Multi-line concurrent processing method for call center system

Abstract

The invention discloses a multi-line concurrent processing method for a call center system, and relates to the technical field of resource optimization scheduling of the call center system, and the method comprises the steps: building a call concurrent model according to line capacity data and historical call record data, and calculating a call load index of each time period; based on the call load index, dividing the call center line into a dynamic line group and a fixed line group; when the call load index is greater than a preset threshold value, allocating an idle line from the fixed line group to the dynamic line group, and carrying out load balancing distribution on lines of the dynamic line group; when the call load index is smaller than or equal to a preset threshold value, idle lines in the dynamic line group are recycled to the fixed line group, and meanwhile, line allocation data are recorded to be used for updating a call concurrency model, so that concurrency processing and optimal scheduling of multiple lines are realized. According to the invention, the problem of unreasonable resource allocation of a traditional call center in a service fluctuation period is solved, and the overall service quality is improved through dynamic resource scheduling.

Description

Technical Field

[0001] The present invention relates to the technical field of call center system resource optimization and scheduling, in particular to a method for concurrently processing multiple lines in a call center system. Background Art

[0002] Traditional call center systems mainly rely on manual scheduling and static resource allocation mechanisms, which makes them prone to performance bottlenecks under high load conditions, such as line blocking, response delays, and reduced service quality. In recent years, with the rapid development of information and communication technology, call center systems have gradually developed in the direction of intelligence and automation, especially scheduling systems based on big data and artificial intelligence have gradually become an important tool for optimizing resource allocation and improving service efficiency. By introducing real-time monitoring technology, data analysis technology, and artificial intelligence algorithms, call center systems can perceive the operating status of each communication line in real time, and dynamically adjust resource allocation according to actual load conditions, greatly improving the system's load processing capacity and resource utilization efficiency.

[0003] However, existing technologies still have certain deficiencies in concurrent processing and optimized scheduling. Traditional load forecasting models mostly rely on simple linear regression or rule-based scheduling algorithms, which are difficult to deal with complex multi-dimensional resource scheduling problems in call centers. Although existing systems can perform a certain degree of resource scheduling through simple load monitoring methods, they usually ignore the impact of multiple factors such as line quality and response delay on resource allocation, resulting in inaccurate scheduling decisions and waste of resources. At the same time, existing load forecasting models fail to fully consider the dynamic change characteristics in time series data, lack sufficient prediction accuracy, and cannot cope with load fluctuations during peak hours. In addition, traditional methods usually only consider a single scheduling goal when scheduling resources, such as minimizing response delay or maximizing system throughput, lack comprehensiveness and flexibility, and it is difficult to achieve balanced optimization of multiple goals.

[0004] Therefore, the existing technology faces problems such as uneven resource allocation, long response delay and insufficient system processing capacity when processing concurrent requests of multiple lines, and cannot effectively guarantee the stable operation and service quality of the call center under high load. In view of the above problems, the present invention proposes a method for concurrent processing of multiple lines for a call center system, which realizes efficient concurrent processing and dynamic optimization scheduling of multiple lines, solves the shortcomings of the existing technology in complex load prediction and multi-objective scheduling optimization, can allocate resources more accurately, and improve the overall performance and response speed of the system. Summary of the invention

[0005] In view of the problems of uneven load distribution, long response delay and waste of resources in the concurrent processing of multiple lines in the existing call center system, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to dynamically adjust the resource allocation of call center lines through load prediction and multi-dimensional resource allocation optimization, realize efficient concurrent processing of multiple lines, improve the overall performance and response speed of the system, and effectively deal with resource bottlenecks and performance degradation problems under high concurrency and high load conditions.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a method for concurrently processing multiple lines in a call center system, comprising: establishing a call concurrency model based on line capacity data and historical call record data, and calculating call load indicators for each time period; based on the call load indicators, dividing the call center lines into dynamic line groups and fixed line groups; when the call load indicators are greater than a preset threshold, allocating idle lines from the fixed line group to the dynamic line group, and performing load balancing on the lines of the dynamic line group; when the call load indicators are less than or equal to the preset threshold, recycling the idle lines in the dynamic line group to the fixed line group, and recording the line allocation data for updating the call concurrency model, thereby realizing concurrent processing and optimized scheduling of multiple lines.

