Index weighting method for service quality evaluation problem

By constructing a multivariate linear regression model and using the coefficient of variation to correct the weight, the problem of lack of scientificity and user heterogeneity considerations in the existing technology is solved, and the scientificity and accuracy of service quality evaluation is improved, and the improvement is achieved. For precise and reasonable combination weights.

CN119090338APending Publication Date: 2024-12-06SUZHOU CITY UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411112293.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The way of combining subjective and objective weights in the prior art lacks scientificity, and user heterogeneity is rarely considered when combining empowerment, resulting in insufficient accuracy and reliability of service quality evaluation results.

Method used

Through investigation, different users' perceived scores and importance ratios for service quality indicators are obtained, multiple linear regression models are constructed, subjective and objective weights are calculated, and the weights are corrected using the coefficient of variation to obtain more scientific and reasonable combination weights.

Benefits of technology

It improves the scientificity and accuracy of service quality evaluation, takes into account user heterogeneity, and obtains more accurate and reasonable combination weights, enhancing the reliability of evaluation results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090338A_ABST
    Figure CN119090338A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of service quality evaluation, in particular to an index weighting method for a service quality evaluation problem. The method comprises the following steps: calculating the subjective weight of each user to each index based on the importance degree ratio of each group of adjacent indexes given by each user; constructing a multiple linear regression model by taking the perception score of the user on the overall service quality as a dependent variable and taking the product of the subjective weight of each index and the perception score of the service quality of each index as an independent variable, calibrating to-be-estimated parameters of the model, and obtaining the objective weight of each index; and based on the objective weight of each index and the subjective weight of each user for each index, calculating a mean value and a variable coefficient of products of the subjective weight and the objective weight of each index, correcting the mean value by using the variable coefficient, and calculating a combined weight of each index. The invention provides a novel subjective and objective weight coupling method, the influence of the user heterogeneity on the index combination weight is considered, and a new thought is provided for comprehensively and accurately calculating the weight of each index.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of service quality evaluation, and in particular to an indicator weighting method for service quality evaluation problems. Background Art

[0002] Service quality evaluation is a process of measuring and evaluating the service level provided by service providers through a series of tools and methods. One of the key tasks in this process is to determine the weight of each evaluation indicator, which is one of the important factors affecting the accuracy of service quality evaluation results. The methods for determining indicator weights can be roughly divided into three categories: subjective weighting method, objective weighting method, and subjective and objective combined weighting method.

[0003] The subjective weighting method is to obtain the importance of each indicator based on the subjective judgment of decision makers and experts, and then calculate the weight of each indicator. For example, expert survey method (Delphi method), hierarchical analysis method, etc. It is suitable for situations where there is a lack of objective data on relevant indicators or it is difficult to quantify. It can make full use of the knowledge and experience of decision makers, experts, etc. to make decisions quickly, but it is easily affected by the knowledge structure and preferences of decision makers, experts, etc. The objective weighting method is to obtain the weight of each indicator based on the objective data related to each indicator using a scientific analysis method. For example, principal component analysis, linear regression, entropy weight method, etc. Compared with the subjective weighting method, it is more objective and scientific, but it requires reliable and sufficient data support, and the selection of analysis methods is also crucial. The subjective and objective combined weighting method determines the weight of the indicator by combining the subjective weighting method and the objective weighting method. It reduces the deviation that may be caused by a single weighting method, makes up for the limitations of a single weighting method, thereby obtaining a more comprehensive weighting result and improving the accuracy and reliability of the weight.

[0004] The combination strategy of subjective weights and objective weights in the combined weighting method is crucial. However, in existing studies, the method of combining subjective weights and objective weights mostly adopts a linear weighting method, and the weighting coefficients are often determined based on the subjective experience of decision makers, experts, etc., which lacks scientificity, and user heterogeneity is rarely considered when combining weights. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is to overcome the technical problems that the method of combining subjective weights and objective weights in the prior art lacks scientificity and rarely considers user heterogeneity when combining and weighting.

[0006] In order to solve the above technical problems, the present invention provides an indicator weighting method for service quality evaluation problem, comprising the following steps:

[0007] Through the survey, we obtain the perceived scores of n service quality indicators given by different users, the perceived scores of overall service quality given by different users, and the importance ratios of each group of adjacent indicators given by different users, where n is the number of indicators;

[0008] Based on the importance ratio of each group of adjacent indicators given by each user, the subjective weight of each user for each indicator is calculated;

[0009] The user's perceived score of the overall service quality is taken as the dependent variable, and the product of the subjective weight of each indicator and the perceived score of the service quality of each indicator is taken as the independent variable. A multivariate linear regression model is constructed, and the parameters to be estimated of the multivariate linear regression model are calibrated to obtain the objective weight of each indicator.

