Outbound call control method and device, electronic device and storage medium

Abstract

The present application provides an outbound call control method and device, an electronic device and a storage medium, and relates to the field of data processing technology. The method inputs the waiting time and / or call loss amount at the current moment into a pre-trained outbound call idle seat prediction model to obtain the predicted idle seat number of the second category outbound call seats at the next moment; uses a preset control algorithm to calculate the idle seat number that needs to be adjusted when the idle seat number of the second category outbound call seats at the next moment is controlled from the predicted idle seat number to the preset value; calculates the outbound call parameters at the next moment based on the idle seat number that needs to be adjusted at the next moment; and adjusts the outbound call parameters at the current moment to the outbound call parameters at the next moment to make an outbound call. The embodiment of the present application can not only reduce the loss of intended users when the first category outbound call seats switch to the second category outbound call seats, but also shorten the waiting time between two calls of the second category outbound call seats, thereby improving the outbound call efficiency and order conversion rate, and improving user satisfaction with the service.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an outbound call control method and device, an electronic device, and a storage medium. Background Art

[0002] Effectively screening high-intent users can significantly improve conversion rates, especially in the insurance sales sector. In recent years, artificial intelligence has achieved unprecedented development in the financial sector. The widespread application of technologies such as ASR (Automatic Speech Recognition) and NLP (Natural Language Processing) provides technical support for robot online sales functions. Although the industry's predictive outbound calling system can directly enable agents to communicate with users, reducing the time agents spend on frequent dialing, it still does not filter out uninterested users. The AI ​​(Artificial Intelligence)-to-human outbound calling system uses a robot to communicate with the connected user, and after determining that the user has a strong intention to purchase insurance, it transfers the call to a human agent, further improving the conversion rate to a single order.

[0003] While AI-to-human outbound calls can effectively increase conversion rates, due to limited human agent capacity, conversations between bots and users can be transferred to human agents when all other agents are busy, leading to call losses. Call losses are inevitable, and how to minimize excessive call loss rates while reducing the average idle time of human agents has become a pressing technical challenge. Summary of the Invention

[0004] In view of the above problems, this application is proposed to provide an outbound call control method and device, electronic device, and storage medium that overcome the above problems or at least partially solve the above problems. The technical solution is as follows:

[0005] In a first aspect, an outbound call control method is provided, which is applied to a scenario in which an outbound call center uses a first-class outbound call agent to communicate with a connected user. When the connected user is determined to meet set conditions and is a desired user, the desired user is submitted to a waiting queue to be transferred to a second-class outbound call agent, and the second-class outbound call agent communicates with the desired user. The method includes:

[0006] Obtain the current waiting time for transferring the intended user from the waiting queue to the second-class outbound call agent and / or the call loss amount caused by the lack of a second-class outbound call agent available for transfer;

[0007] Inputting the current waiting time and / or call loss into a pre-trained outbound call idle agent prediction model to predict the number of idle agents of the second category of outbound call agents at the next moment, thereby obtaining the predicted number of idle agents of the second category of outbound call agents at the next moment;

[0008] A preset control algorithm is used to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second type of outbound call seats at the next moment is controlled from the predicted number of idle seats to the preset value;

[0009] Calculate the outbound call parameters for the next moment based on the number of idle seats that need to be adjusted at the next moment;

[0010] Adjust the outbound call parameters at the current moment to the outbound call parameters at the next moment to make an outbound call.

[0011] In one possible implementation, the outbound call idle agent prediction model is trained through the following steps:

[0012] Build an initial outbound call idle agent prediction model;

[0013] The historical waiting time and / or call loss amount, as well as the number of available seats of the second type of outbound call agents corresponding to the historical waiting time and / or call loss amount, are input into an initial outbound call available seat prediction model, and the initial outbound call available seat prediction model is trained to obtain a trained outbound call available seat prediction model.

[0014] In one possible implementation, the preset control algorithm includes a discretized PID algorithm. The method of using the preset control algorithm to calculate the number of idle seats that need to be adjusted when controlling the number of idle seats of the second type of outbound call agents at the next moment from the predicted number of idle seats to the preset value includes:

[0015] The discretized PID algorithm is used to calculate the deviation, integral deviation, and differential deviation of the number of idle seats of the second type of outbound call seats at the next moment from the predicted number of idle seats when it is controlled at the preset value, and the number of idle seats that needs to be adjusted is obtained.