[0009] As a preferred solution of the method for concurrently processing multiple lines in a call center system described in the present invention, the method for calculating the call load index is as follows: collecting line capacity data of the call center system and obtaining historical call record data, wherein the line capacity data includes the maximum concurrency number, peak load capacity and available duration of each communication line; establishing a capacity constraint matrix based on the line capacity data, and performing time series analysis on the historical call record data, and extracting call pattern features using an exponential smoothing method; based on the capacity constraint matrix and the call pattern features, a support vector regression algorithm is used to construct a call concurrency model to calculate the call load index.

[0010] As a preferred solution of the method for concurrently processing multiple lines in a call center system of the present invention, the historical call record data includes call time distribution, call duration distribution and call density distribution; the capacity constraint matrix is ​​divided into 48 time segments with a 24-hour cycle; the call concurrency model uses a Markov chain to describe the call transfer probability; the specific formula of the call load index is as follows:

[0011]

[0012] Where L(t) is the call load index at time t, ω i is the weight coefficient of the ith line, λ is the time decay factor, t iis the activation time of the ith line, C i is the maximum number of concurrent connections on the ith line, Δ is the time window size, and P ij is the call transfer probability matrix element, H j (τ) is the historical call density function, k is the number of observation time segments, D s is the call density deviation in the sth time segment, m is the total number of lines in the system, and n is the number of call states.

[0013] As a preferred solution of the method for concurrently processing multiple lines of a call center system described in the present invention, the call center lines are divided into dynamic line groups and fixed line groups based on the call load index, including: setting a line grouping threshold according to the fluctuation law of the call load index, analyzing the call load index of each time segment, wherein the line grouping threshold includes a load reference value, a fluctuation upper limit value and a fluctuation lower limit value; classifying the time segment when the call load index is greater than the load reference value and the fluctuation amplitude exceeds 30% as a dynamic regulation interval; at the same time, classifying the time segment when the call load index is less than or equal to the load reference value and the fluctuation amplitude is within 15% as a fixed guarantee interval; setting a dynamic line group and a fixed line group based on the dynamic regulation interval and the fixed guarantee interval; establishing a coordination mechanism between the dynamic line group and the fixed line group, and when the dynamic line group has insufficient capacity, the idle resources of the fixed line group can be temporarily called, and a line group switching buffer can be set, wherein the line group switching buffer is used to smooth the load fluctuation during the switching process.

[0014] As a preferred solution of the method for concurrently processing multiple lines for a call center system of the present invention, when the call load index is greater than a preset threshold, the idle lines are allocated from the fixed line group to the dynamic line group, and the lines of the dynamic line group are load balanced, including: monitoring the change trend of the call load index, when it is detected that the call load index exceeds the preset threshold for three consecutive sampling periods, the line allocation mechanism is triggered, and the real-time load rate of the dynamic line group is calculated; if the real-time load rate exceeds 85% and still has an increasing trend, the cross-group line allocation program is started; the idle line status in the fixed line group is detected, and the lines that are continuously idle within the expected time and have good performance indicators are selected as resources to be allocated; the line allocation operation is implemented, and the number of allocated lines is determined according to the load excess ratio, and a single allocation does not exceed 20% of the total capacity of the fixed line group; the performance of the newly allocated lines is preheated, and call tasks are allocated to the new lines by gradual loading, and the load balancing algorithm is executed on all lines of the dynamic line group.

[0015] As a preferred solution of the method for concurrently processing multiple lines in a call center system described in the present invention, the line allocation mechanism includes the following steps: collecting the numerical changes of the call load index and calculating the real-time load rate of the dynamic line group, wherein the real-time load rate is calculated by the ratio of the number of active calls in the current period to the total number of available lines; performing trend analysis on the real-time load rate, if the real-time load rate exceeds 85% and shows an upward trend, automatically detecting the idle line status in the fixed line group, screening out continuously idle and good-performance lines as allocation resources, and determining the specific number of lines to be allocated based on the current load excess ratio; preheating the performance of the selected new lines, and allocating call tasks in a progressive manner, while executing a load balancing algorithm based on the principle of minimum number of connections for all lines in the dynamic line group to achieve optimal resource allocation.