[0010] Based on the objective weight of each indicator and each user's subjective weight of each indicator, the mean and coefficient of variation of the product of the subjective weight and objective weight of each indicator are calculated;

[0011] The coefficient of variation of the product of the subjective weight and objective weight of each indicator is used to correct the mean of the product of the subjective weight and objective weight of each indicator, so as to obtain the corrected value of the mean of the product of the subjective weight and objective weight of each indicator; based on the corrected value of the mean of the product of the subjective weight and objective weight of each indicator, the combined weight of each indicator is calculated.

[0012] Preferably, the step of calculating each user's subjective weight for each indicator based on the importance ratio of each group of adjacent indicators given by each user comprises the following steps:

[0013] Taking each user's importance score for the nth indicator as the base, based on the importance ratio of each group of adjacent indicators given by each user, each user's importance score for the nth indicator is used to represent each user's importance score for the first n-1 indicators;

[0014] Based on each user's score on the importance of each indicator, the subjective weight of each user on each indicator is calculated.

[0015] Preferably, the importance of each user to the nth indicator is scored As a base, based on the importance ratio of each group of adjacent indicators given by each user Use each user's importance score for the nth indicator Represents each user's importance score for the first n-1 indicators The calculation formula is:

[0016]

[0017] Preferably, the importance of each indicator is scored based on each user Calculate each user's subjective weight for each indicator The calculation formula is:

[0018]

[0019] in, is the subjective weight of the i-th user on the j-th indicator, Score the importance of the jth indicator for the i-th user, is the importance ratio of the kth indicator given by the i-th user to the k+1th indicator, The importance score of the nth indicator by the ith user, where n is the number of indicators and m is the number of users.

[0020] Preferably, the user's perceived score of the overall service quality is used as the dependent variable, and the product of the subjective weight of each indicator and the perceived score of the service quality of each indicator is used as the independent variable to construct a multivariate linear regression model:

[0021]

[0022] Among them, Q i is the perceived score of the overall service quality by the i-th user, b and β j (j=1,2,…,n) is the estimated parameter of the multivariate linear regression model, b is the constant term, β j is the objective weight of the jth indicator, satisfying β j >0 and is the subjective weight of the i-th user on the j-th indicator, is the perceived rating of the service quality of the jth indicator by the i-th user, ε i is the error term, n is the number of indicators, and m is the number of users.

[0023] Preferably, the calibration of the estimated parameters of the multivariate linear regression model and obtaining the objective weight of each indicator actually requires solving:

[0024]

[0025]

[0026] The Sequential Least Squares Programming (SLSQP) method is used to solve the above constrained nonlinear minimization problem and obtain the estimated parameters b and β of the multivariate linear regression model. j ;

[0027] Among them, Q iis the perceived score of the overall service quality by the i-th user, b and β j (j=1,2,…,n) is the estimated parameter of the multivariate linear regression model, b is the constant term, β j is the objective weight of the jth indicator, satisfying β j >0 and is the subjective weight of the i-th user on the j-th indicator, is the perceived rating of the service quality of the jth indicator by the i-th user, ε i is the error term, n is the number of indicators, and m is the number of users.

[0028] Preferably, based on the objective weight of each indicator and the subjective weight of each user for each indicator, the mean and coefficient of variation of the product of the subjective weight and the objective weight of each indicator are calculated, and the calculation formula is:

[0029]

[0030] Among them, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, σ j is the standard deviation of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, is the subjective weight of the i-th user on the j-th indicator, β j is the objective weight of the jth indicator, n is the number of indicators, and m is the number of users.

[0031] Preferably, the coefficient of variation of the product of the subjective weight and objective weight of each indicator is used to correct the mean of the product of the subjective weight and objective weight of each indicator to obtain the corrected value of the mean of the product of the subjective weight and objective weight of each indicator, and the calculation formula is:

[0032]

[0033] Among them, μ j ′ is the corrected value of the mean of the product of the subjective weight and objective weight of the jth indicator, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, and n is the number of indicators.

