Intelligent emergency processing method and system capable of self-adapting to telephone traffic outbreak
Through intelligent emergency response methods and automation technology, we predict the surge in traffic volume and take emergency measures, which solves the problem of rapid traffic volume in call center systems in emergencies, improves the connection rate and operational efficiency, and reduces customers complaint.
Patent Information
- Application Number
- CN202411153745.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-12-06
AI Technical Summary
The call center system is facing a surge in traffic volume due to emergencies, resulting in excessive waiting time for customers, declining service quality, and tight resource of manual agents, making it difficult to meet customer demands.
Adaptable intelligent emergency response method for traffic surges, through intelligent interception strategies, predict traffic volume, automation technology and remote seats, reduce customer transfer, improve manual connection rate, avoid secondary congestion, and Actively notify customers of fault information through SMS, smart notifications, etc.
It effectively improves the processing speed and resource utilization rate of manual seats, ensures the connection rate, reduces customer complaints, improves the operational efficiency and service level of the customer service center, and solves the problem of tight agent resources.
Smart Images

Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of intelligent call systems, and in particular relates to an intelligent emergency processing method and system capable of adaptively handling sudden increase in call traffic. Background Art
[0002] The call center system is a comprehensive customer service solution that integrates multiple technologies and functions. It helps enterprises improve customer satisfaction and loyalty and achieve sustainable development by providing efficient, convenient and personalized services. The stable and normal operation of the call center system is crucial, but every call center system will have a very common and important problem, which is "the surge in call volume". How to deal with the surge in call volume of the call center system will be very important, because it is directly related to customer experience, customer satisfaction and the operational efficiency of the enterprise. When the call volume surges, if these calls cannot be handled effectively, it may cause customers to wait too long, the service quality will decline, and even customers may be lost. For example, a gas customer service hotline, this system is designed to provide users with one-stop gas services, including business handling, business consultation, fault reporting, after-sales service, cost inquiry, hidden danger investigation, complaint handling and other functions. In the daily use of gas, affected by equipment failures, severe natural disasters, industry hotspots, urban engineering construction and other emergencies, the typical feature of the gas customer service system is that the call volume fluctuates greatly. According to incomplete statistics, when an emergency occurs, the peak call period is between 10 and 20 times the normal period, which is much higher than the daily call volume. For example, on a certain day in April 2024, a city experienced severe weather such as hail and continuous heavy rain, causing multiple gas equipment failures in the city. The system's incoming calls on that day reached more than 50,000, resulting in a significant drop in the level of manual telephone connections and a sharp increase in customer dissatisfaction and complaints. How to change call routing and strategies in a timely manner, limit the occupation of seat resources by calls in emergency areas, meet the development of customer service business in non-emergency areas, and on this basis, effectively use self-service voice (IVR), SMS, intelligent notifications, messages and other resources to effectively respond to telephone demands in emergency areas, is of great significance to improving the operational efficiency and service level of the customer service center. Summary of the invention
[0003] The purpose of the present invention is to provide an intelligent emergency handling method and system that can adapt to sudden increase in traffic, so as to solve the sharp contradiction between satisfying customer demands as much as possible and solving the tense contradiction between artificial seat resources in emergencies.
[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows: An intelligent emergency handling method capable of adaptively handling a surge in traffic, comprising: ① After the user dials the customer service hotline and enters the IVR process, the welcome message is played, and the system determines whether the intelligent interception strategy for traffic surge is activated; ② If the traffic surge intelligent interception strategy has been activated, determine whether the customer is from the fault area and proceed to step ③; if the traffic surge intelligent interception strategy has not been activated, continue to the subsequent IVR process; ③ If the customer is from the fault area, determine whether it is a repeated call. If it is a repeated call, the customer will be given priority to enter the manual queue or enter step ④; if it is the first call, push the fault voice to the customer. If it meets the customer's needs, the customer will hang up and reduce the call volume, otherwise continue to follow the subsequent IVR process; if the customer is from a non-fault area, guide the customer to enter the automatic query function to reduce the call saturation; ④ In extreme cases, when the human agent cannot continue to accept new customers, the customer will be directly guided to the message module or a voice notification will be pushed to the customer to inform the customer that the human agent is busy, the current gas area has a fault, and how long it is expected to be restored; at the same time, a text message will be pushed to the customer to inform the customer that the system is busy, the current gas area has a fault, and how long it is expected to be restored; ⑤ When the saturation of manual call traffic remains high, continue to consider adding seats, adding remote seats, or even immediately connecting remote mobile seats to ensure call traffic.
