of specific application scenarios.

[0067] The method shown in the embodiments of the present application can be divided into three parts: model construction, scheme output and example analysis (such as Figure 2(As shown in the figure), during model construction, the objective weights of each evaluation indicator are calculated using the entropy weight method, and the subjective weights are calculated using the Analytic Hierarchy Process (AHP). Adjustable coefficients are added and weights are combined. Using the KL divergence criterion, an optimization model is established and integrated weights are calculated. The operator grid team evaluation value is calculated based on the integrated weights. The solution output uses the entropy weight method, the AHP method, and the KL divergence criterion to generate a weight table, which includes both combined and integrated weights. In case analysis, the combined and integrated weights corresponding to the comprehensive capabilities of a specific operator grid team are calculated, and the reasons and significance of the weight changes are analyzed.

[0068] 1. Model construction

[0069] This solution uses the operator grid team as the business team. The model uses information such as the operator grid team's personnel qualifications, market effectiveness, marketing expansion performance, and work output. By utilizing the KL divergence criterion, a grid evaluation algorithm model is constructed to evaluate the comprehensive capabilities of the operator grid team.

[0070] The specific steps are as follows:

[0071] Step 1: Data processing and analysis

[0072] (1) Data input: An evaluation system based on the comprehensive capabilities of the operator's grid team (e.g. Figure 3 As shown in the figure, this plan conducts data analysis and modeling from the dimensions of grid team (direct and follow-up sales) personnel qualifications, market supervision efficiency, marketing development performance, work output, discipline, service evaluation satisfaction, etc.

[0073] (2) Sample selection: Eliminate samples with data that are too discrete and do not obey the distribution law, and assign reasonable data after making sample estimates based on the distribution of other data.

[0074] (3) Data processing: Normalize the data to eliminate the dimensional effect.

[0075] Step 2: Build a mathematical model

[0076] The grid team comprehensive capability evaluation problem is divided into three evaluation elements: (A, C, X). Among them, A={a i |i=1,2,…,m} is the grid team comprehensive capability set; C={c j |j=1,2,…,n} is the evaluation indicator set. The objective matrix X is constructed based on the objective conditions of the grid team's actual business, such as the number of customers contacted, the number of marketing activities, the number of direct sales, the timeliness of work order processing, the number of work orders per capita, the work order completion rate, the number of training sessions per capita, the exam completion rate, the market share, and the churn rate of existing customer service staff. That is:

[0077]

[0078] wherein x mn represents the value of the jth evaluation index of the ith grid operator team, the multi-index group evaluation refers to that the number of indexes cannot be less than 2, and the group evaluation model needs to consider the differences in actual situations and the influence degree on the group decision. Under the condition of group decision, the weight of the index will change, and the change can be described by a mathematical model, i.e. the comprehensive ability score value of the ith grid team can be calculated by using the weighted sum formula:

[0079]

[0080] wherein ω j is the integrated weight of the index c j , and satisfies 0≤ω j ≤1, j=1, 2, …, n. In group evaluation, the integrated weight is related to the objective weight and the subjective weight of the evaluation index. Therefore, how to solve the integrated weight ω j using the known decision information is particularly important.

[0081] Step three: determining the objective weight of the evaluation index

[0082] In order to more objectively evaluate the grid team of the operator, the objective weight of the evaluation index is first determined by using the entropy weight method. The entropy weight method is an objective weighting method based on the principle of information entropy, and is widely used in multi-index comprehensive evaluation problems. It determines the weight of the index by calculating the information entropy (uncertainty of the index) of each index, so as to reduce the interference of subjective factors and improve the objectivity of the evaluation result. The basic steps of the entropy weight method are as follows: according to the principle of information entropy, the probability of the jth index appearing is:

[0083]

[0084] wherein x ij represents the original value of the ith evaluated person under the jth evaluation index, r ij represents the standardized value of the ith evaluated person under the jth evaluation index, and the weight is inversely proportional to the information entropy, i.e. the smaller the information entropy, the greater the weight. That is, the greater the uncertainty of the index, the lower the weight it occupies. Therefore, the entropy of the jth index is:

[0085]