[0034] Preferably, the combined weight of each indicator is calculated based on the correction value of the mean of the product of the subjective weight and the objective weight of each indicator, and the calculation formula is:

[0035]

[0036] Among them, w j is the combined weight value of the jth indicator, μ j ′ is the corrected value of the mean of the product of the subjective weight and objective weight of the jth indicator, and n is the number of indicators.

[0037] Preferably, the method further comprises: when obtaining, through investigation, perceived scores given by different users to n service quality indicators and perceived scores given by different users to the overall service quality, a Likert 5-level scoring method is used.

[0038] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0039] The indicator weighting method for the service quality evaluation problem described in the present invention calculates the subjective weight of each user for each indicator based on the importance ratio of each group of adjacent indicators given by each user; takes the user's perceived score of the overall service quality as the dependent variable, and the product of the subjective weight of each indicator and the perceived score of the service quality of each indicator as the independent variable, constructs a multivariate linear regression model, calibrates the parameters to be estimated of the multivariate linear regression model, and obtains the objective weight of each indicator. When solving the objective weight, the subjective weight is taken into account, which helps to calculate the objective weight of each indicator more scientifically, and provides a new method for coupling subjective and objective weights.

[0040] In addition, based on the objective weight of each indicator and the subjective weight of each user for each indicator, the mean and coefficient of variation of the product of the subjective weight and objective weight of each indicator are calculated, the mean is corrected using the coefficient of variation, and the combined weight of each indicator is calculated based on the corrected value, taking into account the impact of user heterogeneity on the combined weight. Considering user heterogeneity makes the combined weight of each indicator more scientific, reasonable and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0042] Figure 1 It is a flow chart of the steps of an indicator weighting method for service quality evaluation problem of the present invention. DETAILED DESCRIPTION

[0043] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0044] Reference Figure 1 As shown, the first embodiment of the present invention provides an indicator weighting method for a service quality evaluation problem, comprising the following steps:

[0045] Step S1: obtaining, through investigation, the perceived scores given by different users for n service quality indicators, the perceived scores given by different users for the overall service quality, and the importance ratios of each group of adjacent indicators given by different users, where n is the number of indicators;

[0046] In this embodiment, preferably, it also includes: obtaining the perceived score of the service quality of the jth indicator by the i-th user through a survey The i-th user's perceived rating of the overall service quality is Q i When (i=1,2,…,m), the Likert 5-level rating method is used, with scores ranging from 1 to 5. The larger the value, the better the service quality. n is the number of indicators, and m is the number of users. The Likert 5-level rating method is a simple and widely used rating method that users can easily understand and respond to quickly. The data generated by the Likert 5-level rating method is quantitative and can be directly used for mathematical and statistical analysis, making the analysis of each service quality indicator and the overall score more objective and scientific. The importance ratio of the kth indicator given by the i-th user to the k+1th indicator is obtained through the survey. When , the indicators are randomly arranged in sequence, allowing users to compare the relative importance of two adjacent indicators and give the importance ratio of the two adjacent indicators. The value range of the ratio is {1 / 5, 1 / 4, 1 / 3, 1 / 2, 1, 2, 3, 4, 5}.

[0047] Step S2: Calculate each user's subjective weight for each indicator based on the importance ratio of each group of adjacent indicators given by each user;

[0048] In this embodiment, based on the importance ratios of each group of adjacent indicators given by each user, the subjective weight of each user for each indicator is calculated. The importance score of each user for the nth indicator is first taken as the base, and based on the importance ratios of each group of adjacent indicators given by each user, the importance score of each user for the nth indicator is used to represent the importance score of each user for the first n-1 indicators. Then, based on the importance score of each user for each indicator, the subjective weight of each user for each indicator is calculated.

[0049] Specifically, each user is scored based on the importance of the nth indicator. As a base, based on the importance ratio of each group of adjacent indicators given by each user Use each user's importance score for the nth indicator Represents each user's importance score for the first n-1 indicators The calculation formula is:

[0050]

[0051] Then based on each user's rating of the importance of each indicator Calculate each user's subjective weight for each indicator The calculation formula is:

[0052]

[0053] in, is the subjective weight of the i-th user on the j-th indicator, Score the importance of the jth indicator for the i-th user, is the importance ratio of the kth indicator given by the i-th user to the k+1th indicator, The importance score of the nth indicator by the ith user, where n is the number of indicators and m is the number of users.