[0005] In the present invention, the most important thing about the intelligent interception strategy for sudden increase in call traffic is to predict the sudden increase in call traffic, foresee it in advance, and make correct response measures in advance. Through emergency measures including text messages, intelligent notifications, messages, and pre-sent text messages and other processing measures, the number of customers transferring to manual services can be reduced, secondary congestion can be avoided, the processing speed of manual seats can be improved, and the manual connection rate can be guaranteed without causing customer complaints.
[0006] As a further explanation of the method of the present invention, the activation of the intelligent interception strategy for traffic surge is based on the traffic saturation forecast data for the next 24 hours. When the predicted traffic volume exceeds the online seats by more than 2 times, the interception strategy needs to be activated in time.
[0007] As a further illustration of the method of the present invention, the traffic saturation prediction data for the next 24 hours is obtained by the following steps: (1) Select the historical traffic data of the last three years for statistics, and calculate the basic traffic volume M1, M2, M3, ..., M24 at 24 hours a day under the condition of no faults; (2) Select the historical call data of the past three years, analyze and count the call problem types from the dimensions of hot events, customer system failures, extreme weather, and sudden gas failures, and use mathematical statistical models to calculate and deduce the corresponding impact value mean parameters as follows: the impact parameter of industry hot events on call traffic is α, the impact parameter of customer system failure events on call traffic is β, the impact parameter of extreme special weather events on call traffic is γ, and the number of lines affected by sudden failure events is N and the impact parameter is δ; (3) Extract the average traffic volume at 24 hours per day in the last three months for comparison, and determine the value of the percentage increase / decrease of the average traffic volume at 24 hours per day over the same period λ, which are marked as λ1, λ2, λ3, ..., λ24 respectively; (4) Based on the corresponding forecast parameters and the percentage increase / decrease of the hourly average traffic volume in the corresponding month of the previous year, the traffic load volume at 24 hours on the day is predicted to be S1, S2, S3, ..., S24; S1=M1*(1+α+β+γ+δ*N)+λ1; S2=M2*(1+α+β+γ+δ*N)+λ2; S3=M3*(1+α+β+γ+δ*N)+λ3; ………… S24=M24*(1+α+β+γ+δ*N)+λ24; (5) Combined with the current scheduling information, it is estimated that the number of manual seats per hour on that day is ð1, ð2, ð3, …, ð24; (6) The saturation of manual agent calls at 24 hours on the same day is Σ1, Σ2, Σ3, ..., Σ24; the calculation method is: Σ1=S1 / ð1*100%=(M1*(1+α+β+γ+δ*N)+λ1) / ð1*100%, Σ2=S2 / ð2*100%=(M2*(1+α+β+γ+δ*N)+λ2) / ð2*100%, Σ3=S3 / ð3*100%=(M3*(1+α+β+γ+δ*N)+λ3) / ð3*100%, …………, Σ24=S24 / ð24*100%=(M24*(1+α+β+γ+δ*N)+λ24) / ð24*100%; The value of Σ can be used to determine the traffic forecast for the next day and to make correct measures based on the forecast: If the traffic saturation is >120%: Mild, If the traffic saturation is >150%: Moderate, If the traffic saturation is >200%: severe; When the call saturation is mild, the detection frequency needs to be increased; when the call saturation is moderate, while increasing the detection frequency, it is necessary to remind the operators to improve the service rhythm, end the service as soon as possible, reduce the break time and make calls as soon as possible; when the call saturation is severe, it is necessary to do the first two things and promptly initiate the traffic surge interception strategy, monitor the call volume in real time, and if necessary, activate the remote seat Shenzhen to send additional operators to save the plummeting connection rate.
[0008] The present invention also provides an intelligent emergency processing system capable of self-adapting to sudden increase in traffic, which is used to implement the intelligent emergency processing method capable of self-adapting to sudden increase in traffic as described above.
[0009] The system of the present invention comprises: a relay platform, an intelligent voice response system, a call center system, a CRM system, a fault work order system, a system background database and a manual seat; the call center system is respectively connected to the intelligent voice response system, the CRM system, the fault work order system, the system background database and the manual seat; the intelligent voice response system is connected to the communication network through the relay platform.
[0010] As a further description of the system of the present invention, the call center system is used to implement steps ③ and ④ of the intelligent emergency handling method capable of adaptively handling sudden increase in call traffic as described above.
[0011] As a further explanation of the system of the present invention, the intelligent voice response system is also connected to the automatic speech recognition ASR component and the text-to-speech TTS component respectively through the MRCP interface. The automation technology of the two components reduces the impact of the surge in call traffic on the system, reduces the pressure on manual agents to answer calls, and achieves the effect of improving the connection rate.