[0086] Define the information utility value θ = 1-E j , the smaller the entropy value, the greater the information utility value, and the more the amount of information. After normalizing the information utility value, the entropy weight of each index can be obtained, i.e. the objective weight of the evaluation index, let it be β j , and it can be known that:

[0087]

[0088] The entropy weight method determines the weight of each evaluation index by objectively evaluating the distribution of the evaluation index. The overall performance of the evaluation index is not stable, and the weight is less. The entropy weight method provides a quantitative method to determine the weight, avoids the influence of subjective judgment, and makes the comprehensive evaluation more scientific and reasonable. However, the entropy weight method also has certain limitations, such as sensitivity to extreme values, etc. Therefore, the subjective weight is used to adjust the sensitivity in the method shown in the embodiments of the present application.

[0089] Step four: determining the subjective weight of the evaluation index

[0090] Since it is an evaluation algorithm, according to the actual situation, it cannot be simply objectively analyzed, so the method shown in the embodiments of the present application will also perform subjective analysis, and the subjective weight of the evaluation index is calculated by using the analytic hierarchy process. The analytic hierarchy process is a quantitative method that can decompose relatively complex safety evaluation problems into hierarchical structures, sort them according to importance, and improve the quantification and clarity of the evaluation. Therefore, the method shown in the embodiments of the present application uses the analytic hierarchy process to construct the subjective weight of the evaluation index.

[0091] Hierarchical evaluation problem, according to the nature of the problem and the total target to be achieved, the problem is decomposed into different constituent factors, i.e. evaluation index, according to the mutual relationship and membership relationship between the evaluation indexes, the evaluation indexes are aggregated and combined according to different levels, forming a multi-level analysis structure model, and finally concluding the relative importance degree of the weight or the relative order of the problem of the lowest layer (secondary evaluation index) relative to the highest layer (evaluation total score), i.e. the evaluation of the importance of the evaluation indexes.

[0092] 1. Construct the judgment matrix A of the criterion layer to the target layer, as shown in Table 1.

[0093] Table 1

[0094]

[0095] 2. According to the construction criteria of Table 1, construct the judgment matrix B of the secondary evaluation index to the primary evaluation index.

[0096] 3. Calculate the weight of the matrix, specifically: for the judgment matrix A, calculate the characteristic root and characteristic vector of ω that satisfies Aω = λ max ω, and the characteristic weight ω of this level can be calculated. (2)

[0097] 4. Calculate the weight of the judgment matrix B: normalize each column vector of the matrix B to obtain the matrix, and then sum the rows to obtain the matrix The matrix​ Normalization, Calculate The weight of each judgment matrix B is obtained as the approximation of the largest eigenvalue, which is combined together to form the weight ω of the scheme layer (3) .

[0098] 5. Multiply the weight matrix ω obtained before each layer (2) and ω (3) to obtain the final subjective weight of each index

[0099] The Analytic Hierarchy Process (AHP) is a method of subjective evaluation to determine the importance of evaluation indexes and obtain the final weight. That is, the more attention paid to the evaluation index in expert evaluation, the higher the weight. AHP provides a systematic method to solve multi-criteria decision-making problems, which not only considers qualitative judgments but also quantitatively analyzes through mathematical methods, making the decision-making process more scientific and reasonable. However, AHP also has some limitations, such as the accuracy of expert judgment and the consistency of the standard, therefore, a probability formula, Kullback-Leibler divergence, is introduced.

[0100] Step five: Determine the integrated weight of group evaluation

[0101] Not only the real situation is considered, but also subjective analysis is conducted to calculate the objective weight and subjective weight, and then the calculated objective and subjective weights are combined to obtain the combined weight. The integrated weight is calculated using Kullback-Leibler divergence. KL divergence is a measure of the difference between two probability distributions, mainly used to compare the similarity of two probability distributions.

