Telephone traffic scheduling method and system based on telephone traffic platform

Abstract

The invention discloses a telephone traffic scheduling method based on a telephone traffic platform, and the method comprises the following steps: obtaining a to-be-allocated telephone traffic within a preset dialing time, and carrying out the service extraction of the to-be-allocated telephone traffic, and obtaining a first service and a first service volume corresponding to the first service; the first service is any service in the telephone traffic to be distributed; based on the first idle seat group and the first service volume, determining a target to-be-distributed volume in the first service volume; the first idle seat group is any idle seat group corresponding to the first service in the seat cluster; when the target to-be-allocated quantity is greater than a first preset value, scheduling to-be-allocated telephone traffic in the target to-be-allocated quantity based on a second idle seat group; the second idle seat group is any idle seat group corresponding to a second service associated with the first service in the seat cluster. According to the method and the device, the incoming call can be quickly allocated to the idle seats, idle seats or incoming call congestion is avoided, telephone traffic resources are fully utilized, and thus the satisfaction degree of a telephone traffic platform is improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a traffic scheduling method based on a traffic platform and a system thereof. Background Art

[0002] With the development of society and the advancement of information technology, telephone services have become an indispensable part of people's daily life and work. Especially in call centers, customer service hotlines and other scenarios, efficient telephone services can not only improve customer satisfaction, but also significantly improve the operational efficiency of enterprises.

[0003] However, when the call center conducts real-time dispatch of incoming calls, the traditional dispatch method generally relies on personal operation experience to dispatch call resources, and the dispatch process often lacks accurate data basis. When the call peak comes, due to relying on personal experience to dispatch calls, it is difficult for managers to know the specific content of each seat's call in real time and estimate when the call will end. At the same time, the number of incoming calls and the number of idle seats in real time are simply counted, and no in-depth coordination is made based on the call details. This leads to long waiting times for incoming calls, unreasonable call distribution, and the inability to use idle seats in a timely and efficient manner, resulting in low customer satisfaction. Summary of the invention

[0004] The purpose of the present invention is to solve the problems in the prior art and to propose a traffic scheduling method and system based on a traffic platform.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A traffic scheduling method based on a traffic platform comprises the following steps:

[0007] Acquire the volume of traffic to be allocated within the preset dialing time, and perform service extraction on the volume of traffic to be allocated to obtain a first service and its corresponding first service volume; the first service is any service in the volume of traffic to be allocated;

[0008] Based on a first idle seat group and the first business volume, determining a target amount to be allocated in the first business volume; the first idle seat group is any idle seat group in the seat cluster corresponding to the first business;

[0009] When the target amount to be allocated is greater than a first preset value, the to-be-allocated traffic in the target amount to be allocated is scheduled based on a second idle agent group; the second idle agent group is any idle agent group corresponding to a second service associated with the first service in an agent cluster; the second service is any service in a call center except the first service.

[0010] According to a traffic scheduling method based on a traffic platform provided by the present invention, the scheduling of the to-be-allocated traffic in the target to-be-allocated amount based on the second idle agent group includes:

[0011] Performing correlation processing on the first business and the plurality of the second businesses to obtain a plurality of target correlation values;

[0012] Arrange the target association values ​​in descending order to obtain a target association sequence;

[0013] Determining, based on the multiple target association values ​​and the target association sequence, an allocation coefficient corresponding to each of the second services in the target association sequence;

[0014] The target amount of to-be-allocated traffic is scheduled based on the allocation coefficient, the target association sequence and the second idle agent group.

[0015] According to a traffic scheduling method based on a traffic platform provided by the present invention, the first service and a plurality of the second services are subjected to correlation processing to obtain a plurality of target correlation values, including:

[0016] Performing business semantic similarity processing on the first business and multiple second businesses to obtain multiple business semantic similarity values;

[0017] Performing business process correlation processing on the first business and multiple second businesses to obtain multiple business process correlation values;

[0018] Performing historical transfer probability processing based on the first service and multiple second services to obtain multiple historical transfer probability values;

[0019] The multiple target relevance values ​​are obtained by performing weighted summation based on the multiple business semantic similarity values, the multiple business process relevance values, and the multiple historical transfer probability values.

[0020] According to a traffic scheduling method based on a traffic platform provided by the present invention, the calculation formula of the allocation coefficient is:

[0021]

[0022] Among them, λ l represents the allocation coefficient of the lth service in allocating the traffic to be allocated; A represents the first service; B sl represents the business in the first position after being sorted by relevance among all the second businesses associated with the first business; R(A,B sl ) indicates the first business A and the lth associated business B after sorting slThe degree of association between them; m represents the total number of second services associated with the first service; R(A,B sk ) represents the first business and the sorted kth associated business B sk The degree of correlation between them.

