A big data-based customer service center dispatching system and method

Abstract

The application discloses a kind of based on big data's customer service center dispatch system and method, and the application belongs to dispatch center technical field.System includes artificial customer service analysis module, data storage module, customer consultation time detection module, data analysis module, text input frequency detection module, task allocation module and user experience detection module.Artificial customer service business processing speed and the state of work line are obtained by artificial customer service analysis module;The relationship of customer and intelligent customer service chat time and customer transfer artificial customer service time is obtained by data analysis module;Customer and intelligent customer service consultation time is obtained by customer consultation time, and the prediction value of customer transfer artificial customer service time can be obtained by relationship formula, the prediction value of customer transfer artificial customer service time is subtracted from customer and intelligent customer service consultation time, and the waiting time of customer in transfer artificial customer service can be obtained, and customer is informed of the waiting time, can save customer time, and high customer experience.

Description

Technical Field

[0001] This invention relates to the field of dispatch center technology, specifically to a customer service center dispatch system and method based on big data. Background Technology

[0002] Currently, customer service centers offer two service methods: intelligent customer service and human customer service. Intelligent customer service can provide timely assistance to customers, but it cannot accurately solve problems, so human customer service is still needed.

[0003] When faced with sudden surges in customer traffic or a large volume of customer service during work hours, customers may experience queues. If customers are unaware of the waiting time for human assistance, excessively long queues may cause them to give up on seeking help, resulting in their problems not being resolved in a timely manner and creating a negative customer experience. Summary of the Invention

[0004] The purpose of this invention is to provide a customer service center scheduling system and method based on big data to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a customer service center scheduling system based on big data, the system including a human customer service analysis module, a data storage module, a customer consultation time detection module, a data analysis module, a text input frequency detection module, a task allocation module and a user experience detection module;

[0006] The human customer service analysis module analyzes the speed and workflow of human customer service operations.

[0007] The data storage module stores the processing speed and workflow status of human customer service in the database.

[0008] The customer consultation time detection module detects the consultation time between customer service and intelligent customer service.

[0009] The data analysis module analyzes the time of customer inquiries with the intelligent customer service and the historical data of customer service transfers to human customer service.

[0010] The text input frequency detection module detects the speed at which customers input text during consultations with customer service.

[0011] The task allocation module assigns customers to corresponding human customer service representatives.

[0012] The user experience detection module is used to detect the quality of the user experience.

[0013] The output of the human customer service analysis module is connected to the input of the data storage module, the output of the data storage module is connected to the input of the task allocation module, the output of the customer consultation detection module is connected to the input of the data analysis module, the output of the data analysis module is connected to the input of the task allocation module, the output of the text input frequency detection module is connected to the input of the customer consultation detection module, and the output of the task allocation module is connected to the input of the user experience detection module.

[0014] The human customer service analysis module includes a human customer service analysis unit and a human customer service workflow detection unit.

[0015] The human customer service analysis unit analyzes the speed of human customer service processing.

[0016] The manual customer service work line detection unit is used to detect whether the manual customer service work line is idle.

[0017] The data analysis module includes a data acquisition unit and a data analysis unit;

[0018] The data collection unit collects historical data on the time customers spend consulting with the intelligent customer service and historical data on the time customers need to be transferred to human customer service.

[0019] The data analysis unit analyzes the relationship between the time customers consult with the intelligent customer service system and the time customers need to transfer to human customer service, based on historical data collected from customers' consultation time with the intelligent customer service system and the time customers need to transfer to human customer service.

[0020] The text input frequency detection module is used to detect the frequency of the customer's text input and analyze the speed of the customer's text input and whether the user is hanging up based on the input frequency.

[0021] A big data-based customer service center scheduling method is described below:

[0022] S5-1. Analyze the processing speed of human customer service and the work routes of human customer service representatives, store the analysis results in the database, and update the work route status of human customer service representatives in real time.

[0023] S5-2. Collect historical data on the time customers spend consulting with the intelligent customer service and the time customers need to be transferred to human customer service, and analyze the relationship between the time customers spend consulting with the intelligent customer service and the time customers transfer to human customer service.

[0024] S5-3, The customer consultation time detection module can predict the time it will take for a customer to receive a human customer service consultation by using the relationship between the customer's consultation time with the intelligent customer service and the time it takes for the customer to be transferred to human customer service. After the customer ends the consultation with the intelligent customer service, the intelligent customer service will inform the customer of the waiting time for the transfer to human customer service and ask the customer to wait patiently.