[0016] In one possible implementation, the discrete PID algorithm is used to calculate the deviation, integral deviation, and differential deviation of the number of available seats of the second type of outbound call agents at the next moment from the predicted number of available seats when the number is controlled at a preset value, and the number of available seats that needs to be adjusted is obtained, including:

[0017] Using the discretized PID algorithm, assuming that the sampling time interval is T, then at the next time t, when the number of idle seats of the second type of outbound call seats at the next time t is controlled from the predicted idle seats to the preset value, the deviation between the predicted idle seats at the next time t and the preset value is e(k), the deviation between the predicted idle seats at time tT and the preset value is e(k-1), the deviation between the predicted idle seats at time t-2T and the preset value is e(k-2), and so on;

[0018] The integral deviation is e(k)+e(k-1)+e(k-2)+...+e(0);

[0019] The differential deviation is (e(k)-e(k-1)) / T;

[0020] The number of idle seats that need to be adjusted at the next moment t is u(k), and the number of idle seats that need to be adjusted at moment tT is u(k-1);

[0021]

[0022]

[0023] u(k)=u(k-1)+△u(k);

[0024] where increment ∆u(k) = α(e(k) - e(k-1)) + βe(k) + θ(e(k) - 2e(k-1) + e(k-2));

[0025] α is an input parameter, which indicates the desired ratio to compensate for the gap between the predicted number of available seats and the preset value. It is a value between 0 and 1.

[0026] β = α × T / Ti, Ti is the statistical interval of the integral deviation, which is an integer multiple of T;

[0027] θ = α × Td / T, where Td is the statistical interval of differential deviation and is an integer multiple of T.

[0028] In one possible implementation, the outbound call parameters at the next moment are calculated according to the number of idle seats that need to be adjusted at the next moment using the following formula, including:

[0029] Number of outbound calls at the next moment = u(k) / (Current moment call loss rate × Transfer rate of potential users submitted to the waiting queue for transfer to the second-class outbound call agent × Connection rate of outbound calls initiated by the outbound call center and calls between the first-class outbound call agent and the user)

[0030] The outbound call multiplier at the next moment = the number of outbound calls at the next moment / 100.

[0031] In a second aspect, an outbound call control device is provided, which is applied to a scenario in which an outbound call center uses a first-class outbound call agent to communicate with a connected user. When the connected user is determined to meet set conditions and is an intended user, the intended user is submitted to a waiting queue to be transferred to a second-class outbound call agent, and the second-class outbound call agent communicates with the intended user. The device includes:

[0032] An acquisition module is used to obtain the waiting time for transferring the intended user from the waiting queue to the second-class outbound call agent for a call at the current moment and / or the call loss amount caused by the lack of a second-class outbound call agent available for transfer at the current moment;

[0033] A prediction module is configured to input the current waiting time and / or call loss into a pre-trained outbound call idle agent prediction model to predict the number of idle seats of the second category of outbound call agents at the next moment, thereby obtaining a predicted number of idle seats of the second category of outbound call agents at the next moment;

[0034] a control module for calculating, by using a preset control algorithm, the number of idle seats that needs to be adjusted when the number of idle seats of the second type of outbound call agents at the next moment is controlled from the predicted number of idle seats to the preset value;

[0035] A calculation module is used to calculate the outbound call parameters at the next moment according to the number of idle seats that need to be adjusted at the next moment;

[0036] The adjustment module is used to adjust the outbound call parameters at the current moment to the outbound call parameters at the next moment to make an outbound call.

[0037] In a possible implementation, the apparatus further includes a training module configured to:

[0038] Build an initial outbound call idle agent prediction model;

[0039] The historical waiting time and / or call loss amount, as well as the number of available seats of the second type of outbound call agents corresponding to the historical waiting time and / or call loss amount, are input into an initial outbound call available seat prediction model, and the initial outbound call available seat prediction model is trained to obtain a trained outbound call available seat prediction model.

[0040] In a possible implementation, the preset control algorithm includes a discretized PID algorithm; and the control module is further configured to:

[0041] The discretized PID algorithm is used to calculate the deviation, integral deviation, and differential deviation of the number of idle seats of the second type of outbound call seats at the next moment from the predicted number of idle seats when it is controlled at the preset value, and the number of idle seats that needs to be adjusted is obtained.

[0042] In a third aspect, an electronic device is provided, comprising a processor and a memory, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any of the above-mentioned outbound call control methods.

[0043] In a fourth aspect, a storage medium is provided, wherein the storage medium stores a computer program, wherein the computer program is configured to execute any of the above-mentioned outbound call control methods when running.

[0044] By means of the above-mentioned technical solution, the outbound call control method and device, electronic device and storage medium provided in the embodiments of the present application can obtain the waiting time for transferring the intended user from the waiting queue to the second-class outbound call agent for a call at the current moment and / or the call loss amount caused by the lack of a second-class outbound call agent to transfer the call at the current moment; input the waiting time and / or call loss amount at the current moment into a pre-trained outbound call idle seat prediction model to predict the number of idle seats of the second-class outbound call agents at the next moment, and obtain the predicted idle seats of the second-class outbound call agents at the next moment; use a preset control algorithm to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second-class outbound call agents at the next moment is controlled from the predicted idle seats to a preset value; calculate the outbound call parameters at the next moment based on the number of idle seats that need to be adjusted at the next moment; and adjust the outbound call parameters at the current moment to the outbound call parameters at the next moment to make an outbound call. It can be seen that the embodiment of the present application adopts a preset control algorithm to perform real-time calculation of the number of idle seats that need to be adjusted, controls the predicted number of idle seats at a preset value, and then calculates the outbound call parameters at the next moment based on the number of idle seats that need to be adjusted, and adjusts the outbound call parameters at the current moment to the outbound call parameters at the next moment to perform outbound calls, thereby achieving the purpose of reducing the loss of potential users who switch from the first category of outbound call seats to the second category of outbound call seats, and shortening the waiting time interval between two calls of the second category of outbound call seats, thereby improving outbound call efficiency and order conversion rate, and improving user satisfaction with the service. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments of the present application.