[0023] According to a traffic scheduling method based on a traffic platform provided by the present invention, the method schedules the target amount of to-be-allocated traffic based on the allocation coefficient, the target association sequence and the second idle agent group, including:

[0024] Obtaining an average scheduling time for each traffic in the first service, and predicting upcoming traffic volumes of multiple second services within the average scheduling time based on the average scheduling time and the historical second service volume;

[0025] Based on the upcoming traffic volume, the allocation coefficient and the target association sequence, obtaining the maximum carrying capacity of the second idle agent group;

[0026] Determine a difference result based on the maximum carrying capacity and the target amount to be allocated;

[0027] The target amount of to-be-allocated traffic is scheduled based on the difference result and the acquired second idle seat group.

[0028] According to a traffic scheduling method based on a traffic platform provided by the present invention, scheduling the target amount of to-be-allocated traffic based on the difference result and the acquired second idle seat group includes:

[0029] If the difference result is less than or equal to the second preset value, dispatching the target amount of to-be-allocated traffic to the second idle seat;

[0030] If the difference result is greater than the second preset value, the calls with the former maximum load in the target amount to be allocated will be dispatched to the second idle seat, and based on the average scheduling time, the queuing time of each call with the corresponding difference result in the target amount to be allocated will be obtained, and the queuing time will be fed back to the calling customer.

[0031] According to a traffic scheduling method based on a traffic platform provided by the present invention, the calculation formula for obtaining the queuing time of each traffic corresponding to the difference result in the target amount to be allocated is:

[0032]

[0033] Among them, T que (i) represents the queuing time of the i-th traffic in the target volume to be allocated; n represents the number of batches of historical traffic involved in the calculation; S jrepresents the average processing speed of the jth batch of historical traffic, that is, the number of traffic processed per unit time; σ j It represents the standard deviation of the processing speed of the jth batch of historical traffic, reflecting the fluctuation of the processing speed. represents the average processing speed of the jth batch of traffic; L j represents the traffic volume of the jth batch of historical traffic; Δ represents the difference result; r represents the real-time processing rate of the current system.

[0034] A traffic dispatching system based on a traffic platform, comprising:

[0035] The acquisition unit is used to: acquire the volume of traffic to be allocated within a preset dialing time, and perform service extraction on the volume of traffic to be allocated to obtain a first service and its corresponding first service volume; the first service is any service in the volume of traffic to be allocated;

[0036] The processing unit is used to: determine the target amount to be allocated in the first business volume based on the first idle seat group and the first business volume; the first idle seat group is any idle seat group in the seat cluster corresponding to the first business;

[0037] The scheduling unit is used to: when the target amount to be allocated is greater than a first preset value, schedule the to-be-allocated traffic in the target amount to be allocated based on a second idle seat group; the second idle seat group is any idle seat group corresponding to a second service associated with the first service in an agent cluster; the second service is any service in a call center except the first service.

[0038] The present invention also provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned traffic scheduling method based on a traffic platform when executing the program.

[0039] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned traffic scheduling method based on a traffic platform are implemented.

[0040] Compared with the prior art, the present invention has the following advantages:

[0041] The present invention provides a traffic scheduling method and system based on a traffic platform, which obtains the traffic volume to be allocated within a preset dialing time, extracts a first service and its corresponding first service volume, then combines a first idle seat group with the first service volume to determine a target volume to be allocated in the first service volume, and then, based on the result of the target volume to be allocated, a second idle seat group corresponding to a plurality of second services is used to quickly allocate the traffic to be allocated in the target volume to be allocated, so that incoming calls can be quickly allocated to idle seats, avoiding idle seats or call congestion, and making full use of traffic resources, thereby improving satisfaction with the traffic platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are 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.

[0043] Figure 1 A flow chart of a method for dispatching a call traffic based on a call traffic platform provided by an embodiment of the present invention;

[0044] Figure 2 A schematic diagram of the structure of a traffic dispatching system based on a traffic platform provided in an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of the structure of the electronic device proposed by the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Combine the following Figure 1 - Figure 3 The present invention describes a traffic scheduling method and system based on a traffic platform.