[0025] S5-4. After the customer ends the chat with the intelligent customer service, the text input frequency detection module is used to detect the frequency of the customer's text input to see if the customer's typing speed and hanging up time caused the customer's consultation time with the intelligent customer service to be too long.

[0026] S5-5. Excluding cases where the consultation time with the intelligent customer service is too long due to the customer's typing speed or hanging up, the customer's understanding of the problem is judged by the consultation time with the intelligent customer service. Based on the work line status and business processing speed of the human customer service in the database, the customer is assigned. The longer the consultation time with the intelligent customer service, the more complex the customer's problem is, and the faster the business processing speed of the human customer service is needed to solve the customer's doubts. The customer is then assigned to the corresponding human customer service.

[0027] S5-6. Analyze the customer's user experience value based on the time required for the transition between intelligent customer service and human customer service; if the customer's user experience is poor, it is necessary to increase the number of human customer service routes.

[0028] The specific method for analyzing the customer service work routes in S5-1 is as follows: the customer service work route detection unit is used to detect whether the customer service work routes are idle, the customer service work routes are detected in real time, the idle work routes are stored in the database, and customers are assigned to customer service representatives whose work routes are idle.

[0029] The specific method for analyzing the relationship between customer consultation time and intelligent customer service time, and customer transfer time to human customer service in S5-2, is as follows:

[0030] S7-1. Establish a mathematical model to collect the time customers spend consulting with the intelligent customer service system and the time customers need to be transferred to human customer service. Let these samples be (x...). i y i ), i = 1, 2, 3, ..., n; where x i For the time customers spend consulting with the intelligent customer service, y i The time required for a customer to be transferred to a live customer service representative;

[0031] S7-2. Set the fitted curve y = kx + b, and let the fitted value... Based on the definition conditions of sample points and fitted curves able to obtain Where k is the curve coefficient and b is the error term. These are the predicted values ​​for the curve coefficients. Error term predicted value, The estimated time required for a customer to be transferred to a human agent;

[0032] S7-3, Settings The goal is to minimize the value of L, where L is the sum of squared residuals;

[0033] S7-4, Calculation

[0034]

[0035]

[0036]

[0037]

[0038]

[0039] Similarly, we can obtain

[0040]

[0041] S7-5. Obtain the fitted curve through calculation.

[0042]

[0043] Where x i For the time customers spend consulting with the intelligent customer service, y i This refers to the time required for a customer to be transferred to a live customer service representative.

[0044] The specific waiting time for a customer to be transferred to a live customer service representative in S5-3 is as follows:

[0045] Let P be the waiting time for a customer to be transferred to a human agent, and let x be the time for the customer to consult with the automated customer service. i By fitting the curve, we can obtain the time y for a customer to be transferred to a human customer service representative. i The waiting time for a customer to be transferred to a live agent is P = y. i -x i .

[0046] The calculation method for the user experience value in S5-6 is as follows:

[0047] Let c be the frequency of customer text input, w = 1 / P be the customer's patience level, and F = C1c + C2c + εw be the user experience value; where C1, C2, and ε represent coefficients; the frequency of customer text input is... At that time, C2 = 0, the frequency of customer service text input. When C1 = 0, C2 > C1; where The threshold for text input frequency set by the system is used to determine the customer's patience level (w), which is inversely proportional to the time it takes for the customer to be transferred to a human agent. The longer the time it takes for the customer to be transferred to a human agent, the lower the customer's patience level will be.

[0048] Compared with the prior art, the beneficial effects achieved by the present invention are: after a customer consults with the intelligent customer service, the present invention will inform the customer of the waiting time required to transfer to human customer service. In the case of congestion in the consultation channel, the customer's time is reduced, the utilization rate and service level of human customer service are improved, and a better user experience is obtained with fewer human customer service resources. Attached Figure Description

[0049] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0050] Figure 1 This is a schematic diagram of the structure of a customer service center dispatch system based on big data; Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Please see Figure 1 The present invention provides a technical solution: a customer service center scheduling system based on big data, the system including a human customer service analysis module, a data storage module, a customer consultation time detection module, a data analysis module, a text input frequency detection module, a task allocation module and a user experience detection module;

[0053] The human customer service analysis module analyzes the speed and workflow of human customer service operations.