[0046] Figure 1 A flowchart of the outbound call control method provided by an embodiment of the present application is shown;

[0047] Figure 2 A schematic diagram showing an application scenario provided by an embodiment of the present application is shown;

[0048] Figure 3 The structure diagram of the outbound call control device provided in an embodiment of the present application is shown;

[0049] Figure 4 A structural diagram of an outbound call control device provided by another embodiment of the present application is shown;

[0050] Figure 5 A structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0051] The following describes exemplary embodiments of the present application in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0052] It should be noted that the terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that such usage is interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the term "including" and its variations are to be interpreted as open-ended terms meaning "including but not limited to."

[0053] To address the aforementioned technical issues, an embodiment of the present application provides an outbound call control method that can be applied to a scenario where an outbound call center uses a first-class outbound call agent to communicate with a connected user. When the connected user is determined to meet set conditions and is an intended user, the intended user is submitted to a waiting queue to be transferred to a second-class outbound call agent, who then communicates with the intended user. The set conditions here can be set according to actual needs. For example, if the call duration between the first-class outbound call agent and the connected user is greater than 5 seconds, or if the call between the first-class outbound call agent and the connected user contains set keywords, etc., then the connected user can be determined to be an intended user.

[0054] The outbound call control method provided in the embodiment of the present application can be applied to electronic devices such as servers, personal computers, smart phones, tablet computers, etc. Figure 1 As shown, the outbound call control method may specifically include the following steps S101 to S105:

[0055] Step S101: obtaining the waiting time for transferring the intended user from the waiting queue to the second type outbound call agent for a call and / or the call loss amount caused by the lack of the second type outbound call agent available for transfer at the current moment.

[0056] As mentioned above, when it is determined that the connected user meets the set conditions and is an intended user, the intended user is submitted to the waiting queue to be transferred to the second type of outbound call agent. The waiting time for transferring the intended user from the waiting queue to the second type of outbound call agent for a call at the current moment can be obtained. For example, at the current moment, there are 10 potential users in the waiting queue, namely, potential users 1 to 10, and there are 5 second-class outbound call seats. The waiting time for transferring potential user 1 to the second-class outbound call seat for a call at the current moment can be obtained, which is used as the waiting time for transferring the potential user from the waiting queue to the second-class outbound call seat for a call at the current moment; the average waiting time for transferring potential users 1 and 2 to the second-class outbound call seat for a call at the current moment can also be obtained, which is used as the waiting time for transferring the potential user from the waiting queue to the second-class outbound call seat for a call at the current moment; the average waiting time for transferring potential users 1 to 3 to the second-class outbound call seat for a call at the current moment can also be obtained, which is used as the waiting time for transferring the potential user from the waiting queue to the second-class outbound call seat for a call at the current moment.

[0057] For another example, there are currently five interested users in the waiting queue, namely, interested users 1 to 5, and no second-category outbound call agent is available for transfer. Therefore, the call loss amount caused by the lack of a second-category outbound call agent is 5. It should be noted that the examples here are merely illustrative and do not limit the embodiments of the present application.

[0058] Step S102: Input the current waiting time and / or call loss into a pre-trained outbound call idle agent prediction model to predict the number of idle seats of the second category of outbound call agents at the next moment, thereby obtaining the predicted number of idle seats of the second category of outbound call agents at the next moment.

[0059] Step S103 , using a preset control algorithm to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second type of outbound call seats at the next moment is controlled from the predicted number of idle seats to the preset value.

[0060] Step S104: Calculate the outbound call parameters at the next moment according to the number of idle seats that need to be adjusted at the next moment.

[0061] Step S105: Adjust the outbound call parameters at the current moment to the outbound call parameters at the next moment to make an outbound call.

[0062] The embodiment of the present application adopts a preset control algorithm to perform real-time calculation of the number of idle seats that need to be adjusted, controls the predicted number of idle seats to a preset value, and then calculates the outbound call parameters at the next moment based on the number of idle seats that need to be adjusted. The outbound call parameters at the current moment are adjusted to the outbound call parameters at the next moment for outbound calling. This achieves the purpose of reducing the loss of potential users when the first type of outbound call seats switch to the second type of outbound call seats, and shortening the waiting time interval between two calls of the second type of outbound call seats, thereby improving outbound call efficiency and order conversion rate, and increasing user satisfaction with the service.