[0048] Figure 1 FIG. 1 is a flow chart of a method for dispatching traffic based on a traffic platform provided by the present invention. Figure 1 As shown, the method includes:

[0049] Step 101, obtaining the traffic volume to be allocated within a preset dialing time, and performing service extraction on the traffic volume to be allocated to obtain a first service and its corresponding first service volume; the first service is any service in the traffic volume to be allocated.

[0050] Specifically, the preset dialing time refers to the time between the caller making a call and the time when the call is connected to an artificial customer. The call center platform presets intelligent voice navigation in the dialing system. When the caller enters the voice navigation, the customer is guided to talk to the corresponding artificial customer according to the corresponding business. Therefore, during this time, all call requests entering the call center platform are collected in real time through the communication interface to form the call volume to be allocated.

[0051] Furthermore, after the call platform collects all the call volumes to be allocated, it uses natural language processing technology and business classification algorithms to analyze each call request. A call request refers to the business request spoken by the caller during the intelligent customer guidance process. Natural language processing technology converts the received call request into text to understand the user's call intention. Business classification algorithms, such as text classification models, classify the text converted by natural language processing technology, and ultimately realize the classification of future calls into different business types. For example, for a call content: Check last month's water and electricity bills, the model identifies keywords such as water and electricity bills and determines it as a water and electricity bill inquiry business.

[0052] Furthermore, after determining the service type of the incoming call service, each service type is calculated and counted to obtain the service volume corresponding to each service, such as setting the set of all service categories as S = [s1, s2, ..., s μ ], by traversing all call requests, counting the business s i The number of calls is recorded as q i , where q i For business i Corresponding business volume, select one from all businesses as the first business, such as the first business selected is s j , then the corresponding business volume q j The first business volume.

[0053] Step 102: determine a target amount to be allocated in the first business volume based on a first idle seat group and a first business volume; the first idle seat group is any idle seat group in the seat cluster corresponding to the first business.

[0054] Specifically, while obtaining the first service volume, the call center platform also monitors the status of the seat cluster in real time, obtains all idle seats corresponding to the first service, and forms a first idle seat group. After obtaining the total number N of idle seats in the first idle seat group, the first service volume q j Compare to obtain the target amount Q to be allocated in the first business volumetar .

[0055] Step 103, when the target amount to be allocated is greater than the first preset value, the to-be-allocated traffic in the target amount to be allocated is scheduled based on the second idle seat group; the second idle seat group is any idle seat group corresponding to the second business associated with the first business in the seat cluster; the second business is any business in the call center except the first business.

[0056] Specifically, the first preset value is set first. If the first preset value is set to zero, it indicates that the total number of idle seats corresponding to the incoming call traffic is insufficient to cope with the total number of calls. Then, the idle seat group corresponding to the second service associated with the first service, i.e., the second idle seat group, is searched to receive the excess incoming calls of the first service. If the first preset value is set to 10, it indicates that the idle seat group corresponding to the second service associated with the first service is searched only when the target amount to be allocated is greater than 10. In addition, when comparing the total number of idle seats N in the first idle seat group with a traffic volume q j When the size of tar =q j -N is calculated, when Q tar <0, indicating that the number of idle seats is greater than the business volume received within the preset dialing time. The call platform directly assigns the corresponding calls to the idle seats of the corresponding business, without the caller waiting for the call. tar >0, indicating that the number of idle seats for the current service is insufficient to cope with the incoming call volume within the current preset dialing time. tar The judgment is made with the size of the first preset value to complete the scheduling and allocation of the excess traffic volume, that is, the target volume to be allocated, thereby reducing the waiting time of the incoming caller.

[0057] The present invention relates to the field of communications, and proposes a traffic scheduling method based on a traffic platform. In the present invention, the traffic scheduling method proposed obtains the traffic volume to be allocated within a preset dialing time, extracts a first service and its corresponding first service volume, then combines a first idle seat group with the first service volume to determine a target volume to be allocated in the first service volume, and then according to the result of the target volume to be allocated, a second idle seat group corresponding to a plurality of second services is used to quickly allocate the traffic to be allocated in the target volume to be allocated, so that incoming calls can be quickly allocated to idle seats, avoiding idle seats or call congestion, and making full use of traffic resources, thereby improving satisfaction with the traffic platform.

[0058] Step 103, in one embodiment, scheduling the to-be-allocated traffic in the target to-be-allocated amount based on the second idle seat group includes:

[0059] Step 1031 : performing correlation processing based on the first service and the plurality of second services to obtain a plurality of target correlation values.