[0054] By deriving the importance score based on the ratio, the consistency and coherence of the score can be ensured, and the ratio-based derivation scoring method can better adapt to the changes in the relative importance of different indicators and individual differences among users.

[0055] Step S3: Taking the user's perceived score of the overall service quality as the dependent variable, and the product of the subjective weight of each indicator and the perceived score of the service quality of each indicator as the independent variable, construct a multiple linear regression model, calibrate the parameters to be estimated of the multiple linear regression model, and obtain the objective weight of each indicator;

[0056] Specifically, the multivariate linear regression model is constructed as follows:

[0057]

[0058] Among them, Q i is the perceived score of the overall service quality by the i-th user, b and β j (j=1,2,…,n) is the estimated parameter of the multivariate linear regression model, b is the constant term, β j is the objective weight of the jth indicator, satisfying β j >0 and is the subjective weight of the i-th user on the j-th indicator, is the perceived rating of the service quality of the jth indicator by the i-th user, ε i is the error term, n is the number of indicators, and m is the number of users.

[0059] Specifically, the calibration of the estimated parameters of the multivariate linear regression model and obtaining the objective weights of each indicator actually requires solving:

[0060]

[0061] The Sequential Least Squares Programming (SLSQP) method is used to solve the above constrained nonlinear minimization problem and obtain the estimated parameters b and β of the multivariate linear regression model. j ;

[0062] Among them, Q i is the perceived score of the overall service quality by the i-th user, b and β j (j=1,2,…,n) is the estimated parameter of the multivariate linear regression model, b is the constant term, β j is the objective weight of the jth indicator, satisfying β j >0 and is the subjective weight of the i-th user on the j-th indicator, is the perceived rating of the service quality of the jth indicator by the i-th user, ε i is the error term, n is the number of indicators, and m is the number of users.

[0063] When solving the objective weight, the subjective weight is considered, and the product of the user's subjective weight of each indicator and the perceived score of the service quality of each indicator is used as the independent variable. A multivariate linear regression model is constructed to solve the objective weight, which helps to calculate the objective weight of each indicator more scientifically and provides a new method for coupling subjective and objective weights. The sequential least squares programming method is used to solve the above-mentioned constrained nonlinear minimization problem. The sequential least squares programming method can quickly and reliably find the global or local optimal solution by iteratively optimizing the objective function while satisfying the constraints. The sequential least squares programming method can effectively handle these complex constraints, ensure the legitimacy and practical feasibility of the parameters in the optimization process, and make the regression model fit better.

[0064] Step S4: Based on the objective weight of each indicator and each user's subjective weight of each indicator, the mean and coefficient of variation of the product of the subjective weight and the objective weight of each indicator are calculated;

[0065] The calculation formula for calculating the mean and coefficient of variation of the product of the subjective weight and objective weight of each indicator is:

[0066]

[0067] Among them, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, σ j is the standard deviation of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, is the subjective weight of the i-th user on the j-th indicator, β j is the objective weight of the jth indicator, n is the number of indicators, and m is the number of users.

[0068] Step S5: using the coefficient of variation of the product of the subjective weight and objective weight of each indicator to correct the mean of the product of the subjective weight and objective weight of each indicator, to obtain a corrected value of the mean of the product of the subjective weight and objective weight of each indicator;

[0069] The coefficient of variation of the product of the subjective weight and objective weight of each indicator is used to correct the mean of the product of the subjective weight and objective weight of each indicator. The calculation formula is:

[0070]

[0071] Among them, μ j ′ is the corrected value of the mean of the product of the subjective weight and objective weight of the jth indicator, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, and n is the number of indicators.

[0072] Step S6: Calculate the combined weight of each indicator based on the corrected value of the product of the subjective weight and the objective weight of each indicator.

[0073] The calculation formula for calculating the combined weight of each indicator is:

[0074]

[0075] Among them, w j is the combined weight value of the jth indicator, μ j ′ is the corrected value of the mean of the product of the subjective weight and objective weight of the jth indicator, and n is the number of indicators.

[0076] Based on the objective weight of each indicator and the subjective weight of each user for each indicator, the mean and coefficient of variation of the product of the subjective weight and objective weight of each indicator are calculated, the mean is corrected using the coefficient of variation, and the combined weight of each indicator is calculated based on the corrected value, taking into account the impact of user heterogeneity on the combined weight. Considering user heterogeneity makes the combined weight of each indicator more scientific, reasonable and accurate.