[0012] Advantages of the present invention: The emergency handling method of the present invention enables an intelligent interception strategy for traffic surges, predicts future traffic volume based on historical data analysis, current traffic volume, fault conditions, agent scheduling, and periodic trends, and when a traffic surge is predicted, the monitoring dispatcher can predict in advance and take correct response measures in advance, changing the traditional mode of directly cutting off the relay line channel. While significantly improving the manual connection rate, it actively notifies the customer of the fault information in the area through text messages, intelligent notifications, etc. at the first time, avoiding secondary congestion caused by customers repeatedly calling the hotline, improving the processing speed of manual agents and resource utilization, ensuring the manual connection rate without causing customer complaints, improving the operating efficiency and service level of the customer service center, and greatly resolving the contradiction between customer demands and tight agent resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1It is the system architecture diagram of the call center system customer service center.
[0014] Figure 2 It is a flow chart of the call center system after the customer dials the access number and enters the system.
[0015] Figure 3 This is a chart comparing the connection rates before and after the intelligent interception strategy for sudden increase in call traffic is activated. DETAILED DESCRIPTION
[0016] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. Example
[0017] like Figure 2 As shown, an intelligent emergency handling method capable of adaptively handling sudden increase in traffic volume comprises: ① After the user dials the customer service hotline and enters the IVR process, the welcome message is played, and the system determines whether the intelligent interception strategy for traffic surge is activated; ② If the traffic surge intelligent interception strategy has been activated, determine whether the customer is from the fault area and proceed to step ③; if the traffic surge intelligent interception strategy has not been activated, continue to the subsequent IVR process; ③ If the customer is from the fault area, determine whether it is a repeated call. If it is a repeated call, the customer will be given priority to enter the manual queue or enter step ④; if it is the first call, push the fault voice to the customer. If it meets the customer's needs, the customer will hang up and reduce the call volume, otherwise continue to follow the subsequent IVR process; if the customer is from a non-fault area, guide the customer to enter the automatic query function to reduce the call saturation; ④ In extreme cases, when the human agent cannot continue to accept new customers, the customer will be directly guided to the message module or a voice notification will be pushed to the customer to inform the customer that the human agent is busy, the current gas area has a fault, and how long it is expected to be restored; at the same time, a text message will be pushed to the customer to inform the customer that the system is busy, the current gas area has a fault, and how long it is expected to be restored; ⑤ When the saturation of manual call traffic remains high, continue to consider adding seats, adding remote seats, or even immediately connecting remote mobile seats to ensure call traffic.
[0018] It is further explained that the activation of the intelligent interception strategy for traffic surge is based on the forecast data of traffic saturation in the next 24 hours. When the predicted traffic volume exceeds 2 times the number of online seats, the interception strategy needs to be activated in time.
[0019] The traffic saturation prediction data for the next 24 hours is obtained by the following steps: (1) Select the historical traffic data of the last three years and calculate the traffic volume M1, M2, M3, ..., M24 at 24 hours a day under the condition of no failure; (2) Select the historical call data of the past three years, analyze and count the call problem types from the dimensions of hot events, customer system failures, extreme weather, and sudden gas failures, and use mathematical statistical models to calculate and deduce the corresponding impact value mean parameters as follows: the impact parameter of industry hot events on call traffic is α, the impact parameter of customer system failure events on call traffic is β, the impact parameter of extreme special weather events on call traffic is γ, and the number of lines affected by sudden failure events is N and the impact parameter is δ; (3) Extract the average traffic volume at 24 hours per day in the last three months for comparison, and determine the value of the percentage increase / decrease of the average traffic volume at 24 hours per day over the same period λ, which are marked as λ1, λ2, λ3, ..., λ24 respectively; (4) Based on the corresponding forecast parameters and the percentage increase / decrease of the hourly average traffic volume in the corresponding month of the previous year, the traffic load volume at 24 hours on the day is predicted to be S1, S2, S3, ..., S24; S1=M1*(1+α+β+γ+δ*N)+λ1; S2=M2*(1+α+β+γ+δ*N)+λ2; S3=M3*(1+α+β+γ+δ*N)+λ3; ………… S24=M24*(1+α+β+γ+δ*N)+λ24; (5) Combined with the current scheduling information, it is estimated that the number of manual seats per hour on that day is ð1, ð2, ð3, …, ð24; (6) The saturation of manual agent calls at 24 hours on the same day is Σ1, Σ2, Σ3, ..., Σ24; the calculation method is: Σ1=S1 / ð1*100%=(M1*(1+α+β+γ+δ*N)+λ1) / ð1*100%, Σ2=S2 / ð2*100%=(M2*(1+α+β+γ+δ*N)+λ2) / ð2*100%, Σ3=S3 / ð3*100%=(M3*(1+α+β+γ+δ*N)+λ3) / ð3*100%, ..., Σ24=S24 / ð24*100%=(M24*(1+α+β+γ+δ*N)+λ24) / ð24*100%; The traffic forecast for the next day can be determined based on the value of Σ, and correct measures can be taken based on the forecast: when the traffic saturation is mild, the detection frequency needs to be increased; when the traffic saturation is moderate, the detection frequency needs to be increased while reminding the operators to improve the service rhythm, end the service as soon as possible, reduce the break time and the service interval time; when the traffic saturation is severe, the first two items need to be done and the traffic surge interception strategy needs to be activated in time, the traffic volume needs to be monitored in real time, and if necessary, additional operators need to be dispatched and remote seats need to be activated to save the plummeting connection rate; If the traffic saturation is >120%: Mild, If the traffic saturation is >150%: Moderate, If the traffic saturation is > 200%: severe.