[0060] Specifically, the correlation degree needs to be comprehensively considered in the calculation process, such as business semantic similarity, business process correlation degree and historical transfer probability. Determining the correlation degree between the second business and the first business from multiple dimensions makes the calculation of the correlation degree more accurate and comprehensive, and can more truly reflect the internal connection between the businesses.

[0061] Further, it includes: step 10311, performing business semantic similarity processing based on the first business and multiple second businesses to obtain multiple business semantic similarity values.

[0062] Specifically, firstly, natural language processing is used to convert the text descriptions of the first service and the second service into vector representations, such as using the Word2Vec model to map the words in the description text of each service to a low-dimensional vector space. 11 ,ω 12 ,…,ω 1θ} and the text T of the i-th second service 2i ={ω 2i1 ,ω 2i2 ,…,ω 2iδ}, calculate the vector representation of each word in the text respectively, and then get the text vector by averaging, etc. and Then the cosine similarity formula is used to calculate the business semantic similarity value of the two business description texts in

[0063] Step 10312: Perform business process correlation processing based on the first business and the plurality of second businesses to obtain a plurality of business process correlation values.

[0064] Specifically, when processing the correlation of business processes, the call center platform first models the processes of the first business and the second business, and decomposes the business processes into a series of steps. For example, the first business is to handle the change of mobile phone packages, and its process includes identity authentication, package query, package selection, confirmation of change and other steps; the second business is to change the mobile phone data package, and the process may include identity authentication, data package query, data package selection, confirmation of change and other steps. Then, the proportion of the number of identical steps in the two business processes to the total number of steps is compared to calculate the correlation of the business processes. Set the first business process step set to P1, the second business process step set to P2, and the set of identical steps to P com , then the business process correlation

[0065] Step 10313: Perform historical transfer probability processing based on the first service and the plurality of second services to obtain a plurality of historical transfer probability values.

[0066] Specifically, the call platform records the number of transfers from the first service to the second service in historical data and the total number of transfers of the first service. For example, if the total number of transfers of the first service in the past month is 100, and the number of transfers to a second service is 20, then the historical transfer probability is Among them, Z O Indicates the number of times the service is transferred to the second service. U Indicates the total number of transfers of the first service.

[0067] Step 10314, performing weighted summation based on multiple business semantic similarity values, multiple business process association values, and multiple historical transfer probability values ​​to obtain multiple target association values.

[0068] Specifically, the weight corresponding to the business semantic similarity value is set to ω s , the weight corresponding to the business process association value is set to ω p , the weight corresponding to the historical transfer probability value is ω h , and ω s +ω p +ω p =1. Then for the ith service, its target association value R i =ω s *S sem +ω p *S pro +ω h *P tran .

[0069] Furthermore, the cosine similarity formula is used for business semantic similarity because cosine similarity can measure the directional similarity of two vectors. In the text word vector representation, directional similarity means similar semantics. The business process association formula can intuitively reflect the similarity of business processes by calculating the proportion of the same steps in the total steps. The historical transfer probability formula is based on actual transfer data and reflects the correlation between businesses in past transfer behaviors. Therefore, the correlations of different dimensions are integrated to obtain a comprehensive target correlation value, which comprehensively considers the characteristics of different dimensions of the business and avoids the limitations of a single factor. By setting weights reasonably, the importance of different aspects can be highlighted according to actual needs, so that the target correlation value obtained can more comprehensively reflect the correlation between businesses. In addition, for the weight ω s ,ω p and ω hThe determination can be made based on the knowledge and experience of industry experts. For example, in a financial business association analysis scenario, experts believe that business semantic similarity is the most important for judging business association based on their understanding of financial business for many years, followed by business process association, and finally historical transfer probability. It can also be determined by using the hierarchical analysis method, specifically by constructing a hierarchical model, taking the target association value as the target layer, and business semantic similarity, business process association, and historical transfer probability as the criterion layer. Then, by comparing the importance of each element of the criterion layer to the target layer, a judgment matrix is ​​constructed. Then, the maximum eigenvalue of the judgment matrix and its corresponding eigenvector are calculated, and the weight vector is obtained after normalization.

[0070] Step 1032: Arrange in descending order based on the multiple target association values ​​to obtain a target association sequence.

[0071] Specifically, after obtaining multiple target association values, the multiple target association values ​​are stored in an array, and then the multiple target association values ​​are arranged in descending order through a sorting algorithm such as a quick sort algorithm. After sorting, a target association sequence is obtained. For example, the target association values ​​of the three second businesses and the second businesses are R1=0.8, R2=0.6, and R3=0.7, respectively. Then, after sorting, the target association sequence obtained is [08, 0.7, 0.6]. At the same time, the corresponding second businesses are also sorted in this order.