[0077] After obtaining the combined weights of each indicator, it can be applied to the evaluation of service quality. The following two methods can be used to evaluate service quality:

[0078] (1) Calculate the mean of the service quality perception scores of different users for each indicator, and use the mean of the service quality perception scores of each indicator and the combined weight of each indicator to evaluate using the IPA method;

[0079] (2) Calculate the average of the service quality perception scores of different users for each indicator, and multiply the average by the corresponding combined weight of each indicator to obtain the overall service quality score.

[0080] Example 2 of the present invention takes the indicator weighting of the online car-hailing service quality evaluation problem in Suzhou as an example, and its specific implementation steps are as follows:

[0081] (1) Design questionnaire and collect data

[0082] The known evaluation indicators for the quality of online car-hailing services in Suzhou include: driver safety driving, difficulty of hailing a taxi, vehicle hygiene environment and condition, facility use and configuration, dress and etiquette, App usage experience, deviation, honest and reasonable charges, driver picking up and dropping off passengers midway, reasonable driving route, driver response time, and passenger waiting time.

[0083] The number of indicators n=12.

[0084] According to the above evaluation indicators, a questionnaire was designed to investigate users' perceived scores of the service quality of each indicator, the perceived score of the overall service quality, and the perceived relative importance ratio of two adjacent indicators. The questionnaire is shown in Table 1:

[0085] Table 1

[0086]

[0087]

[0088]

[0089]

[0090] A total of 598 questionnaires were collected, 75 invalid questionnaires were removed, and a total of 523 valid questionnaires were collected. The number of users m = 523, the number of indicators n = 12. The questionnaire survey results are as follows:

[0091]

[0092] (2) Calculation of subjective weight

[0093] Each user scores the importance of the 12th indicator As a base, based on the importance ratio of each group of adjacent indicators given by each user Use each user's rating of the importance of the 12th indicator Represents each user's importance score for the first 11 indicators The calculation formula is:

[0094]

[0095] Based on each user's rating of the importance of each indicator Calculate each user's subjective weight for each indicator The calculation formula is:

[0096]

[0097] in, is the subjective weight of the i-th user on the j-th indicator, Score the importance of the jth indicator for the i-th user, is the importance ratio of the kth and k+1th indicators given by the i-th user, Score the importance of the 12th indicator for the i-th user.

[0098] Taking the first user as an example, that is, i=1, the importance ratio of each group of adjacent indicators given by the first user is:

[0099]

[0100] Score the importance of the 12th indicator by the first user As a base number, combined with the importance ratio of each group of adjacent indicators given by the first user above Calculate the importance score of the first user for the first 11 indicators

[0101] The first user rates the importance of the first 11 indicators The calculation process is:

[0102]

[0103] Calculate the subjective weight of each indicator by the first user based on the first user's importance score for each indicator The calculation process is as follows:

[0104]

[0105] The calculation method of the subjective weights of the 2nd to 523rd users for each indicator is the same as above, and finally the subjective weights of all users for each indicator are obtained.

[0106]

[0107] (3) Calculation of objective weights

[0108] The user's perception of the overall service quality is scored as Q i As the dependent variable, the subjective weight of each indicator Perception score of service quality of each indicator The product of As an independent variable, a multiple linear regression model is constructed, the parameters to be estimated in the multiple linear regression model are calibrated, and the objective weight β of each indicator is obtained. j ;

[0109]

[0110] Among them, [Q i ] m*1 Ratings for 523 users’ perception of overall service quality.

[0111]

[0112] in, 523 users' perceived scores on the service quality of each indicator.

[0113]

[0114] in, It is the subjective weight of each indicator given by 523 users.

[0115] This can be used to calculate the user's subjective weight of each indicator Perception score of service quality of each indicator The product of

[0116]

[0117] The multiple linear regression model expression is:

[0118]

[0119] [Q i ] m*1 , Substitute into the multiple linear regression model expression:

[0120]

[0121] To calibrate the estimated parameters of the above multivariate linear regression model, we actually need to solve:

[0122]

[0123]

[0124] To solve the above constrained nonlinear minimization problem, the Sequential Least Squares Programming (SLSQP) method is used to solve the parameter β 1 =0.155, β 2 =0.117, β 3 =0.047, β 4 =0.028, β 5 =0.074, β 6 =0.053, β 7 =0.082, β 8 =0.091, β 9 =0.099, β 10 =0.109, β 11 =0.071, β 12 =0.074. The objective weights of each indicator are shown in Table 2.