[0020] This embodiment also provides an intelligent emergency handling system capable of self-adapting to sudden increase in traffic, which is used to implement the intelligent emergency handling method capable of self-adapting to sudden increase in traffic as described above.
[0021] like Figure 1 As shown, the system includes: a relay platform, an intelligent voice response system, a call center system, a CRM system, a fault work order system, a system background database and manual seats; the call center system is connected to the intelligent voice response system, the CRM system, the fault work order system, the system background database and the manual seats respectively; the intelligent voice response system is connected to the communication network through the relay platform.
[0022] It is further explained that the call center system is used to implement steps ③ and ④ of the intelligent emergency handling method capable of adaptively handling sudden increase in call traffic as described above.
[0023] It is further explained that the intelligent voice response system is also connected to the automatic speech recognition ASR component and the text-to-speech TTS component through the MRCP interface. The automation technology of the two components reduces the impact of the surge in call traffic on the system, reduces the pressure on manual agents to answer calls, and achieves the effect of improving the connection rate.
[0024] The system is a customer service center based on computer telephone integration (CTI) and customer relationship management (CRM) systems, providing customers with consultation, complaints, after-sales service and other support through multiple channels such as telephone and Internet. The system architecture can be divided into two parts: the front desk and the back desk. The front desk is mainly responsible for interaction with customers, while the back desk is responsible for the management and analysis of customer information. With the continuous advancement of technology, the call center customer service system is also constantly developing and moving towards intelligence and automation.
[0025] The above emergency handling method and system are applied to a certain telecom smart cloud call center system, and the comparison data before and after the activation of the intelligent interception strategy on May 7, 2024 is recorded, such as Figure 3 As shown, starting at 9:00 a.m. on May 7, the customer service system experienced an abnormal surge in call traffic, causing the manual connection rate to drop rapidly to below 85%. The monitoring dispatcher activated the intelligent interception function at 11:00 a.m., which quickly restored the connection rate to a normal state of over 96%.
[0026] Obviously, the above embodiments are only examples for clearly explaining the present invention, and are not intended to limit the implementation of the present invention. For ordinary technicians in the relevant technical field, other different forms of changes or modifications can be made based on the above description; it is not necessary and impossible to list all the implementation methods here; and the obvious changes or modifications derived from them are still within the protection scope of the present invention.
Claims
1. An intelligent emergency handling method capable of adaptively responding to sudden increase in traffic, characterized in that include: ① After the user dials the customer service hotline and enters the IVR process, the welcome message is played, and the system determines whether the intelligent interception strategy for traffic surge is activated; ② If the traffic surge intelligent interception strategy has been activated, determine whether the customer is from the fault area and proceed to step ③; If the intelligent interception strategy for traffic surge is not activated, the subsequent IVR process will continue; ③If the customer is from the fault area, determine whether it is a repeated call. If it is a repeated call, the customer will be sent to the manual queue first, or go to step ④; If it is the first call, the fault voice will be pushed to the customer. If it meets the customer's needs, the customer will hang up and reduce the call volume. Otherwise, continue to follow the subsequent IVR process. If the customer is from a non-fault area, the customer will be guided to enter the automatic query function to reduce the call saturation. ④ In extreme cases, when the human agent cannot continue to accept new customers, the customer will be directly guided to the message module or a voice notification will be pushed to the customer to inform the customer that the human agent is busy, the current gas area has a fault, and how long it is expected to be restored; at the same time, a text message will be pushed to the customer to inform the customer that the system is busy, the current gas area has a fault, and how long it is expected to be restored; ⑤ When the saturation of manual call traffic remains high, continue to consider adding seats, adding remote seats, or even immediately connecting remote mobile seats to ensure call traffic.