[0054] The data storage module stores the processing speed and workflow status of human customer service in the database.

[0055] The customer consultation time detection module detects the consultation time between customer service and intelligent customer service.

[0056] The data analysis module analyzes the time of customer inquiries with the intelligent customer service and the historical data of customer service transfers to human customer service.

[0057] The text input frequency detection module detects the speed at which customers input text during consultations with customer service.

[0058] The task allocation module assigns customers to corresponding human customer service representatives.

[0059] The user experience detection module is used to detect the quality of the user experience.

[0060] The output of the human customer service analysis module is connected to the input of the data storage module, the output of the data storage module is connected to the input of the task allocation module, the output of the customer consultation detection module is connected to the input of the data analysis module, the output of the data analysis module is connected to the input of the task allocation module, the output of the text input frequency detection module is connected to the input of the customer consultation detection module, and the output of the task allocation module is connected to the input of the user experience detection module.

[0061] The human customer service analysis module includes a human customer service analysis unit and a human customer service workflow detection unit.

[0062] The human customer service analysis unit analyzes the speed of human customer service processing.

[0063] The manual customer service work line detection unit is used to detect whether the manual customer service work line is idle.

[0064] The data analysis module includes a data acquisition unit and a data analysis unit;

[0065] The data collection unit collects historical data on the time customers spend consulting with the intelligent customer service and historical data on the time customers need to be transferred to human customer service.

[0066] The data analysis unit analyzes the relationship between the time customers consult with the intelligent customer service system and the time customers need to transfer to human customer service, based on historical data collected from customers' consultation time with the intelligent customer service system and the time customers need to transfer to human customer service.

[0067] The text input frequency detection module is used to detect the frequency of the customer's text input and analyze the speed of the customer's text input and whether the user is hanging up based on the input frequency.

[0068] A big data-based customer service center scheduling method is described below:

[0069] S5-1. Analyze the processing speed of human customer service and the work routes of human customer service representatives, store the analysis results in the database, and update the work route status of human customer service representatives in real time.

[0070] S5-2. Collect historical data on the time customers spend consulting with the intelligent customer service and the time customers need to be transferred to human customer service, and analyze the relationship between the time customers spend consulting with the intelligent customer service and the time customers transfer to human customer service.

[0071] S5-3, The customer consultation time detection module can predict the time it will take for a customer to receive a human customer service consultation by using the relationship between the customer's consultation time with the intelligent customer service and the time it takes for the customer to be transferred to human customer service. After the customer ends the consultation with the intelligent customer service, the intelligent customer service will inform the customer of the waiting time for the transfer to human customer service and ask the customer to wait patiently.

[0072] S5-4. After the customer ends the chat with the intelligent customer service, the text input frequency detection module is used to detect the frequency of the customer's text input to see if the customer's typing speed and hanging up time caused the customer's consultation time with the intelligent customer service to be too long.

[0073] S5-5. Excluding cases where the consultation time with the intelligent customer service is too long due to the customer's typing speed or hanging up, the customer's understanding of the problem is judged by the consultation time with the intelligent customer service. Based on the work line status and business processing speed of the human customer service in the database, the customer is assigned. The longer the consultation time with the intelligent customer service, the more complex the customer's problem is, and the faster the business processing speed of the human customer service is needed to solve the customer's doubts. The customer is then assigned to the corresponding human customer service.

[0074] S5-6. Analyze the customer's user experience value based on the time required for the transition between intelligent customer service and human customer service; if the customer's user experience is poor, it is necessary to increase the number of human customer service routes.

[0075] The specific method for analyzing the customer service work routes in S5-1 is as follows: the customer service work route detection unit is used to detect whether the customer service work routes are idle, the customer service work routes are detected in real time, the idle work routes are stored in the database, and customers are assigned to customer service representatives whose work routes are idle.