[0102] Suppose p(x) = (p1, p2, …, p n ) and q(x) = (q1, q2, …, q n ) are two probability distributions of the same random event X, the information entropy of p(x) distribution is:

[0103]

[0104] The information entropy of q(x) distribution is:

[0105]

[0106] Then the KL divergence is:

[0107]

[0108] The smaller the KL divergence is, the closer the two probability distributions p(x) and q(x) are. In order to evaluate the comprehensive ability of the grid team objectively, the subjective weight and the objective weight are combined to form a new combined weight μ j The integrated weight is calculated by using the new weight μ j and the KL divergence criterion. From the practical point of view, the evaluation system should be in a relatively stable state, that is, when the objective conditions do not change greatly, the evaluation results should be in a relatively stable state. Therefore, the concept of KL divergence is introduced to minimize the evaluation change. First, the combined weight μ j of the objective weight and the subjective weight is calculated:

[0109]

[0110] wherein α is an adjustment coefficient, satisfying 0 < α < 1, and the specific value will be adjusted according to the actual results. By using the KL divergence criterion, the following optimization model is established, and the purpose is to minimize the evaluation change and make the evaluation tend to be stable:

[0111] wherein

[0112] In order to solve the model, the Lagrange function is constructed:

[0113]

[0114] wherein (λ-1) is the Lagrange multiplier, and the partial derivatives of ω j and λ are taken to obtain a system of equations.

[0115]

[0116] The integrated weight is finally calculated:

[0117]

[0118] The integrated weight obtained by the optimization model is substituted into to calculate the score value of the comprehensive ability of the grid team.

[0119] The KL divergence is combined with the subjective and objective weights to make the evaluation results tend to be stable. This method organically combines the subjective and objective weights, avoids the instability caused by the extreme value influence of the objective entropy weight method and the human bias of the subjective analytic hierarchy process, and formulates a stable, unified, scientific and efficient evaluation system to provide support for evaluating the grid team construction.

[0120] 2. Scheme output

[0121] According to the evaluation index of the operator grid team, an evaluation score matrix is constructed, the objective weight of each evaluation index is determined by using the entropy weight method, the judgment matrix is constructed by using the analytic hierarchy process, and the subjective weight of each evaluation index is calculated; the objective weight and the subjective weight are combined, the integrated weight is calculated and solved by using the KL divergence criterion, and the weight table is obtained, including the evaluation index and the corresponding combined weight and integrated weight. The integrated weight in the weight table is used to score the comprehensive ability of the operator grid team, and the teams are sorted according to the scoring results, so as to compare the comprehensive abilities of different teams. The grid manager can identify the strengths and weaknesses of the team according to the evaluation index and the weight in the weight table, and provide a basis for subsequent team rectification and ability improvement.

[0122] 3. Example analysis

[0123] Compared with the traditional algorithm, the method shown in the embodiments of the application can improve the calculation efficiency, reduce the human factors, improve the objectivity of evaluation, and simplify the calculation process, which is easy to implement and effectively optimizes the group evaluation decision.

[0124] Example analysis scheme

[0125] The example analysis scheme has the following advantages: the scheme studies the multi-index group evaluation problem, and in view of the problems of complex mathematical model, large amount of calculation and repeated adjustment of the subjective and objective weighting method, a weight integration method based on information entropy theory is proposed, which can be well applied to the comprehensive ability evaluation of the operator grid team. In actual application, the evaluation of the comprehensive ability of the operator grid team often faces the challenges of non-uniform evaluation system and complex index, which not only increases the difficulty of index standardization processing, but also affects the efficiency and accuracy of evaluation. In order to overcome these problems, the scheme proposes an improved evaluation method, which optimizes the complex evaluation index through a unified evaluation standard and process, and uses an efficient calculation method such as entropy weight method, analytic hierarchy process and KL divergence criterion to simplify the weight calculation process and improve the evaluation efficiency. This method not only improves the objectivity and accuracy of evaluation, but also reduces the dependence on subjective judgment, ensures the consistency and comparability of the evaluation of different operator grid teams, and provides a more efficient and simple solution for the comprehensive ability evaluation of the operator grid team.

[0126] The example analysis scheme evaluates the comprehensive ability of the operator grid team from the team satisfaction, team output, team composition, work order processing, learning / discipline and market service, and the evaluation index system is shown in Table 2.

