Flow control method for intelligent voice outbound

Abstract

The invention discloses a flow control method for intelligent voice outbound, and the method comprises the steps: introducing a statistical classification algorithm in the aspect of an outbound flow control algorithm, taking various outbound flow control factors as events, and calculating the probability that the outbound flow reaches a certain level under a certain condition; flow upper limit calculation and flow adjustment dynamic decision are introduced for flow control of intelligent voice outbound; and an outbound traffic controller based on an outbound traffic control algorithm is introduced to complete a full-intelligent outbound system. According to the method, accurate prediction and intelligent adjustment of the flow of the intelligent outbound system are realized, dynamic balance between the outbound task flow and available resources of the outbound voice platform is realized, the resource utilization rate of the outbound voice platform is maximized, and the capacity of the intelligent outbound system is further released, the cost is reduced and the efficiency is improved on the premise that the outbound resources are kept unchanged.

Description

Technical Field

[0001] The present invention relates to the field of traffic control in intelligent outbound call scenarios, and in particular to a traffic control method for intelligent voice outbound calls. Background Art

[0002] The outbound call system relies on the voice platform to initiate calls. When the voice platform makes an outbound call, it first pulls a set of pending outbound call tasks from the business system before executing the call. The execution of these tasks consumes outbound call resources. Because a single outbound call task can be executed in a variety of situations (customer answers, customer rings for a long time but doesn't answer, customer rejects the call, etc.), there is uncertainty between the influx of outbound call tasks pulled from the voice platform and the amount of inbound and outbound call resources occupied by the voice platform when multiple rounds of outbound call tasks are executed:

[0003] 1. When the incoming outbound call traffic exceeds the number of available outbound call resources on the current voice platform, the excess outbound call tasks will fail due to a lack of outbound call resources, and the outbound call list will be invalidated.

[0004] 2. When the incoming outbound call task traffic is less than the number of available inbound and outbound call resources on the current voice platform, the number of pulled outbound call tasks is insufficient, resulting in idle outbound call resources and a sharp drop in the number of outbound call tasks executed per unit time.

[0005] During an outbound call execution cycle, due to indiscriminate flow control, the above two situations occur repeatedly and alternately, resulting in a reduction in the production capacity of the outbound call system during the execution cycle, accompanied by the problem of a large number of invalid outbound call lists, which directly affects the outbound call system's high-efficiency and high-quality execution of outbound call tasks. Summary of the Invention

[0006] In view of the above problems, the present invention is proposed to provide a flow control method for intelligent voice outbound calls that overcomes the above problems or at least partially solves the above problems.

[0007] According to one aspect of the present invention, a flow control method for intelligent voice outbound calls is provided, the flow control method comprising:

[0008] In terms of outbound call flow control algorithms, a statistical classification algorithm is introduced, which treats various outbound call flow control factors as events and calculates the probability of outbound call flow reaching a certain level under certain conditions;

[0009] Introducing traffic limit calculation and dynamic traffic adjustment decision-making for traffic control of intelligent voice outbound calls;

[0010] An outbound call flow controller based on an outbound call flow control algorithm is introduced to complete a fully intelligent outbound call system.

[0011] Optionally, the calculation under certain conditions specifically includes: time, and outbound call execution result.

[0012] Optionally, in terms of the outbound call flow control algorithm, introducing a statistical classification algorithm specifically includes:

[0013] Flow-related factors were defined as variables;

[0014] Construct an outbound call flow control algorithm model;

[0015] Flow control decision calculation.

[0016] Optionally, the traffic-related factors are defined as variables including:

[0017] Call success rate factor: Set the call success rate as an event A and use historical data to determine the prior probability P(A) of the call success rate;

[0018] Customer response rate factor: Define the customer response rate as event C, and the prior probability P(C) is obtained from historical data;

[0019] The type of customer affects the customer response rate. The probability of customer response P(C|D) for different customer types is calculated. The probability data is used as the conditional probability in the outbound call flow control algorithm.

[0020] Optionally, the building of the outbound call flow control algorithm model specifically includes:

[0021] Construct vector data networks to represent the relationships between factors;

[0022] Call time affects call success rate. Customers in a sensitive period of multiple outbound calls will affect the customer answer rate. The call success rate and customer answer rate will jointly affect the effect of outbound call traffic.

[0023] In a Bayesian network, nodes represent events and edges represent causal relationships between events;

[0024] Through the network, analyze and calculate the flow control related probabilities under different combinations of conditions.

[0025] Optionally, the flow control decision calculation specifically includes:

[0026] Calculate the traffic limit;

[0027] Dynamically adjust the flow control strategy of the outbound call system.

[0028] Optionally, the calculation of the flow rate upper limit specifically includes:

[0029] Determine an ideal call success rate and customer response rate based on business goals;

[0030] Use the outbound traffic control algorithm to calculate the probability of meeting the requirements under different conditions;

[0031] Assume that event B is a known call time between 9 and 10 am, and event D is a customer who is within a sensitive period for multiple outbound calls. Calculate the call success rate P(A|B) and customer response rate P(C|D). Consider these two probabilities to determine whether to increase outbound call traffic.