[0072] Step 1033: Determine a distribution coefficient corresponding to each second service in the target association sequence based on the multiple target association values ​​and the target association sequence.

[0073] Specifically, the calculation formula of the distribution coefficient is:

[0074] Among them, λ l represents the allocation coefficient of the lth service in allocating the traffic to be allocated; A represents the first service; B sl represents the business in the first position after being sorted by relevance among all the second businesses associated with the first business; R(A,B sl ) indicates the first business A and the lth associated business B after sorting sl The degree of association between them; m represents the total number of second services associated with the first service; R(A,B sk ) represents the first business and the sorted kth associated business B sk The degree of correlation between them.

[0075] Furthermore, the distribution coefficient λ l It reflects the relative importance of each business compared to other businesses in terms of relevance, such as business B s1The correlation degree with the first service A is significantly higher than that with other services, so the corresponding λ1 value is larger. When allocating the traffic to be allocated, service B s1 It can obtain relatively more call volume, thus providing a quantitative basis for the reasonable allocation of the call volume to be allocated through the allocation coefficient, ensuring that the call volume can be scientifically allocated to the idle seat groups corresponding to each related business according to the degree of business relevance, thereby improving the rationality and efficiency of the overall scheduling.

[0076] Step 1034: Schedule the target amount of to-be-allocated traffic based on the allocation coefficient, the target association sequence, and the second idle agent group.

[0077] By comprehensively considering the business relevance and the processing capacity of the idle seat group, it is possible to allocate calls more reasonably, avoid waste or overload of agent resources, and optimize the resource utilization efficiency of the entire call platform.

[0078] In one embodiment, step 1034, scheduling the target amount of to-be-allocated traffic based on the allocation coefficient, the target association sequence, and the second idle agent group, includes:

[0079] Step 10341, obtain the average scheduling time of each traffic in the first service, and based on the average scheduling time and the historical second service volume, predict the upcoming traffic volume of multiple second services within the average scheduling time.

[0080] Specifically, the traffic platform calculates the total dispatch time T of the first service from historical data. zz , and the total dispatch traffic quantity s s , through the total scheduling time T of the first service zz The total dispatch traffic volume s s , calculate the average scheduling time for each call If the first business is customer service consulting business, the total scheduling time T is obtained from historical data zz = 300 minutes, total dispatch traffic volume s s =60, then the average scheduling time for each call is T avg = 5 minutes, that is, each call takes an average of 5 minutes to process. And the historical data of the most recent month is preferred. For the historical second business volume, the call platform also collects and organizes the business volume data of each second business in the past month. For example, the historical second business set is B z = {B1, B2, ..., B q}, the corresponding historical business volume is Q hi (B i ), represents the business volume of the i-th second business in the past period of time (such as the past month). Through the business volume and the corresponding time, the business volume Q corresponding to the second business in unit time can be obtaineddd (B i ), and then the average scheduling time T of the first service avg , we can get the average scheduling event T avg The upcoming traffic volume Q of multiple second services in pred (B i )=T avg *Q dd (B i ), by calculating the upcoming traffic volume of the second service, it is possible to reserve at least the upcoming traffic volume Q when the second service receives the traffic volume of the first service. pred (B i ) to avoid that there are not enough second idle seats to receive the subsequent calls corresponding to the second service, thereby achieving scheduling assistance for the target to-be-allocated amount of the first service and at the same time not affecting the subsequent call reception status of the second service itself.

[0081] Step 10342: Based on the incoming traffic volume, the allocation coefficient and the target association sequence, the maximum carrying capacity of the second idle agent group is obtained.