[0076] The specific method for analyzing the relationship between customer consultation time and intelligent customer service time, and customer transfer time to human customer service in S5-2, is as follows:

[0077] S7-1. Establish a mathematical model to collect the time customers spend consulting with the intelligent customer service system and the time customers need to be transferred to human customer service. Let these samples be (x...). i y i ), i = 1, 2, 3, ..., n; where x i For the time customers spend consulting with the intelligent customer service, y i The time required for a customer to be transferred to a live customer service representative;

[0078] S7-2. Set the fitted curve y = kx + b, and let the fitted value... Based on the definition conditions of sample points and fitted curves able to obtain Where k is the curve coefficient and b is the error term. These are the predicted values ​​for the curve coefficients. Error term predicted value, The estimated time required for a customer to be transferred to a human agent;

[0079] S7-3, Settings The goal is to minimize the value of L, where L is the sum of squared residuals;

[0080] S7-4, Calculation

[0081]

[0082]

[0083]

[0084]

[0085]

[0086] Similarly, we can obtain

[0087]

[0088] S7-5. Obtain the fitted curve through calculation.

[0089]

[0090] Where x i For the time customers spend consulting with the intelligent customer service, y i This refers to the time required for a customer to be transferred to a live customer service representative.

[0091] The specific waiting time for a customer to be transferred to a live customer service representative in S5-3 is as follows:

[0092] Let P be the waiting time for a customer to be transferred to a human agent, and let x be the time for the customer to consult with the automated customer service. i By fitting the curve, we can obtain the time y for a customer to be transferred to a human customer service representative. i The waiting time for a customer to be transferred to a live agent is P = y. i -x i .

[0093] The calculation method for the user experience value in S5-6 is as follows:

[0094] Let c be the frequency of customer text input, w = 1 / P be the customer's patience level, and F = C1c + C2c + εw be the user experience value; where C1, C2, and ε represent coefficients; the frequency of customer text input is... At that time, C2 = 0, the frequency of customer service text input. When C1 = 0, C2 > C1; where The threshold for text input frequency set by the system is used to determine the customer's patience level (w), which is inversely proportional to the time it takes for the customer to be transferred to a human agent. The longer the time it takes for the customer to be transferred to a human agent, the lower the customer's patience level will be.

[0095] In this implementation case:

[0096] The time for customer consultation with the intelligent customer service and the time for customers to be transferred to human agents are set as follows: (12, 18), (16, 20), (15, 21), (17, 19), (18, 26);

[0097] According to the formula Substituting the duration of customer consultation with the intelligent customer service system and the time required for the customer to be transferred to a human agent, we can obtain y = 0.925x + 6.377;

[0098] If the customer's consultation with the AI ​​customer service lasts for 20 minutes, the customer will be successfully transferred to a human customer service representative at 24.877 minutes, and the customer will only need to wait for 4.877 minutes.

[0099] The frequency of customer text input is set to 0.7, the threshold for text input frequency set by the system is 0.6, and the coefficients C1, C2 and ε are 0.1, 0.2 and 0.7 respectively; the user experience value is divided into 0-0.05 (poor experience value), 0.05-0.1 (medium experience value), and greater than or equal to 0.1 (good experience value).

[0100] Therefore, the user experience value F = 0.1 × 0.7 + 0.7 × 0.21 = 0.217, which is a good user experience value.

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0102] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A customer service center scheduling method based on big data, characterized in that: The specific methods for dispatching customer service centers are as follows: S5-1. Analyze the processing speed of human customer service and the work routes of human customer service representatives, store the analysis results in the database, and update the work route status of human customer service representatives in real time. S5-2. Collect historical data on the time customers spend consulting with the intelligent customer service and the time customers need to be transferred to human customer service, and analyze the relationship between the time customers spend consulting with the intelligent customer service and the time customers transfer to human customer service. S5-3, The customer consultation time detection module can predict the time it will take for a customer to receive a human customer service consultation by using the relationship between the customer's consultation time with the intelligent customer service and the time it takes for the customer to be transferred to human customer service. After the customer ends the consultation with the intelligent customer service, the intelligent customer service will inform the customer of the waiting time for the transfer to human customer service and ask the customer to wait patiently. S5-4. After the customer ends the chat with the intelligent customer service, the text input frequency detection module detects the frequency of the customer's text input to see if the customer's typing speed and hanging up time caused the customer's consultation time with the intelligent customer service to be too long. S5-5. Excluding cases where the consultation time with the intelligent customer service is too long due to the customer's typing speed or hanging up, the customer's understanding of the problem is judged by the consultation time with the intelligent customer service. Based on the work line status and business processing speed of the human customer service in the database, the customer is assigned to the corresponding human customer service. S5-6. Analyze the customer's user experience value based on the waiting time required during the transition from intelligent customer service to human customer service; if the customer's user experience is poor, it is necessary to increase the number of human customer service routes. The specific method for analyzing the relationship between customer consultation time and intelligent customer service time, and customer transfer time to human customer service in S5-2, is as follows: S7-1. Establish a mathematical model to collect data on the time customers spend consulting with the intelligent customer service system and the time customers need to be transferred to a human customer service representative. Set these samples as follows: , ;in The time allotted for customer consultation with the intelligent customer service system. The time required for a customer to be transferred to a live customer service representative; S7-2, Setting the Fitting Curve Let the fitted value According to the definition conditions of sample points and fitted curves able to obtain ,in For curve coefficients, For error terms, These are the predicted values ​​for the curve coefficients. Error term predicted value, The estimated time required for a customer to be transferred to a human agent; S7-3, Settings It is necessary to make The value is the smallest, among which It is the sum of squared residuals; S7-4. Obtain the fitted curve through calculation. ,in The time allotted for customer consultation with the intelligent customer service system. The time required for a customer to be transferred to a live customer service representative; The calculation method for the user experience value in S5-6 is as follows: Set the frequency of customer text input to The customer's patience level is User experience value ;in , and Representative coefficient; frequency of text input by the customer. hour, Customer text input frequency hour, , ;in The threshold for text input frequency set by the system, and the customer's patience level. This is inversely proportional to the time it takes for a customer to be transferred to a human customer service representative.