[0063] This embodiment of the present application provides a possible implementation method, which can train the outbound call idle agent prediction model mentioned in step S102 above by following the steps A1 and A2:

[0064] Step A1, building an initial prediction model;

[0065] In step A2, the historical waiting time and / or call loss volume, as well as the number of available seats of the second category of outbound call agents corresponding to the historical waiting time and / or call loss volume, are input into an initial outbound available seat prediction model, and the initial outbound available seat prediction model is trained to obtain a trained outbound available seat prediction model.

[0066] In this embodiment, an initial outbound call idle agent prediction model is constructed and trained to obtain a trained outbound call idle agent prediction model. This model can accurately and effectively predict the number of idle seats of the second category of outbound call agents, so that the predicted number of idle seats of the second category of outbound call agents can be adjusted in a timely manner.

[0067] In one embodiment of the present application, a possible implementation method is provided. The preset control algorithm mentioned in step S103 above may be a discretized PID (Proportion Integral Derivative) algorithm. Step S103 uses the preset control algorithm to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second type of outbound call agents at the next moment is controlled from the predicted number of idle seats to the preset value. Specifically, the method may include the following steps B1:

[0068] Step B1, using a discretized PID algorithm, calculates the deviation, integral deviation, and differential deviation of the number of idle seats of the second type of outbound call seats at the next moment from the predicted number of idle seats when the number is controlled at a preset value, and obtains the number of idle seats that needs to be adjusted.

[0069] The preset value here can be set according to actual needs, such as the preset value being 0, etc., which is not limited in this embodiment.

[0070] This embodiment uses a discretized PID algorithm to calculate the deviation, integral deviation, and differential deviation of the number of available seats for the second type of outbound call agents at the next moment from the predicted number of available seats when the number is controlled at a preset value. This calculates the number of available seats that needs to be adjusted, thereby achieving the purpose of effectively controlling the number of available seats.

[0071] In an embodiment of the present application, a possible implementation method is provided. In the above step B1, a discretized PID algorithm is used to calculate the deviation, integral deviation, and differential deviation of the number of available seats of the second type of outbound call agents at the next moment from the predicted number of available seats to be controlled at a preset value, and the number of available seats to be adjusted is obtained. Specifically, the method may include step B11:

[0072] Step B11: Using a discretized PID algorithm, assuming the sampling interval is T, at the next time t, when the number of available seats of the second-category outbound call agents at the next time t is controlled from the predicted number of available seats to the preset value, the deviation between the predicted number of available seats at the next time t and the preset value is e(k), the deviation between the predicted number of available seats at time tT and the preset value is e(k-1), the deviation between the predicted number of available seats at time t-2T and the preset value is e(k-2), and so on.

[0073] The integral deviation is e(k)+e(k-1)+e(k-2)+...+e(0);

[0074] The differential deviation is (e(k)-e(k-1)) / T;

[0075] The number of idle seats that need to be adjusted at the next moment t is u(k), and the number of idle seats that need to be adjusted at moment tT is u(k-1);

[0076]

[0077]

[0078] u(k)=u(k-1)+△u(k);

[0079] where increment ∆u(k) = α(e(k) - e(k-1)) + βe(k) + θ(e(k) - 2e(k-1) + e(k-2));

[0080] α is an input parameter, which indicates the desired ratio to compensate for the gap between the predicted number of available seats and the preset value. It is a value between 0 and 1.

[0081] β = α × T / Ti, Ti is the statistical interval of the integral deviation, which is an integer multiple of T;

[0082] θ = α × Td / T, where Td is the statistical interval of differential deviation and is an integer multiple of T.

[0083] This embodiment uses a PID control algorithm to calculate the number of idle seats that need to be adjusted in real time, controls the predicted number of idle seats to a preset value, and then calculates the outbound call parameters at the next moment based on the number of idle seats that need to be adjusted. The outbound call parameters at the current moment are adjusted to the outbound call parameters at the next moment, thereby achieving the purpose of reducing the loss of potential users when the first type of outbound call agents switch to the second type of outbound call agents, and shortening the waiting time interval between two calls for the second type of outbound call agents.

[0084] The embodiment of the present application provides a possible implementation method. Step S104 calculates the outbound call parameters at the next moment based on the number of idle seats that need to be adjusted at the next moment. The outbound call parameters here can be the number of outbound calls or the outbound call multiplier, etc. Therefore, the outbound call parameters at the next moment can be calculated based on the number of idle seats that need to be adjusted at the next moment using the following formula:

[0085] Number of outbound calls at the next moment = u(k) / (Current moment call loss rate × Transfer rate of potential users submitted to the waiting queue for transfer to the second-class outbound call agent × Connection rate of outbound calls initiated by the outbound call center and calls between the first-class outbound call agent and the user)

[0086] The outbound call multiplier at the next moment = the number of outbound calls at the next moment / 100.

[0087] This embodiment calculates the outbound call parameters at the next moment based on the number of idle agents that need to be adjusted, and adjusts the outbound call parameters at the current moment to the outbound call parameters at the next moment for the outbound call. This achieves the purpose of reducing the loss of potential users when the first type of outbound call agents switch to the second type of outbound call agents, and shortening the waiting time interval between two calls for the second type of outbound call agents.