[0032] If, under existing conditions, both the call success rate and the customer response rate are greater than the set threshold, the upper limit of outbound call traffic is increased; conversely, if the probability is less than the set threshold, the upper limit of outbound call traffic is lowered.

[0033] Optionally, the dynamic adjustment of the traffic control strategy of the outbound call system specifically includes:

[0034] During system operation, outbound call data is collected in real time;

[0035] New data is used to update prior probabilities and conditional probabilities;

[0036] If the newly collected data shows that the call success rate has dropped in a certain time period, the flow control parameters for this time period will be recalculated;

[0037] By continuously updating probability estimates, the outbound call system's flow control strategy is dynamically adjusted to adapt to the ever-changing actual situation.

[0038] The present invention provides a flow control method for intelligent outbound voice calls. The flow control method includes: introducing a statistical classification algorithm into the outbound flow control algorithm, treating various outbound flow control factors as events, and calculating the probability of outbound flow reaching a certain level under certain conditions; introducing flow upper limit calculation and dynamic flow adjustment decision-making for intelligent outbound voice call flow control; and introducing an outbound flow controller based on the outbound flow control algorithm to complete a fully intelligent outbound call system. This method achieves accurate prediction and intelligent adjustment of flow for the intelligent outbound call system, achieves a dynamic balance between outbound call task flow and available resources of the outbound voice platform, maximizes the resource utilization of the outbound voice platform, and further releases the production capacity of the intelligent outbound call system while maintaining the outbound call resources, thereby reducing costs and increasing efficiency.

[0039] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 A flow chart of a method for controlling traffic in an intelligent voice outbound call is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure 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 disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0043] The terms "comprises" and "comprising" and any variations thereof in the description, embodiments, claims and drawings of the present invention are intended to cover non-exclusive inclusions, for example, including a series of steps or units.

[0044] The technical solution of the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0045] The overall technical solution includes two aspects: in terms of outbound call flow control algorithm, a statistical classification algorithm is introduced to regard various outbound call flow control factors as events, and the probability of outbound call flow reaching a certain level under certain conditions (such as different time, outbound call execution results, etc.) is calculated. Based on this, the flow upper limit calculation and flow adjustment dynamic decision-making are introduced to propose a flow control algorithm for intelligent voice outbound calls; in terms of implementation method, an outbound call flow controller based on the outbound call flow control algorithm is introduced to complete the specific technical implementation method of the fully intelligent outbound call system.

[0046] 1. Outbound call flow control algorithm

[0047] In the traffic control scenario, various outbound call traffic control factors can be regarded as events. The above method can be used to calculate the probability of outbound call traffic reaching a certain level under certain conditions (such as different time, outbound call execution results, etc.).

[0048] Traffic-related factors are defined as variables (events)

[0049] Call success rate factor: Consider the call success rate as an event (A). Historical data can be used to determine the prior probability of the call success rate (P(A)). For example, the number of successful calls in the past week divided by the total number of calls gives the basic probability of the call success rate. Factors influencing the call success rate, such as the time of the call (set as event (B)), are also considered. By calculating the probability of call success in different time periods (P(A|B)), for example, the call success rate between 9:00 AM and 10:00 AM is 60% and the call success rate between 7:00 PM and 8:00 PM is 40% (P(A|B)). This data can be used for subsequent calculations.

[0050] Customer response rate factor: Define the customer response rate as an event (C). The prior probability (P(C)) is obtained from historical data, such as the percentage of customer responses among all outbound calls. The customer type (defined as the event (D)) affects the customer response rate. For example, the response rate may be higher for first-time outbound customers, while the response rate may be lower for repeat outbound customers. Therefore, the probability of customer response (P(C|D)) for different customer types is calculated. These probabilities serve as conditional probabilities in the outbound traffic control algorithm.

[0051] Constructing an outbound call flow control algorithm model

[0052] A vector data network is constructed to represent the relationships between these factors. For example, call duration affects call success rate, and customers within a sensitive period of multiple outbound calls will affect their answer rate. These two factors, in turn, jointly influence outbound traffic flow. In a Bayesian network, nodes represent events (such as call success rate and customer response rate), and edges represent causal relationships between events. This network allows for more intuitive analysis and calculation of traffic control probabilities under different combinations of conditions.

[0053] Flow control decision calculation

[0054] Traffic Cap Calculation: Determine an ideal call success rate and customer response rate based on business objectives. For example, a business requirement might be a call success rate of at least 50% and a customer response rate of at least 30%. The outbound traffic control algorithm calculates the probability of meeting these requirements under different conditions (e.g., specific time of day and customer type combination). Assuming the call time is between 9:00 AM and 10:00 AM (event (B)), and the customer type is within a sensitive period for multiple outbound calls (event (D)), calculate the call success rate (P(A|B)) and customer response rate (P(C|D)) at this time. These two probabilities are then considered to determine whether outbound traffic can be increased. If both the call success rate and customer response rate are high under these conditions, the outbound traffic cap can be appropriately increased. Conversely, if the probabilities are low, the outbound traffic cap should be lowered.