[0082] Specifically, by obtaining the distribution coefficient λ = [λ1, λ2, ..., λ l ], the corresponding second service set B x = {B1, B2, ..., B g}, and according to the upcoming traffic volume Q obtained in step 10341 pred (B i ) and the distribution coefficient λ i , combined with the processing capacity of the second service agent group C (B i ) (calculated by the number of idle seats and the processing volume per unit time of each seat), first calculate the estimated allocated traffic volume Q of each second service all (B i )=λ i *Q tar , where Q tar is the target amount to be allocated for the first business, such as Q tar =100, λ1 = 0.3, then Q all (B1) = 0.3 * 100 = 30. Then, combined with the upcoming traffic volume Q of the second service pred (B i ), then the maximum capacity of the second idle seat group M(B i )=min(C(B i )-Q pred (B i )), Q all (B i ). In one embodiment, C(B1)=50, Qpred (B1) = 5, Q all (B1) = 30, then M(B1) = min(50-5, 30) = 30. This indicates that the second agent group corresponding to the second service is considering its own upcoming service volume Q pred (B i ) can carry the traffic volume allocated from the first service. Taking the allocation coefficient, the upcoming traffic volume and the agent group processing capacity into consideration, the maximum carrying capacity of the second idle agent group can be calculated more accurately to avoid over-allocation or under-allocation. At the same time, if compared with the estimated allocated traffic volume of the second service, the maximum carrying capacity of the second idle agent group can be calculated based on the upcoming traffic volume Q. pred (B i ) is large, avoid over-allocation. pred (B i ) is relatively small, it can not only meet the expected allocated call volume, but also reserve allocation capacity to cope with emergency allocation.

[0083] Step 10343, determine the difference result based on the maximum load and the target amount to be allocated.

[0084] Specifically, according to the target amount to be distributed Q tar and the maximum carrying capacity M(B) of the second idle seat group calculated in step 10342 i ), for each second service B i , calculate the difference result

[0085] Furthermore, in one embodiment, there are three second services, the maximum carrying capacity is M(B1)=75, M(B2)=60, M(B3)=40, and the target allocation capacity Q tar =200, then Δ(B) = 200-75-60-40 = 25. This means that after this allocation, there are still 25 calls that cannot be carried by the second idle seat.

[0086] Step 10344: Schedule the target amount of to-be-allocated traffic based on the difference result and the acquired second idle agent group.

[0087] The method comprises: if the difference result is less than or equal to a second preset value, dispatching the target amount of to-be-allocated call traffic to the second idle seat.

[0088] Specifically, the second preset value is taken as zero for calculation. When the difference result is less than or equal to zero, it indicates that the ideal state of traffic distribution is that all target traffic to be distributed can be carried by the second idle agent group.

[0089] Furthermore, when the calculated difference result Δ(B)≤0, it indicates that the maximum carrying capacity of the second idle seat group is sufficient to meet the target allocation quantity. At this time, according to the previously calculated allocation coefficient λ i , the target amount to be allocated Q tar Distribute to each second idle seat group according to the proportion. If there are The traffic volume Q allocated to the i-th second idle agent group is i =λ i *Q tar , if there are three second idle seat groups, the allocation coefficients are λ1=0.3, λ2=0.4, λ3=0.3, and the target amount to be allocated Q tar =100, then the traffic volume allocated to the first second idle seat group is Q1=0.3*100=30, the second is Q2=0.4*100=40, and the third is Q3=0.3*100=30. And through the traffic platform, these traffic volumes are respectively dispatched to the corresponding second idle seat groups.

[0090] If the difference result is greater than the second preset value, the calls with the former maximum load in the target amount to be allocated will be dispatched to the second idle seat, and based on the average scheduling time, the queuing time of each call with the corresponding difference result in the target amount to be allocated will be obtained, and the queuing time will be fed back to the calling customer.

[0091] Specifically, when the calculated difference result Δ(B)>0, it indicates that the second idle seat group cannot fully carry the target to-be-allocated amount. At this time, the part of the target to-be-allocated amount that can be carried by the second idle seat group (i.e., the maximum carrying capacity) is allocated to the second idle seat group according to the allocation coefficient. Assuming that the maximum carrying capacity Traffic volume allocated to the i-th second idle agent group

[0092] Furthermore, for the part exceeding the maximum carrying capacity, that is, the traffic volume corresponding to the difference result Δ(B), its queuing time needs to be calculated, and the calculation formula is: Among them, T que (i) represents the queuing time of the i-th traffic in the target volume to be allocated; n represents the number of batches of historical traffic involved in the calculation; S j represents the average processing speed of the jth batch of historical traffic, that is, the number of traffic processed per unit time; σ j It represents the standard deviation of the processing speed of the jth batch of historical traffic, reflecting the fluctuation of the processing speed. represents the average processing speed of the jth batch of traffic; L jrepresents the volume of the jth batch of historical calls; Δ represents the difference result; r represents the real-time processing rate of the current system. Then the queuing time of each call is fed back to the calling customer through the notification system of the call platform, such as through voice prompts or SMS notifications. By feeding back the queuing time to the customer, the customer has a clear understanding of the waiting time, reduces the customer's anxiety, and improves the customer experience. In the case of being unable to fully carry the call, the borne call volume is reasonably allocated, and the calls that exceed the carrying capacity are reasonably queued to ensure the orderliness of call scheduling. And the above calculation is recalculated every preset dialing time to update the queuing time.