[0016] As a preferred solution of the method for concurrently processing multiple lines in a call center system of the present invention, when the call load index is less than or equal to the preset threshold, the idle lines in the dynamic line group are recycled to the fixed line group, and the line allocation data is recorded for updating the call concurrency model to achieve concurrent processing and optimized scheduling of multiple lines, including: monitoring the change of the call load index, when it is detected that the call load index is lower than the preset threshold for 5 consecutive sampling periods, triggering the line recycling mechanism, and analyzing the resource occupancy of each line in the dynamic line group, identifying the call volume that is continuously lower than 4 within the expected time. 0% of the lines are taken as objects to be recovered; line recovery preprocessing is performed to orderly migrate the existing call tasks on the lines to be recovered; line recovery operations are implemented in batches, and the number of lines recovered in each batch does not exceed 15% of the current capacity of the dynamic line group, and an observation period is set between adjacent batches; the recovered lines are reintegrated into the fixed line group, and the status and performance are reset and tested, and detailed data of this line recovery are recorded, wherein the detailed data includes the recovery time point, the number of recovered lines, the load change trend and the service quality index; the detailed data is input into the call concurrency model for online learning to optimize the model's prediction ability for load changes.

[0017] In a second aspect, an embodiment of the present invention provides a system for concurrently processing multiple lines in a call center system, comprising: a call concurrency model building module, which is used to establish a call concurrency model based on line capacity data and historical call record data, and calculate call load indicators for each time period; a division module, which divides the call center lines into dynamic line groups and fixed line groups based on the call load indicators; a load allocation and balancing module, which is used to allocate idle lines from the fixed line group to the dynamic line group when the call load indicator is greater than a preset threshold, and to load balance the lines in the dynamic line group; a line recovery and model update module, which is used to recover idle lines in the dynamic line group to the fixed line group when the call load indicator is less than or equal to the preset threshold, and at the same time record line allocation data for updating the call concurrency model to achieve concurrent processing and optimized scheduling of multiple lines.

[0018] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for concurrently processing multiple lines in a call center system as described in the first aspect of the present invention are implemented.

[0019] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for concurrently processing multiple lines in a call center system as described in the first aspect of the present invention are implemented.

[0020] The beneficial effects of the present invention are as follows: accurate prediction of business load is achieved by establishing a call concurrency model based on historical data, a line group division mechanism combining dynamic and static is adopted to provide a reliable basis for resource allocation, and the intelligent allocation strategy and load balancing algorithm triggered by threshold are combined to ensure efficient utilization of system resources. Meanwhile, continuous optimization of the model is achieved through timely resource recovery and data feedback mechanism. This method not only solves the problem of unreasonable resource allocation of traditional call centers during business fluctuation periods, but also improves the overall service quality through dynamic resource scheduling, and ultimately achieves the technical effect of improving line resource utilization, reducing operating costs and enhancing system reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing 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 creative work. Among them:

[0022] Figure 1This is a flowchart of a method for concurrently processing multiple lines in a call center system according to Example 1. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0026] Example 1

[0027] Reference Figure 1 , which is the first embodiment of the present invention, provides a method for concurrently processing multiple lines in a call center system, comprising:

[0028] S1: Establish a call concurrency model based on line capacity data and historical call record data, and calculate the call load index for each time period.

[0029] Specifically, the method for calculating the call load index is to collect the line capacity data of the call center system and obtain the historical call record data; establish a capacity constraint matrix based on the line capacity data, perform time series analysis on the historical call record data, and use the exponential smoothing method to extract call pattern characteristics;

[0030] It should be noted that the line capacity data includes the maximum number of concurrent calls, peak load capacity and available time of each communication line; the historical call record data includes call time distribution, call duration distribution and call density distribution; the capacity constraint matrix is ​​divided into 48 time segments with a 24-hour cycle; the call concurrency model uses Markov chain to describe the call transfer probability.

[0031] Furthermore, based on the capacity constraint matrix and call pattern characteristics, the support vector regression algorithm is used to build a call concurrency model and calculate the call load index.

[0032] Furthermore, the specific formula of the call load indicator is as follows:

[0033]

[0034] Where L(t) is the call load index at time t, ω i is the weight coefficient of the ith line, λ is the time decay factor, t i is the activation time of the ith line, C i is the maximum number of concurrent connections on the ith line, Δ is the time window size, and P ij is the call transfer probability matrix element, H j (τ) is the historical call density function, k is the number of observation time segments, D s is the call density deviation in the sth time segment, m is the total number of lines in the system, and n is the number of call states.

[0035] S2: Based on the call load indicator, divide the call center lines into dynamic line groups and fixed line groups.