[0127] Table 2

[0128]

[0129]

[0130] According to the data of the above dimensions, an expert scoring matrix for evaluating the comprehensive ability of the operator grid team is constructed, and a judgment matrix is also constructed. The evaluation index set is normalized, consistent, and dimensionless. Using the expert scoring matrix, the objective weight of each evaluation index is calculated based on the entropy weight method, the subjective weight of each index is calculated based on the analytic hierarchy process using the judgment matrix, the KL divergence criterion is used to establish an optimization model, and the integrated weight is solved. The integrated weight is brought into the formula to obtain the comprehensive ability evaluation value of the operator grid team.

[0131] Example analysis results

[0132] According to the above design steps, the combination weight and integrated weight are calculated and solved. As shown in Table 3.

[0133] Table 3

[0134] Numbering Combination weight Integrated weight A1 0.067 0.140 A2 0.084 0.079 B1 0.091 0.088 B2 0.075 0.070 B3 0.060 0.060 B4 0.057 0.054 C1 0.112 0.065 C2 0.042 0.043 C3 0.038 0.038 D1 0.023 0.100 D2 0.152 0.028 D3 0.087 0.075 E1 0.099 0.090 E2 0.054 0.052 F1 0.066 0.066 F2 0.079 0.082

[0135] According to the data in Table 3, we can see that the existing index system gives higher attention to the number of direct sales (C1) and the number of work orders per capita (D2) in the calculation of the combination weight. This may reflect a traditional concept that performance and efficiency are measured by quantity. However, in the evaluation of the expert group, the calculation of the integrated weight shows that the experts pay more attention to the satisfaction of installation and maintenance (A1) and the work order processing rate (D1). This change may be due to the introduction of the combination mechanism of subjective and objective weights, which encourages the evaluation results to focus more on service quality and response speed, rather than just quantity. The introduction of the weight combination mechanism makes the evaluation process pay more attention to practical application value and service quality. This mechanism emphasizes the evaluation of service satisfaction, such as the satisfaction of installation and maintenance (A1), which is directly related to the experience of consumers and the trust of service providers. At the same time, more attention is paid to the timeliness of work order processing (D1), because it reflects the response speed of service providers to customer demand and the efficiency of problem solving.

[0136] This change shows that the weight combination mechanism tends to give more weight to indicators that can provide high-quality services and fast response. This not only helps to improve customer satisfaction, but also helps to gain an advantage in a competitive market. By paying attention to the satisfaction of installation and maintenance and the timeliness of work order processing, service providers can better meet customer demand, improve brand image, and establish stable customer relationships in the long term.

[0137] In summary, the changes in the index system revealed in Table 2, as well as the shift in the group evaluation, all point to a common goal: while pursuing quantity, more attention is paid to service quality and customer experience, in order to achieve long-term customer satisfaction and business success. This quality-oriented evaluation system helps to promote continuous innovation and improvement of service providers to meet the growing needs of the market and customers.

[0138] In addition, this review method helps the grid manager to well understand the strengths and weaknesses of the team, so as to facilitate future improvement and progress. Through this way, not only a more objective and accurate comprehensive ability evaluation mechanism of grid team is provided, but also the high-quality balanced development of the grid team of the operator is promoted.

[0139] In summary, the algorithm provides an efficient and objective multi-index group evaluation method by combining information entropy theory and KL divergence, which is especially suitable for occasions requiring group decision-making such as comprehensive ability assessment.

[0140] The multi-index evaluation algorithm of the operator grid team based on KL divergence has the following three advantages:

[0141] 1. Reducing subjectivity: The algorithm determines the objective weight of the evaluation index through the entropy weight method, determines the subjective weight of the evaluation index through the analytic hierarchy process, obtains the combined weight, and then determines the integrated weight of the group evaluation based on the KL divergence. This process reduces the intervention of human factors. The algorithm does not need to completely rely on the subjective experience of experts to adjust parameters, so it can more objectively reflect the importance of evaluation indexes and improve the fairness and accuracy of evaluation results.

[0142] 2. Having rectification guidance: The district company can identify the strengths and weaknesses of the team according to the evaluation indexes and their weights in the weight table, and provide a basis for subsequent team rectification and capacity improvement. As can be seen from the example, the direct sales quantity and the per capita work order quantity of the grid team are weak, and the subsequent correction, installation and maintenance satisfaction and work order processing timeliness are advantages, which should be maintained.