[0125] Table 2

[0126] index Weight Driver driving safely 0.155 Difficulty in getting a taxi 0.117 Vehicle sanitation environment and condition 0.047 Facility use and configuration 0.028 Dress and Etiquette 0.074 App Experience 0.053 deviation 0.082 Honest and reasonable charges 0.091 The driver picked up the passenger and dropped him off midway 0.099 Reasonable driving route 0.109 Driver response time 0.071 Passenger waiting time 0.074

[0127] (4) Calculate the combination weight

[0128] Based on the objective weight of each indicator β j , Each user's subjective weight for each indicator Calculate the product of subjective weight and objective weight of each indicator The mean and coefficient of variation are calculated as follows:

[0129]

[0130] Among them, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, σ j is the standard deviation of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, is the subjective weight of the i-th user on the j-th indicator, β j is the objective weight of the jth indicator.

[0131] Using the coefficient of variation v j For the mean μ j Make corrections to obtain the correction value μ of the mean of the product of subjective weight and objective weight of each indicator j ′, the calculation formula is:

[0132]

[0133] Among them, μ j′ is the corrected value of the mean of the product of the subjective weight and objective weight of the jth indicator, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, and n is the number of indicators.

[0134] The correction value μ based on the mean of the product of subjective weight and objective weight of each indicator j ′, calculate the combined weight w of each indicator j , the calculation formula is:

[0135]

[0136] Among them, w j is the combined weight value of the jth indicator, μ j ′ It is the corrected value of the mean of the product of the subjective weight and objective weight of the j-th indicator.

[0137] From step (2), we can know that each user has a subjective weight for each indicator for:

[0138]

[0139] Through the calculation in step (3), we can know that the objective weight β of each indicator is j is: 1 =0.155, β 2 =0.117, β 3 =0.047, β 4 =0.028, β 5 =0.074, β 6 =0.053, β 7 =0.082, β 8 =0.091, β 9 =0.099, β 10 =0.109, β 11 =0.071, β 12 =0.074.

[0140] The product of subjective weight and objective weight of each indicator for:

[0141]

[0142] Find the mean μ for each column of the above matrix j , standard deviation σ j , coefficient of variation v j , and using the coefficient of variation v j The mean value μ of the product of subjective and objective weightsj After correction, the sum is normalized to obtain the combined weight w of the jth indicator. j , respectively:

[0143]

[0144]

[0145]

[0146] In the second embodiment of the present invention, weights are assigned to various evaluation indicators of Suzhou's online car-hailing service quality through an indicator assignment method for service quality evaluation problems of the present invention, which can effectively improve the scientificity and accuracy of the online car-hailing service quality evaluation work, help online car-hailing service providers to discover problems and formulate reasonable and effective measures to improve their service quality, protect the rights and interests of passengers, promote the healthy development of the market, and have a guiding role in government supervision and policy formulation. At the same time, it also helps to promote industry innovation and improve the quality of life of citizens.

[0147] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0148] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0149] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.

[0150] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0151] Obviously, the above embodiments are merely examples for the purpose of clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.

Claims

1. A method for assigning weights to service quality evaluation indicators, characterized in that: The following steps are involved: Through the survey, we obtain the perceived scores of n service quality indicators given by different users, the perceived scores of overall service quality given by different users, and the importance ratios of each group of adjacent indicators given by different users, where n is the number of indicators; Based on the importance ratio of each group of adjacent indicators given by each user, the subjective weight of each user for each indicator is calculated; The user's perceived score of the overall service quality is taken as the dependent variable, and the product of the subjective weight of each indicator and the perceived score of the service quality of each indicator is taken as the independent variable. A multivariate linear regression model is constructed, and the parameters to be estimated of the multivariate linear regression model are calibrated to obtain the objective weight of each indicator. Based on the objective weight of each indicator and each user's subjective weight of each indicator, the mean and coefficient of variation of the product of the subjective weight and objective weight of each indicator are calculated; The coefficient of variation of the product of the subjective weight and objective weight of each indicator is used to correct the mean of the product of the subjective weight and objective weight of each indicator, so as to obtain the corrected value of the mean of the product of the subjective weight and objective weight of each indicator; based on the corrected value of the mean of the product of the subjective weight and objective weight of each indicator, the combined weight of each indicator is calculated.