[0088] The above introduces Figure 1 There are multiple implementation methods for each link of the embodiment shown. The outbound call control method of this embodiment will be further explained below through specific embodiments.

[0089] In this specific embodiment, the first type of outbound call agent is a robot, and the second type of outbound call agent is a human agent. Existing outbound call processes do not have a module for robot-human communication, and existing outbound call process calculation systems fail to adaptively adjust outbound call parameters, such as the number of outbound calls or the outbound call multiplier, based on set targets.

[0090] This application implementation example is based on artificial intelligence application scenarios in the field of online sales, such as Figure 2 As shown, first, the lead library includes multiple lead users, and the outbound call center communicates with the connected users through a robot. There is a connection rate here. For example, the outbound call center initiates 1,000 outbound calls, and 100 users are connected. At this time, the connection rate is 10%. This is just an example and does not limit the embodiments of the present application.

[0091] Next, users interested in purchasing insurance are placed in a waiting queue to be transferred to a human agent, who then speaks with the prospective user and guides them toward closing a purchase. In both AI and human models, excessive outbound calls can result in a lack of available human agents to speak with interested customers, resulting in call losses. Excessive outbound calls can lead to extended idle time for human agents, reducing work efficiency.

[0092] This embodiment uses the PID algorithm to normalize call loss and wait time into a variable X, namely the number of available human agents. The PID control algorithm is used to calculate and adjust the number of outbound calls or the outbound call ratio in real time, maintaining the value of variable X near the target value of 0. This achieves the goal of reducing lead loss during the robot-to-human transition stage and shortening the wait time between two calls for human agents.

[0093] like Figure 2 As shown, in the outbound call system of AI and humans, the fluctuation of the number of outbound calls, which represents the outbound call volume, will cause two parameters to change, and the call loss or waiting time will increase or decrease. Specifically, when the call loss is high, the waiting time is long, and when the call loss is low, the waiting time is short. This embodiment adopts the idea of ​​PID control algorithm, normalizes the call loss and the waiting time into an indicator X, that is, the predicted number of idle seats of manual agents at the next moment, and uses the PID algorithm formula to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second type of outbound call agents at the next moment is controlled from the predicted idle number to the preset value, and then calculates the number of outbound calls or the outbound call ratio at the next moment based on the number of idle seats that need to be adjusted, thereby achieving adaptive adjustment of the number of outbound calls or the outbound call ratio.

[0094] 1. Variable Normalization

[0095] Set the normalized variable to X, the fixed number of outbound calls or outbound call rate adjustment time period T, and record the number of call losses and the length of time the agent waits within time T as X1 and X2, respectively. Based on historical data or actual specific requirements, the default number of call losses Y1 and waiting time Y2 within time T can be set.

[0096] If 0<=X1<=Y1, the number of call losses is within a reasonable range, and the waiting time X2 is used as the representation of X; if 0<=X2<=Y2, the waiting time is within a reasonable range, and the number of call losses is used as the representation of X; if both X1 and X2 are within a reasonable range, then X=0, that is, the system is in an ideal state.

[0097] When using X1 and X2 to represent X, you can use a corresponding function to increase the impact of the variable in a certain direction according to specific needs. For example, if the waiting time indicator is relatively more important, you can use an exponential function to represent the variable X = f(X2-Y2) to increase the penalty imposed by the increased waiting time on the current system state. This embodiment inputs the current waiting time and / or call loss volume into a pre-trained outbound call idle agent prediction model to predict the number of idle seats for manual agents at the next moment, and obtains the predicted number of idle seats for manual agents at the next moment as the normalized variable X.

[0098] 2.PID algorithm

[0099] In this embodiment, the number of idle seats of the manual agents needs to be maintained at a preset value, that is, a target value of 0, through the PID algorithm.

[0100] Using the discretized PID algorithm, assuming that the sampling time interval is T, then at the next time t, when the number of idle seats of the next time t is controlled from the predicted number of idle seats to the preset value:

[0101] Deviation e(k) = 0 (target value) - predicted number of available seats for human agents at the next moment

[0102] The purpose of introducing deviation is to know how much the predicted number of idle seats differs from the target value, and whether to increase or decrease the number of idle seats.

[0103] The differential deviation is (e(k)-e(k-1)) / T

[0104] The differential deviation indicates how big the difference is between the current number of idle seats and the last number of idle seats. In the ideal system state, the e(k) calculated each time is 0, that is, the differential should be 0, so the differential deviation is introduced.

[0105] The integral deviation is e(k)+e(k-1)+e(k-2)+...+e(0)

[0106] Integral deviation refers to the call loss value or waiting time generated during the historical system operation. By introducing historical data weights, the system state is shifted to another state.