[0143] 3. Algorithm simplicity: The process of the algorithm is simple and easy to understand and implement. Through the combination of entropy weight method, analytic hierarchy process and KL divergence, the algorithm provides a direct and effective method to solve the multi-index group evaluation problem, without complex mathematical models and tedious calculation processes. This makes the algorithm not only theoretically attractive, but also highly operable in practical applications.

[0144] Figure 4 A multi-index evaluation device of a business team according to an embodiment of the present application is shown, as shown in Figure 4 The device 40 includes:

[0145] The determining module 401 is configured to determine a service team to be evaluated.

[0146] The executing module 402 is configured to obtain a respective value of each evaluation index of a plurality of evaluation indexes of the service team to be evaluated.

[0147] The respective objective weight and the respective subjective weight of each evaluation index are determined.

[0148] The respective combination weight of each evaluation index is determined based on the objective weight and the subjective weight.

[0149] The respective integrated weight of each evaluation index is determined based on the KL divergence optimization method and the combination weight. The integrated weight is the weight of the combination weight after constraint optimization by the KL divergence optimization method.

[0150] The comprehensive score of the service team to be evaluated is determined based on the respective value of each evaluation index and the respective integrated weight of each evaluation index.

[0151] In a possible implementation, the executing module 402 is further configured to determine the respective objective weight of each evaluation index based on the entropy weight method.

[0152] In a possible implementation, the executing module 402 is further configured to determine the respective subjective weight of each evaluation index based on the analytic hierarchy process.

[0153] In a possible implementation, the executing module 402 is further configured to determine the respective combination weight of each evaluation index based on a preset coefficient, the objective weight and the subjective weight. The preset coefficient is used to represent the respective proportion of the objective weight and the subjective weight in the combination process, and the preset coefficient is in the range of [0, 1].

[0154] In a possible implementation, the executing module 402 is further configured to generate an analysis report based on each evaluation index, the respective integrated weight of each evaluation index, the respective combination weight of each evaluation index and the comprehensive score after determining the comprehensive score of the service team to be evaluated based on the respective value of each evaluation index and the respective integrated weight of each evaluation index. The analysis report is used to indicate the strong points and the weak points of the service team, and provide corresponding suggestions based on the weak points.

[0155] In a possible implementation, when the service team is an operator grid team, the plurality of evaluation indexes include at least two of the following: the number of customers contacted, the number of marketing activities, the number of direct sales, the on-time rate of work order processing, the number of work orders per person, the work order completion rate, the number of training sessions per person, the examination completion rate, the market share, and the stock customer attrition rate.

[0156] In summary, in the existing technical solutions, the evaluation of the operator grid team mainly depends on performance indicators such as product sales, assessment achievement rate, and user satisfaction, etc. These indicators only reflect one side of the team performance. In contrast, the method provided by the embodiments of the present application has the following advantages:

[0157] Comprehensiveness of evaluation: multi-dimensional evaluation: not only performance indicators are considered, but also multiple dimensions such as team composition, discipline, and service are included, thereby providing a more comprehensive perspective to evaluate the overall level of the grid team; combination of subjective and objective weights: an innovative method is adopted in weight setting, i.e., combining subjective and objective factors to determine the weight of each evaluation indicator. This method allows combining expert experience and historical data to verify and adjust the subjective judgment of experts through data analysis, thereby improving the scientificity and objectivity of weight setting.

[0158] Provide in-depth team evaluation analysis and support for team differentiation adjustment decision: personalized evaluation model: considering the specific situation of different grid teams, the method shown in the embodiments of the present application allows customization of the evaluation model to adapt to the characteristics and needs of different teams; decision support system: by providing differentiated analysis results, the method shown in the embodiments of the present application can help management make more accurate differentiated adjustment strategies, such as personnel configuration optimization, training needs analysis, etc.