[0093] Figure 2 FIG. 1 is a structural diagram of a traffic dispatching system based on a traffic platform provided by the present invention. Figure 2 As shown, the system includes: an acquisition unit 10 for acquiring the volume of to-be-allocated traffic within a preset dialing time, and performing business extraction on the volume of to-be-allocated traffic to obtain a first business and its corresponding first business volume; the first business is any business in the volume of to-be-allocated traffic; a processing unit 20 for determining a target volume of to-be-allocated traffic in the first business volume based on a first idle agent group and the first business volume; the first idle agent group is any idle agent group in an agent cluster corresponding to the first business; a scheduling unit 30 for scheduling the to-be-allocated traffic in the target volume of to-be-allocated traffic based on a second idle agent group when the target volume of to-be-allocated traffic is greater than a first preset value; the second idle agent group is any idle agent group corresponding to a second business associated with the first business in an agent cluster; the second business is any business in a call center except the first business.

[0094] Figure 3 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 3As shown, the electronic device may include: a processor (processor) 310, a communication interface (CommunicationsInterface) 320, a memory (memory) 330 and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 33 can call the logic instructions in the memory 330 to execute a traffic scheduling method based on the traffic platform, which method includes: obtaining the traffic volume to be allocated within the preset dialing time, and performing business extraction on the traffic volume to be allocated to obtain a first business and its corresponding first business volume; the first business is any business in the traffic volume to be allocated; based on the first idle seat group and the first business volume, determining the target amount to be allocated in the first business volume; the first idle seat group is any idle seat group in the seat cluster corresponding to the first business; when the target amount to be allocated is greater than the first preset value, scheduling the traffic to be allocated in the target amount to be allocated based on the second idle seat group; the second idle seat group is any idle seat group corresponding to the second business associated with the first business in the seat cluster; the second business is any business in the traffic center except the first business.

[0095] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0096] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute a traffic scheduling method based on a traffic platform provided by the above methods, the method including: obtaining the traffic volume to be allocated within a preset dialing time, and performing business extraction on the traffic volume to be allocated to obtain a first business and its corresponding first business volume; the first business is any business in the traffic volume to be allocated; based on the first idle seat group and the first business volume, determining the target amount to be allocated in the first business volume; the first idle seat group is any idle seat group corresponding to the first business in the seat cluster; when the target amount to be allocated is greater than the first preset value, scheduling the traffic to be allocated in the target amount to be allocated based on the second idle seat group; the second idle seat group is any idle seat group corresponding to the second business associated with the first business in the seat cluster; the second business is any business in the call center except the first business.

[0097] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute a traffic scheduling method based on a traffic platform provided above, the method comprising: obtaining the volume of traffic to be allocated within a preset dialing time, and performing service extraction on the volume of traffic to be allocated to obtain a first service and its corresponding first service volume; the first service is any service in the volume of traffic to be allocated; based on the first idle agent group and the first service volume, determining a target volume to be allocated in the first service volume; the first idle agent group is any idle agent group in the agent cluster corresponding to the first service; when the target volume to be allocated is greater than a first preset value, scheduling the traffic to be allocated in the target volume to be allocated based on the second idle agent group; the second idle agent group is any idle agent group corresponding to a second service associated with the first service in the agent cluster; the second service is any service in the call center except the first service.

[0098] The system embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0100] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A traffic scheduling method based on a traffic platform, characterized in that: The following steps are involved: Acquire the volume of to-be-allocated traffic within a preset dialing time, and perform service extraction on the volume of to-be-allocated traffic to obtain a first service and its corresponding first service volume; The first service is any service in the traffic to be allocated; Determining a target amount to be allocated in the first business volume based on the first idle seat group and the first business volume; The first idle seat group is any idle seat group in the seat cluster corresponding to the first service; When the target amount to be allocated is greater than a first preset value, scheduling the to-be-allocated traffic in the target amount to be allocated based on a second idle agent group; The second idle seat group is any idle seat group corresponding to the second service associated with the first service in the seat cluster; the second service is any service in the call center except the first service.

2. The traffic scheduling method based on the traffic platform according to claim 1, characterized in that: The scheduling of the to-be-allocated traffic in the target to-be-allocated amount based on the second idle seat group includes: Performing correlation processing on the first business and the plurality of the second businesses to obtain a plurality of target correlation values; Arrange the target association values ​​in descending order to obtain a target association sequence; Determining, based on the multiple target association values ​​and the target association sequence, an allocation coefficient corresponding to each of the second services in the target association sequence; The target amount of to-be-allocated traffic is scheduled based on the allocation coefficient, the target association sequence and the second idle agent group.