[0055] Dynamic traffic adjustment decisions: During system operation, outbound call data is collected in real time, including call success and customer response information. This new data is used to update prior probabilities and conditional probabilities. For example, if newly collected data shows a decrease in call success rates during a certain time period, the traffic control parameters for that time period are recalculated. By continuously updating probability estimates, the outbound call system's traffic control strategy is dynamically adjusted to adapt to changing conditions.

[0056] Implementation method: Introduce a traffic controller based on the outbound traffic control algorithm between the outbound call task pool and the outbound call platform, and achieve a dynamic balance between the outbound call task traffic and the available resources of the outbound voice platform through traffic control decision calculation (traffic upper limit calculation and dynamic traffic adjustment decision).

[0057] Beneficial effects: The present invention comprehensively applies statistical classification algorithms to achieve accurate prediction and intelligent adjustment of traffic for intelligent outbound call systems, realize dynamic balance between outbound call task traffic and available resources of the outbound voice platform, maximize the resource utilization of the outbound voice platform, and further release the production capacity of the intelligent outbound call system while keeping the outbound resources unchanged, thereby reducing costs and increasing efficiency.

[0058] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A flow control method for intelligent voice outbound calling, characterized in that: The flow control method comprises: In terms of outbound call flow control algorithms, a statistical classification algorithm is introduced, which treats various outbound call flow control factors as events and calculates the probability of outbound call flow reaching a certain level under certain conditions; Introducing traffic limit calculation and dynamic traffic adjustment decision-making for traffic control of intelligent voice outbound calls; An outbound call flow controller based on an outbound call flow control algorithm is introduced to complete a fully intelligent outbound call system.

2. A flow control method for intelligent voice outbound calling according to claim 1, characterized in that: The calculation under certain conditions specifically includes: time, and outbound call execution result.

3. A flow control method for intelligent voice outbound calling according to claim 1, characterized in that: In terms of the outbound call flow control algorithm, the introduction of the statistical classification algorithm specifically includes: Flow-related factors were defined as variables; Construct an outbound call flow control algorithm model; Flow control decision calculation.

4. A flow control method for intelligent voice outbound calling according to claim 3, characterized in that: The traffic-related factors are defined as variables and specifically include: Call success rate factor: Set the call success rate as an event A and use historical data to determine the prior probability P(A) of the call success rate; Customer response rate factor: Define the customer response rate as event C, and the prior probability P(C) is obtained from historical data; The type of customer affects the customer response rate. The probability of customer response P(C|D) for different customer types is calculated. The probability data is used as the conditional probability in the outbound call flow control algorithm.

5. A flow control method for intelligent voice outbound calling according to claim 3, characterized in that: The construction of the outbound call flow control algorithm model specifically includes: Construct vector data networks to represent the relationships between factors; Call time affects call success rate. Customers in a sensitive period of multiple outbound calls will affect the customer answer rate. The call success rate and customer answer rate will jointly affect the effect of outbound call traffic. In a Bayesian network, nodes represent events and edges represent causal relationships between events; Through the network, analyze and calculate the flow control related probabilities under different combinations of conditions.

6. A flow control method for intelligent voice outbound calling according to claim 3, characterized in that: The flow control decision calculation specifically includes: Calculate the traffic limit; Dynamically adjust the flow control strategy of the outbound call system.

7. A flow control method for intelligent voice outbound calling according to claim 6, characterized in that: The calculation flow upper limit specifically includes: Determine an ideal call success rate and customer response rate based on business goals; Use the outbound traffic control algorithm to calculate the probability of meeting the requirements under different conditions; Assume that event B is a known call time between 9 and 10 am, and event D is a customer who is within a sensitive period for multiple outbound calls. Calculate the call success rate P(A|B) and customer response rate P(C|D). Consider these two probabilities to determine whether to increase outbound call traffic. If, under existing conditions, both the call success rate and the customer response rate are greater than the set threshold, the upper limit of outbound call traffic is increased; conversely, if the probability is less than the set threshold, the upper limit of outbound call traffic is lowered.

8. A flow control method for intelligent voice outbound calling according to claim 6, characterized in that: The dynamic adjustment of the traffic control strategy of the outbound call system specifically includes: During system operation, outbound call data is collected in real time; New data is used to update prior probabilities and conditional probabilities; If the newly collected data shows that the call success rate has dropped in a certain time period, the flow control parameters for this time period will be recalculated; By continuously updating probability estimates, the outbound call system's flow control strategy is dynamically adjusted to adapt to the ever-changing actual situation.

Images