[0159] Adjustable parameter combination weight is used to compound multiple actual scenarios: in terms of weight setting, the technical advantages of the method shown in the embodiments of the present application include: dynamic weight adjustment: the method shown in the embodiments of the present application allows dynamic adjustment of the weight of each evaluation indicator according to actual conditions, making the evaluation result more flexible and able to adapt to changing business environment; scientific basis for weight setting: by using expert consultation or historical data mining, the method shown in the embodiments of the present application can provide a more scientific basis for weight setting, reducing subjectivity while considering actual conditions.

[0160] In summary, the method shown in the embodiments of the present application provides a comprehensive, in-depth, and flexible team evaluation framework, which not only improves the accuracy and objectivity of evaluation, but also enhances the adaptability to different team characteristics and the support ability for decision-making. These technical advantages make the method shown in the embodiments of the present application have significant advantages in the evaluation of operator grid teams.

[0161] In addition, constructing the KL divergence grid team group evaluation model has the following commercial values:

[0162] I. Optimize resource allocation and decision support

[0163] The construction of a KL divergence-based group evaluation model provides enterprise management with an unprecedented multi-dimensional and refined evaluation tool. By comprehensively covering key evaluation dimensions such as satisfaction, output, discipline, market service, work order processing, and team composition, it can more accurately identify the strengths and weaknesses of grid teams, providing data support for optimizing team building, thus scientifically allocating resources, optimizing team structure, and maximizing resource utilization efficiency. This model provides data-driven decision support for management, helping district companies develop more accurate strategic planning and operational strategies.

[0164] II. Enhancing Customer Satisfaction and Market Competitiveness

[0165] The model's in-depth assessment of the market service dimension can directly reflect the performance of grid teams in customer service, including user satisfaction, opportunity collection, and the number of customer touchpoints. By continuously optimizing these aspects, enterprises can significantly improve customer satisfaction, enhance customer loyalty, and thus enhance market competitiveness. In a highly competitive market environment, this ability to continuously optimize based on customer feedback is key to maintaining a leading position for enterprises.

[0166] III. Promoting Team Collaboration and Efficiency

[0167] By introducing team composition, ratio data, and learning and training data, the model not only evaluates the overall performance of the team but also deeply analyzes the collaboration efficiency and growth potential within the team. This helps county companies identify the workload and efficiency of grid members, identify bottlenecks in team collaboration, and conduct detailed assessments of task execution. By precisely adjusting the allocation of time, funds, and manpower, and strengthening training and other measures, it promotes internal circulation and collaboration within the team, thereby efficiently completing work tasks and improving overall efficiency.

[0168] IV. Strengthening Risk Management and Problem Early Warning

[0169] Work order processing is an important part of the model's evaluation, covering multiple key indicators such as the number of work orders, completion rate, saturation, and processing rate. By monitoring and analyzing these data in real time, enterprises can identify and address potential operational risks and service issues in a timely manner, effectively preventing large-scale customer complaints or business disruptions. This forward-looking risk management capability helps enterprises maintain operational stability and sustainability.

[0170] V. Driving Innovation and Continuous Improvement

[0171] The process of constructing the KL divergence group evaluation model itself is a data-driven innovative practice. As the model is continuously optimized and applied, more valuable experience and data assets related to grid team management will be accumulated. These data can not only be used for current evaluation and decision-making, but also serve as a source of future innovation, driving enterprises to continuously explore new management and service models, and achieving sustainable improvement and leap-forward development.

[0172] In summary, constructing the KL divergence group evaluation model has significant commercial value and can help achieve overall improvement in resource allocation, customer satisfaction, team collaboration, risk management, and innovation-driven aspects, laying a solid foundation for sustainable development of enterprises.

[0173] The embodiment of the present application provides a network device 50, such as Figure 5 As shown, the network device 50 includes a processor 501, a memory 502, and a program stored in the memory 502 and executable on the processor 501, which implements the steps of the multi-index evaluation method of a business team as shown in the above embodiment when executed by the processor 501.

[0174] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the multi-index evaluation method of a business team as shown in the above embodiment when executed by a processor, and can achieve the same technical effects. To avoid repetition, this will not be repeated here. The computer readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0175] The embodiment of the present application also provides a computer program product, which includes computer instructions, and the computer instructions implement the steps of the multi-index evaluation method of a business team as shown in the above embodiment when executed by a processor, and can achieve the same technical effects. To avoid repetition, this will not be repeated here.