2. The method for assigning weights to service quality evaluation indicators according to claim 1, characterized in that: The method of calculating each user's subjective weight for each indicator based on the importance ratio of each group of adjacent indicators given by each user includes the following steps: Taking each user's importance score for the nth indicator as the base, based on the importance ratio of each group of adjacent indicators given by each user, each user's importance score for the nth indicator is used to represent each user's importance score for the first n-1 indicators; Based on each user's score on the importance of each indicator, the subjective weight of each user on each indicator is calculated.

3. The method for assigning weights to service quality evaluation indicators according to claim 2, characterized in that: The importance of each user to the nth indicator is scored As a base, based on the importance ratio of each group of adjacent indicators given by each user Use each user's importance score for the nth indicator Represents each user's importance score for the first n-1 indicators The calculation formula is: The importance of each indicator is scored based on each user Calculate each user's subjective weight for each indicator The calculation formula is: in, is the subjective weight of the i-th user on the j-th indicator, Score the importance of the jth indicator for the i-th user, is the importance ratio of the kth indicator given by the i-th user to the k+1th indicator, The importance score of the nth indicator by the ith user, where n is the number of indicators and m is the number of users.

4. The method for assigning weights to service quality evaluation indicators according to claim 1, characterized in that: The user's perceived score of the overall service quality is taken as the dependent variable, and the product of the subjective weight of each indicator and the perceived score of the service quality of each indicator is taken as the independent variable to construct a multivariate linear regression model: Among them, Q i is the perceived score of the overall service quality by the i-th user, b and β j (j=1,2,…,n) is the estimated parameter of the multivariate linear regression model, b is the constant term, β j is the objective weight of the jth indicator, satisfying β j >0 and is the subjective weight of the i-th user on the j-th indicator, is the perceived rating of the service quality of the jth indicator by the i-th user, ε i is the error term, n is the number of indicators, and m is the number of users.

5. The method for assigning weights to service quality evaluation indicators according to claim 1, characterized in that: The calibration of the estimated parameters of the multivariate linear regression model and obtaining the objective weights of each indicator actually requires solving: The Sequential Least Squares Programming (SLSQP) method is used to solve the above constrained nonlinear minimization problem and obtain the estimated parameters b and β of the multivariate linear regression model. j ; Among them, Q i is the perceived score of the overall service quality by the i-th user, b and β j (j=1,2,…,n) is the estimated parameter of the multivariate linear regression model, b is the constant term, β j is the objective weight of the jth indicator, satisfying β j >0 and is the subjective weight of the i-th user on the j-th indicator, is the perceived rating of the service quality of the jth indicator by the i-th user, ε i is the error term, n is the number of indicators, and m is the number of users.

6. The method for assigning weights to service quality evaluation indicators according to claim 1, characterized in that: Based on the objective weight of each indicator and the subjective weight of each user for each indicator, the mean and coefficient of variation of the product of the subjective weight and the objective weight of each indicator are calculated. The calculation formula is: Among them, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, σ j is the standard deviation of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, is the subjective weight of the i-th user on the j-th indicator, β j is the objective weight of the jth indicator, n is the number of indicators, and m is the number of users.

7. The method for assigning weights to service quality evaluation indicators according to claim 1, characterized in that: The coefficient of variation of the product of the subjective weight and objective weight of each indicator is used to correct the mean of the product of the subjective weight and objective weight of each indicator to obtain the corrected value of the mean of the product of the subjective weight and objective weight of each indicator. The calculation formula is: Among them, μ j ′ is the corrected value of the mean of the product of the subjective weight and objective weight of the jth indicator, μ j is the mean of the product of the subjective weight and objective weight of the jth indicator, v j is the coefficient of variation of the product of the subjective weight and objective weight of the jth indicator, and n is the number of indicators.

8. The method for assigning weights to service quality evaluation indicators according to claim 1, characterized in that: The combined weight of each indicator is calculated based on the correction value of the mean of the product of the subjective weight and the objective weight of each indicator. The calculation formula is: Among them, w j is the combined weight value of the jth indicator, μ j ′ is the corrected value of the mean of the product of the subjective weight and objective weight of the jth indicator, and n is the number of indicators.

9. The method for assigning weights to service quality evaluation indicators according to claim 1, characterized in that: Also includes: The Likert 5-level scoring method is used to obtain the perceived scores given by different users on n indicators of service quality and the perceived scores given by different users on the overall service quality through surveys.