2. The intelligent emergency handling method capable of adaptively handling sudden increase in traffic according to claim 1 is characterized in that: The activation basis of the intelligent interception strategy for traffic surge is that the traffic saturation forecast data for the next 24 hours shows a high concurrent traffic surge. When the predicted traffic volume exceeds the number of online seats by more than 2 times, the interception strategy needs to be activated in a timely manner.
3. The intelligent emergency handling method capable of adaptively handling sudden increase in traffic according to claim 2 is characterized in that: The traffic saturation prediction data for the next 24 hours is obtained by the following steps: (1) Select the historical traffic data of the last three years and calculate the traffic volume M1, M2, M3, ..., M24 at 24 hours a day under the condition of no failure; (2) Select the historical call data of the past three years, analyze and count the call problem types from the dimensions of industry hot events, customer system failures, extreme weather, and sudden gas failures, and use mathematical statistical models to calculate and deduce the corresponding impact value mean parameters as follows: the impact parameter of industry hot events on call traffic is α, the impact parameter of customer system failure events on call traffic is β, the impact parameter of extreme special weather events on call traffic is γ, and the number of lines affected by sudden failure events is N and the impact parameter is δ; (3) Extract the average traffic volume at 24 hours per day in the last three months for comparison, and determine the value of the percentage increase / decrease of the average traffic volume at 24 hours per day over the same period λ, which are marked as λ1, λ2, λ3, ..., λ24 respectively; (4) Based on the corresponding forecast parameters and the percentage increase / decrease of the hourly average traffic volume in the corresponding month of the previous year, the traffic load volume at 24 hours on the day is predicted to be S1, S2, S3, ..., S24; S1=M1*(1+α+β+γ+δ*N)+λ1; S2=M2*(1+α+β+γ+δ*N)+λ2; S3=M3*(1+α+β+γ+δ*N)+λ3; ………… S24=M24*(1+α+β+γ+δ*N)+λ24; (5) Combined with the current scheduling information, it is estimated that the number of manual seats per hour on that day is ð1, ð2, ð3, …, ð24; (6) The saturation of manual agent answering at 24 hours on the same day is Σ1, Σ2, Σ3, ..., Σ24; the calculation method is: Σ1=S1 / ð1*100%=(M1*(1+α+β+γ+δ*N)+λ1) / ð1*100%, Σ2=S2 / ð2*100%=(M2*(1+α+β+γ+δ*N)+λ2) / ð2*100%, Σ3=S3 / ð3*100%=(M3*(1+α+β+γ+δ*N)+λ3) / ð3*100%, ……, Σ24=S24 / ð24*100%=(M24*(1+α+β+γ+δ*N)+λ24) / ð24*100%; The value of Σ can be used to determine the traffic forecast for the next day and make correct measures based on the forecast. If the traffic saturation is >120%: Mild, If the traffic saturation is >150%: Moderate, If the traffic saturation is >200%: severe; When the call saturation is mild, the detection frequency needs to be increased; when the call saturation is moderate, while increasing the detection frequency, it is necessary to remind the operators to improve the service rhythm, end the service as soon as possible, reduce the break time and make calls as soon as possible; when the call saturation is severe, it is necessary to do the first two things and promptly initiate the traffic surge interception strategy, monitor the call volume in real time, and if necessary, activate remote seats or even add operators to save the plummeting connection rate.
4. An intelligent emergency handling system capable of adaptively responding to sudden increase in traffic, characterized in that: The system is used to implement the intelligent emergency handling method capable of adaptively handling sudden increase in traffic as described in claim 1.
5. The intelligent emergency handling system capable of adaptively responding to sudden increase in traffic according to claim 4 is characterized in that include: Relay platform, intelligent voice response system, call center system, CRM system, fault ticket system, system background database and manual seats; the call center system is connected to the intelligent voice response system, CRM system, fault ticket system, system background database and manual seats respectively; the intelligent voice response system is connected to the communication network through the relay platform.
6. The intelligent emergency handling system capable of adaptively handling sudden increase in traffic according to claim 5, characterized in that: The call center system is used to implement steps ③ and ④ in the intelligent emergency handling method capable of adaptively handling sudden increase in traffic as described in claim 1.
7. The intelligent emergency handling system capable of self-adapting to sudden increase in traffic according to claim 5, characterized in that: The intelligent voice response system is also connected to the automatic speech recognition ASR component and the text-to-speech TTS component through the MRCP interface. The automation technology of the two components reduces the impact of the surge in call traffic on the system, reduces the pressure on manual agents to answer calls, and achieves the effect of improving the connection rate.