[0107] The number of idle seats that need to be adjusted at the next moment t is u(k), and the number of idle seats that need to be adjusted at moment tT is u(k-1);

[0108]

[0109]

[0110] u(k)=u(k-1)+△u(k);

[0111] where increment ∆u(k) = α(e(k) - e(k-1)) + βe(k) + θ(e(k) - 2e(k-1) + e(k-2));

[0112] α is an input parameter, which indicates the desired ratio to compensate for the gap between the predicted number of available seats and the preset value. It is a value between 0 and 1.

[0113] β = α × T / Ti, Ti is the statistical interval of the integral deviation, which is an integer multiple of T;

[0114] θ = α × Td / T, where Td is the statistical interval of differential deviation and is an integer multiple of T.

[0115] Number of outbound calls at the next moment = u(k) / (Current moment call loss rate × Transfer rate of potential users submitted to the waiting queue for transfer to the second-class outbound call agent × Connection rate of outbound calls initiated by the outbound call center and calls between the first-class outbound call agent and the user)

[0116] The outbound call multiplier at the next moment = the number of outbound calls at the next moment / 100.

[0117] Based on the framework of a robot and human outbound call system, this embodiment normalizes call loss and wait time into a single indicator and employs the PID algorithm to perform adaptive outbound call adjustments in the outbound call control system. This significantly improves outbound call efficiency, increases the human agent call rate by 30%, and reduces the call loss rate from 20-30% to approximately 10%.

[0118] It should be noted that the order of execution of the steps in the above embodiments does not necessarily imply a specific order of execution. The order of execution of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. In practical applications, all possible implementation methods described above can be combined in any manner to form possible embodiments of the present application, and will not be described in detail here.

[0119] Based on the outbound call control methods provided in the above embodiments and based on the same inventive concept, an embodiment of the present application further provides an outbound call control device.

[0120] Figure 3 This is a structural diagram of the outbound call control device provided by the embodiment of the present application. Figure 3 As shown, the outbound call control device can be applied to an outbound call center to communicate with a connected user through a first type of outbound call agent. When it is determined that the connected user meets the set conditions and is an intended user, the intended user is submitted to a waiting queue to be transferred to a second type of outbound call agent, and the second type of outbound call agent communicates with the intended user. Specifically, it may include an acquisition module 310, a prediction module 320, a control module 330, a calculation module 340 and an adjustment module 350.

[0121] An acquisition module 310 is configured to obtain a waiting time for transferring a user from a waiting queue to a second-class outbound call agent for a call and / or a call loss amount caused by a lack of a second-class outbound call agent available for transfer.

[0122] Prediction module 320, configured to input the current waiting time and / or call loss into a pre-trained outbound call available agent prediction model, predict the number of available agents of the second category of outbound call agents at the next moment, and obtain the predicted number of available agents of the second category of outbound call agents at the next moment;

[0123] The control module 330 is configured to use a preset control algorithm to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second type of outbound call agents at the next moment is controlled from the predicted number of idle seats to the preset value;

[0124] Calculation module 340, used to calculate outbound call parameters at the next moment according to the number of idle seats that need to be adjusted at the next moment;

[0125] The adjustment module 350 is configured to adjust the outbound call parameters at the current moment to the outbound call parameters at the next moment for making an outbound call.

[0126] A possible implementation method is provided in the embodiment of the present application, such as Figure 4 As shown above Figure 3 The illustrated apparatus may further include a training module 410 for:

[0127] Build an initial outbound call idle agent prediction model;

[0128] The historical waiting time and / or call loss amount, as well as the number of available seats of the second type of outbound call agents corresponding to the historical waiting time and / or call loss amount, are input into an initial outbound call available seat prediction model, and the initial outbound call available seat prediction model is trained to obtain a trained outbound call available seat prediction model.

[0129] An embodiment of the present application provides a possible implementation method, wherein the preset control algorithm includes a discretized PID algorithm; and the control module 330 is further configured to:

[0130] The discretized PID algorithm is used to calculate the deviation, integral deviation, and differential deviation of the number of idle seats of the second type of outbound call seats at the next moment from the predicted number of idle seats when it is controlled at the preset value, and the number of idle seats that needs to be adjusted is obtained.

[0131] An embodiment of the present application provides a possible implementation method, wherein the control module 330 is further configured to:

[0132] Using the discretized PID algorithm, assuming that the sampling time interval is T, then at the next time t, when the number of idle seats of the second type of outbound call seats at the next time t is controlled from the predicted idle seats to the preset value, the deviation between the predicted idle seats at the next time t and the preset value is e(k), the deviation between the predicted idle seats at time tT and the preset value is e(k-1), the deviation between the predicted idle seats at time t-2T and the preset value is e(k-2), and so on;

[0133] The integral deviation is e(k)+e(k-1)+e(k-2)+...+e(0);

[0134] The differential deviation is (e(k)-e(k-1)) / T;

[0135] The number of idle seats that need to be adjusted at the next moment t is u(k), and the number of idle seats that need to be adjusted at moment tT is u(k-1);

[0136]

[0137]

[0138] u(k)=u(k-1)+△u(k);

[0139] where increment ∆u(k) = α(e(k) - e(k-1)) + βe(k) + θ(e(k) - 2e(k-1) + e(k-2));

[0140] α is an input parameter, which indicates the desired ratio to compensate for the gap between the predicted number of available seats and the preset value. It is a value between 0 and 1.