2. The customer service center scheduling method based on big data according to claim 1, characterized in that: The specific method for analyzing the customer service work routes in S5-1 is as follows: the customer service work route detection unit is used to detect whether the customer service work routes are idle, the customer service work routes are detected in real time, the idle work routes are stored in the database, and customers are assigned to customer service representatives whose work routes are idle.

3. The customer service center scheduling method based on big data according to claim 2, characterized in that: The specific waiting time for a customer to be transferred to a live customer service representative in S5-3 is as follows: Set the waiting time for customers to be transferred to human customer service. The time for customers to consult with the intelligent customer service is By fitting the curve, we can obtain the time it takes for a customer to be transferred to a human customer service representative. The waiting time for a customer to be transferred to a live agent. .

4. A big data-based customer service center scheduling system, used to implement the big data-based customer service center scheduling method as described in claim 3, characterized in that: The system includes a human customer service analysis module, a data storage module, a customer consultation time detection module, a data analysis module, a text input frequency detection module, a task allocation module, and a user experience detection module. The human customer service analysis module analyzes the speed and workflow of human customer service operations. The data storage module stores the processing speed and workflow status of human customer service in the database. The customer consultation time detection module detects the consultation time between customer service and intelligent customer service. The data analysis module analyzes the time of customer inquiries with the intelligent customer service and the historical data of customer service transfers to human customer service. The text input frequency detection module detects the speed at which customers input text during consultations with customer service. The task allocation module assigns customers to corresponding human customer service representatives. The user experience detection module is used to detect the quality of the user experience. The output of the human customer service analysis module is connected to the input of the data storage module, the output of the data storage module is connected to the input of the task allocation module, the output of the customer consultation time detection module is connected to the input of the data analysis module, the output of the data analysis module is connected to the input of the task allocation module, the output of the text input frequency detection module is connected to the input of the customer consultation time detection module, and the output of the task allocation module is connected to the input of the user experience detection module.

5. A customer service center dispatch system based on big data according to claim 4, characterized in that: The human customer service analysis module includes a human customer service analysis unit and a human customer service workflow detection unit; The human customer service analysis unit analyzes the speed of human customer service processing. The manual customer service work line detection unit is used to detect whether the manual customer service work line is idle.

6. A customer service center dispatch system based on big data according to claim 4, characterized in that: The data analysis module includes a data acquisition unit and a data analysis unit; The data collection unit collects historical data on the time customers spend consulting with the intelligent customer service and historical data on the time customers need to be transferred to human customer service. The data analysis unit analyzes the relationship between the time customers consult with the intelligent customer service system and the time customers need to transfer to human customer service, based on historical data collected from customers' consultation time with the intelligent customer service system and the time customers need to transfer to human customer service.

7. A customer service center dispatch system based on big data according to claim 4, characterized in that: The text input frequency detection module is used to detect the frequency of the customer's text input and analyze the speed of the customer's text input and whether the user is hanging up based on the input frequency.

Images