3. The traffic scheduling method based on the traffic platform according to claim 2, characterized in that: The performing correlation processing based on the first service and the plurality of the second services to obtain a plurality of target correlation values ​​includes: Performing business semantic similarity processing on the first business and multiple second businesses to obtain multiple business semantic similarity values; Performing business process correlation processing on the first business and multiple second businesses to obtain multiple business process correlation values; Performing historical transfer probability processing based on the first service and multiple second services to obtain multiple historical transfer probability values; The multiple target relevance values ​​are obtained by performing weighted summation based on the multiple business semantic similarity values, the multiple business process relevance values, and the multiple historical transfer probability values.

4. The traffic scheduling method based on the traffic platform according to claim 2, characterized in that: The calculation formula of the distribution coefficient is: Among them, λ l represents the allocation coefficient of the lth service in allocating the traffic to be allocated; A represents the first service; B sl represents the business in the first position after being sorted by relevance among all the second businesses associated with the first business; R(A,B sl ) indicates the first business A and the lth associated business B after sorting sl The degree of association between them; m represents the total number of second services associated with the first service; R(S,B sk ) represents the first business and the sorted kth associated business B sk The degree of correlation between them.

5. The traffic scheduling method based on the traffic platform according to claim 2, characterized in that: The step of scheduling the target amount of to-be-allocated traffic based on the allocation coefficient, the target association sequence, and the second idle agent group includes: Obtaining an average scheduling time for each traffic in the first service, and predicting upcoming traffic volumes of multiple second services within the average scheduling time based on the average scheduling time and the historical second service volume; Based on the upcoming traffic volume, the allocation coefficient and the target association sequence, obtaining the maximum carrying capacity of the second idle agent group; Determine a difference result based on the maximum carrying capacity and the target amount to be allocated; The target amount of to-be-allocated traffic is scheduled based on the difference result and the acquired second idle seat group.

6. The traffic scheduling method based on the traffic platform according to claim 5, characterized in that: The scheduling of the target amount of to-be-allocated traffic based on the difference result and the acquired second idle seat group includes: If the difference result is less than or equal to the second preset value, dispatching the target amount of to-be-allocated traffic to the second idle seat; If the difference result is greater than a second preset value, the calls with the former maximum load in the target amount to be allocated are dispatched to the second idle seat, and based on the average scheduling time, the queuing time of each call with the corresponding difference result in the target amount to be allocated is obtained, and the queuing time is fed back to the calling customer.

7. The traffic scheduling method based on the traffic platform according to claim 6, characterized in that: The calculation formula for obtaining the queuing time of each traffic corresponding to the difference result in the target amount to be allocated is: Among them, T que (i) represents the queuing time of the i-th traffic in the target volume to be allocated; n represents the number of batches of historical traffic involved in the calculation; S j represents the average processing speed of the jth batch of historical traffic, that is, the number of traffic processed per unit time; σ j It represents the standard deviation of the processing speed of the jth batch of historical traffic, reflecting the fluctuation of the processing speed. represents the average processing speed of the jth batch of traffic; L j represents the traffic volume of the jth batch of historical traffic; Δ represents the difference result; r represents the real-time processing rate of the current system.

8. A traffic dispatching system based on a traffic platform, applied to the traffic dispatching method based on a traffic platform as claimed in any one of claims 1 to 7; The traffic dispatching system based on the traffic platform includes: The acquisition unit is used to: acquire the volume of traffic to be allocated within a preset dialing time, and perform service extraction on the volume of traffic to be allocated to obtain a first service and its corresponding first service volume; the first service is any service in the volume of traffic to be allocated; The processing unit is used to: determine the target amount to be allocated in the first business volume based on the first idle seat group and the first business volume; the first idle seat group is any idle seat group in the seat cluster corresponding to the first business; The scheduling unit is used to: when the target amount to be allocated is greater than a first preset value, schedule the to-be-allocated traffic in the target amount to be allocated based on a second idle seat group; the second idle seat group is any idle seat group corresponding to a second service associated with the first service in an agent cluster; the second service is any service in a call center except the first service.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the traffic scheduling method based on the traffic platform as described in any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a traffic scheduling method based on a traffic platform as described in any one of claims 1 to 7 are implemented.

Images