[0141] β = α × T / Ti, Ti is the statistical interval of the integral deviation, which is an integer multiple of T;

[0142] θ = α × Td / T, where Td is the statistical interval of differential deviation and is an integer multiple of T.

[0143] A possible implementation is provided in an embodiment of the present application, wherein the calculation module 340 is further configured to:

[0144] The outbound call parameters for the next moment are calculated using the following formula based on the number of idle seats that need to be adjusted at the next moment:

[0145] Number of outbound calls at the next moment = u(k) / (Current moment call loss rate × Transfer rate of potential users submitted to the waiting queue for transfer to the second-class outbound call agent × Connection rate of outbound calls initiated by the outbound call center and calls between the first-class outbound call agent and the user)

[0146] The outbound call multiplier at the next moment = the number of outbound calls at the next moment / 100.

[0147] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a processor and a memory, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the outbound call control method of any one of the above embodiments.

[0148] In an exemplary embodiment, an electronic device is provided, such as Figure 5 As shown, Figure 5 The electronic device 500 shown includes a processor 501 and a memory 503. The processor 501 and the memory 503 are connected, for example, via a bus 502. Optionally, the electronic device 500 may further include a transceiver 504. It should be noted that in actual applications, the number of transceivers 504 is not limited to one, and the structure of the electronic device 500 does not constitute a limitation on the embodiments of the present application.

[0149] The processor 501 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 501 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0150] The bus 502 may include a path for transmitting information between the above components. The bus 502 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 502 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0151] The memory 503 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0152] The memory 503 is used to store computer program codes for executing the solution of the present application, and the execution is controlled by the processor 501. The processor 501 is used to execute the computer program codes stored in the memory 503 to implement the contents shown in the above method embodiments.

[0153] Among them, electronic devices include but are not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The electronic device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0154] Based on the same inventive concept, an embodiment of the present application further provides a storage medium, in which a computer program is stored, wherein the computer program is configured to execute the outbound call control method of any one of the above embodiments when running.

[0155] Those skilled in the art will clearly understand that the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the aforementioned method embodiments, and for the sake of brevity, they will not be further described here.

[0156] Those skilled in the art will appreciate that the technical solution of the present application, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes a number of program instructions for causing an electronic device (e.g., a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application when the program instructions are executed. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0157] Alternatively, all or part of the steps of implementing the aforementioned method embodiments may be accomplished by hardware related to program instructions (such as electronic devices such as personal computers, servers, or network devices), and the program instructions may be stored in a computer-readable storage medium. When the program instructions are executed by a processor of an electronic device, the electronic device executes all or part of the steps of the methods described in the various embodiments of the present application.

[0158] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that, within the spirit and principles of the present application, they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate from the protection scope of the present application.

Claims

1. An outbound call control method, characterized in that: The method is applied to a scenario where an outbound call center uses a first-class outbound call agent to communicate with a connected user. When the connected user is determined to meet set conditions and is an intended user, the intended user is submitted to a waiting queue to be transferred to a second-class outbound call agent, and the second-class outbound call agent communicates with the intended user. The method includes: Obtain the current waiting time for transferring the intended user from the waiting queue to a second-class outbound call agent and the call loss amount caused by the lack of a second-class outbound call agent available for transfer. Input the current waiting time and call loss into a pre-trained outbound call idle agent prediction model to predict the number of idle seats of the second category of outbound call agents at the next moment, thereby obtaining the predicted number of idle seats of the second category of outbound call agents at the next moment; A preset control algorithm is used to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second type of outbound call seats at the next moment is controlled from the predicted number of idle seats to the preset value; Calculate the outbound call parameters for the next moment based on the number of idle seats that need to be adjusted at the next moment; Adjust the outbound call parameters at the current moment to the outbound call parameters at the next moment to make an outbound call; The preset control algorithm includes a discretized PID algorithm; the method of using the preset control algorithm to calculate the number of idle seats that need to be adjusted when the number of idle seats of the second type of outbound call seats at the next moment is controlled from the predicted number of idle seats to the preset value includes: Using the discretized PID algorithm, assuming that the sampling time interval is T, then at the next time t, when the number of idle seats of the second type of outbound call seats at the next time t is controlled from the predicted idle seats to the preset value, the deviation between the predicted idle seats at the next time t and the preset value is e(k), the deviation between the predicted idle seats at time tT and the preset value is e(k-1), the deviation between the predicted idle seats at time t-2T and the preset value is e(k-2), and so on; The integral deviation is e(k)+e(k-1)+e(k-2)+...+e(0); The differential deviation is (e(k)-e(k-1)) / T; The number of idle seats that need to be adjusted at the next moment t is u(k), and the number of idle seats that need to be adjusted at moment tT is u(k-1); u(k)=u(k-1)+△u(k); where increment ∆u(k) = α(e(k) - e(k-1)) + βe(k) + θ(e(k) - 2e(k-1) + e(k-2)); α is an input parameter, which indicates the desired ratio to compensate for the gap between the predicted number of available seats and the preset value. It is a value between 0 and 1. β = α × T / Ti, Ti is the statistical interval of the integral deviation, which is an integer multiple of T; θ = α × Td / T, where Td is the statistical interval of differential deviation and is an integer multiple of T; Here, the outbound call parameters at the next moment include the number of outbound calls at the next moment, which is u(k) / (the call loss rate at the current moment × the transfer rate of the intended user to the waiting queue to be transferred to the second type of outbound call agent × the connection rate of the outbound call initiated by the outbound call center and the conversation between the first type of outbound call agent and the user).