[0176] It should be noted that in this document, the term "comprise", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or apparatus including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or apparatus. Without more limitations, the element defined by the statement "comprises a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0177] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned example methods can be realized by means of software and a necessary general hardware platform, and of course, can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product in essence or in the form of a part that contributes to the prior art. The computer software product is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk) and includes a plurality of instructions for causing a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in the various embodiments of the present application.

[0178] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative and not restrictive. Those skilled in the art can make many forms without departing from the scope of the present application under the inspiration of the present application, and all of them belong to the protection of the present application.

Claims

1. A multi-index evaluation method for a business team, characterized by: The method comprises: Identify the business teams to be evaluated; Obtaining a value corresponding to each of the multiple evaluation indicators of the business team to be evaluated; Determining the objective weight corresponding to each evaluation indicator and the subjective weight corresponding to each evaluation indicator; Determine the combined weight corresponding to each evaluation indicator based on the objective weight and the subjective weight; Determine the integrated weight corresponding to each evaluation indicator based on the relative entropy KL divergence optimization method and the combined weight; the integrated weight is the weight obtained by constrained optimization of the combined weight using the KL divergence optimization method; Based on the numerical value corresponding to each evaluation indicator and the integrated weight corresponding to each evaluation indicator, a comprehensive score of the business team to be evaluated is determined.

2. The method according to claim 1, characterized in that Determining the objective weight corresponding to each evaluation indicator includes: Based on the entropy weight method, the objective weight corresponding to each evaluation indicator is determined.

3. The method according to claim 1, characterized in that Determining the subjective weight corresponding to each evaluation indicator includes: Based on the hierarchical analysis method, the subjective weight corresponding to each evaluation indicator is determined.

4. The method according to claim 1, wherein Determining the combined weight corresponding to each evaluation indicator based on the objective weight and the subjective weight includes: Based on the pre-set coefficient, the objective weight and the subjective weight, the combination weight corresponding to each evaluation indicator is determined, wherein the pre-set coefficient is used to represent the respective proportions of the objective weight and the subjective weight in the combination process, and the pre-set coefficient ∈ [0,1].

5. The method according to claim 1, wherein After determining the comprehensive score of the business team to be evaluated based on the numerical value corresponding to each evaluation indicator and the integrated weight corresponding to each evaluation indicator, the method further includes: An analysis report is generated based on each evaluation indicator, the integrated weight corresponding to each evaluation indicator, the combined weight corresponding to each evaluation indicator, and the comprehensive score, wherein the analysis report is used to point out the strengths and weaknesses of the business team and provide corresponding suggestions based on the weaknesses.

6. The method according to any one of claims 1 to 5, characterized in that When the business team is an operator grid team, the multiple evaluation indicators include at least two of the following: number of customer contacts, number of marketing activities, number of direct sales, timeliness of work order processing, number of work orders per capita, work order completion rate, number of training sessions per capita, exam completion rate, market share, and existing customer service churn rate.

7. A multi-index evaluation device for a business team, characterized in that: The device comprises: Determination module, used to determine the business team to be evaluated; An execution module, configured to obtain a numerical value corresponding to each of the multiple evaluation indicators of the business team to be evaluated; Determining the objective weight corresponding to each evaluation indicator and the subjective weight corresponding to each evaluation indicator; Determine the combined weight corresponding to each evaluation indicator based on the objective weight and the subjective weight; Determine the integrated weight corresponding to each evaluation indicator based on the KL divergence optimization method and the combined weight; the integrated weight is the weight obtained by constrained optimization of the combined weight using the KL divergence optimization method; Based on the numerical value corresponding to each evaluation indicator and the integrated weight corresponding to each evaluation indicator, a comprehensive score of the business team to be evaluated is determined.

8. A network device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the steps of a multi-index evaluation method for a business team according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the multi-index evaluation method for a business team according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions, which, when executed by a processor, implement the steps of a multi-index evaluation method for a business team according to any one of claims 1 to 6.

Images