2. The method according to claim 1, characterized in that Train the outbound call idle agent prediction model by following these steps: Build an initial outbound call idle agent prediction model; The historical waiting time and call loss volume, as well as the number of idle seats of the second type of outbound call agents corresponding to the historical waiting time and call loss volume, are input into an initial outbound idle seat prediction model, and the initial outbound idle seat prediction model is trained to obtain a trained outbound idle seat prediction model.

3. The method according to claim 1, characterized in that The outbound call parameters for the next moment are calculated using the following formula based on the number of idle seats that need to be adjusted at the next moment, including: The outbound call multiplier at the next moment = the number of outbound calls at the next moment / 100.

4. An outbound call control device, characterized in that: The device is applied to a scenario where an outbound call center uses a first-class outbound call agent to communicate with a connected user. When the connected user is determined to meet set conditions and is an intended user, the intended user is submitted to a waiting queue to be transferred to a second-class outbound call agent, and the second-class outbound call agent then communicates with the intended user. The device includes: An acquisition module is used to obtain the waiting time for transferring the intended user from the waiting queue to the second-class outbound call agent for a call at the current moment and the call loss amount caused by the lack of a second-class outbound call agent available for transfer at the current moment; A prediction module is used to input the current waiting time and call loss into a pre-trained outbound call idle agent prediction model to predict the number of idle seats of the second category of outbound call agents at the next moment, thereby obtaining the predicted number of idle seats of the second category of outbound call agents at the next moment; a control module for calculating, by using a preset control algorithm, the number of idle seats that needs to be adjusted when the number of idle seats of the second type of outbound call agents at the next moment is controlled from the predicted number of idle seats to the preset value; A calculation module is used to calculate the outbound call parameters at the next moment according to the number of idle seats that need to be adjusted at the next moment; An adjustment module, used to adjust the outbound call parameters at the current moment to the outbound call parameters at the next moment for making an outbound call; The preset control algorithm includes a discretized PID algorithm; the control module is further configured to: Using the discretized PID algorithm, assuming that the sampling time interval is T, then at the next time t, when the number of idle seats of the second type of outbound call seats at the next time t is controlled from the predicted idle seats to the preset value, the deviation between the predicted idle seats at the next time t and the preset value is e(k), the deviation between the predicted idle seats at time tT and the preset value is e(k-1), the deviation between the predicted idle seats at time t-2T and the preset value is e(k-2), and so on; The integral deviation is e(k)+e(k-1)+e(k-2)+...+e(0); The differential deviation is (e(k)-e(k-1)) / T; The number of idle seats that need to be adjusted at the next moment t is u(k), and the number of idle seats that need to be adjusted at moment tT is u(k-1); u(k)=u(k-1)+△u(k); where increment ∆u(k) = α(e(k) - e(k-1)) + βe(k) + θ(e(k) - 2e(k-1) + e(k-2)); α is an input parameter, which indicates the desired ratio to compensate for the gap between the predicted number of available seats and the preset value. It is a value between 0 and 1. β = α × T / Ti, Ti is the statistical interval of the integral deviation, which is an integer multiple of T; θ = α × Td / T, where Td is the statistical interval of differential deviation and is an integer multiple of T; Here, the outbound call parameters at the next moment include the number of outbound calls at the next moment, which is u(k) / (the call loss rate at the current moment × the transfer rate of the intended user to the waiting queue to be transferred to the second type of outbound call agent × the connection rate of the outbound call initiated by the outbound call center and the conversation between the first type of outbound call agent and the user).

5. The device according to claim 4, characterized in that Also includes training modules for: Build an initial outbound call idle agent prediction model; The historical waiting time and call loss volume, as well as the number of idle seats of the second type of outbound call agents corresponding to the historical waiting time and call loss volume, are input into an initial outbound idle seat prediction model, and the initial outbound idle seat prediction model is trained to obtain a trained outbound idle seat prediction model.

6. An electronic device, characterized in that: The invention comprises a processor and a memory, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the outbound call control method according to any one of claims 1 to 3.

7. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the outbound call control method according to any one of claims 1 to 3 when running.

Images