[0036] Specifically, according to the fluctuation pattern of call load indicators, the line grouping threshold is set, and the call load indicators of each time segment are analyzed; the time segments when the call load indicators are greater than the load reference value and the fluctuation range exceeds 30% are classified as dynamic control intervals; at the same time, the time segments when the call load indicators are less than or equal to the load reference value and the fluctuation range is within 15% are classified as fixed guarantee intervals.

[0037] Furthermore, based on the dynamic control interval and the fixed guarantee interval, dynamic line groups and fixed line groups are set up; a coordination mechanism between the dynamic line groups and the fixed line groups is established. When the dynamic line group has insufficient capacity, the idle resources of the fixed line group can be temporarily called, and a line group switching buffer is set, wherein the line group switching buffer is used to smooth load fluctuations during the switching process.

[0038] It should be noted that the dynamic line group adjusts the number of lines as the call load index fluctuates; the fixed line group maintains a basic line scale of no less than 40% of the total number of lines; the line grouping threshold includes the load reference value, the upper limit of fluctuation and the lower limit of fluctuation.

[0039] S3: When the call load index is greater than a preset threshold, allocate idle lines from the fixed line group to the dynamic line group, and perform load balancing on the lines of the dynamic line group.

[0040] Specifically, the changing trend of the call load indicator is monitored. When it is detected that the call load indicator exceeds the preset threshold for three consecutive sampling cycles, the line allocation mechanism is triggered to calculate the real-time load rate of the dynamic line group; if the real-time load rate exceeds 85% and there is still a growth trend, the cross-group line allocation procedure is started.

[0041] It should be noted that the preset threshold is set by analyzing the maximum concurrency during the peak hour, calculating the call connection rate per unit time, evaluating the performance overhead of line switching, and combining the fluctuation trend of business volume in historical data to set a dynamic threshold range that can balance system stability and resource utilization.

[0042] Furthermore, the line allocation mechanism includes the following steps: collecting the numerical changes of call load indicators, calculating the real-time load rate of the dynamic line group, wherein the real-time load rate is calculated by the ratio of the number of active calls in the current period to the total number of available lines; performing trend analysis on the real-time load rate, if the real-time load rate exceeds 85% and shows an upward trend, automatically detecting the idle line status in the fixed line group, screening out continuously idle and high-performance lines as allocation resources, and determining the specific number of lines to be allocated based on the current load excess ratio; preheating the performance of the selected new lines, and allocating call tasks in a progressive manner, while executing a load balancing algorithm based on the principle of minimum number of connections for all lines in the dynamic line group to achieve optimal resource allocation.

[0043] Furthermore, the idle line status in the fixed line group is detected, and the lines that are continuously idle within the expected time and have good performance indicators are screened out as resources to be allocated; the line allocation operation is implemented, and the number of lines to be allocated is determined according to the load excess ratio, and a single allocation does not exceed 20% of the total capacity of the fixed line group; the performance of the newly allocated lines is preheated, and call tasks are allocated to the new lines by gradual loading, and at the same time, the load balancing algorithm is executed on all lines of the dynamic line group.

[0044] S4: When the call load index is less than or equal to the preset threshold, the idle lines in the dynamic line group are recycled to the fixed line group, and the line allocation data is recorded to update the call concurrency model to achieve concurrent processing and optimized scheduling of multiple lines.

[0045] Specifically, the changes in the call load indicator are monitored. When it is detected that the call load indicator is lower than the preset threshold for five consecutive sampling cycles, the line recovery mechanism is triggered, and the resource occupancy of each line in the dynamic line group is analyzed. The lines whose call volume continues to be lower than 40% within the expected time are identified as objects to be recovered.

[0046] Furthermore, line recovery preprocessing is performed to orderly migrate existing call tasks on the lines to be recovered; line recovery operations are implemented in batches, and the number of lines recovered in each batch does not exceed 15% of the current capacity of the dynamic line group, and an observation period is set between adjacent batches.

[0047] Furthermore, the recovered lines are reintegrated into the fixed line group, the status is reset and the performance is tested, and the detailed data of this line recovery is recorded; the detailed data is input into the call concurrency model for online learning to optimize the model's ability to predict load changes.

[0048] It should be noted that the detailed data includes the recovery time point, the number of recovery lines, the load change trend and the service quality indicators.

[0049] In summary, the present invention realizes accurate prediction of business load by establishing a call concurrency model based on historical data, adopts a line group division mechanism combining dynamic and static to provide a reliable basis for resource allocation, and combines threshold-triggered intelligent allocation strategy and load balancing algorithm to ensure efficient utilization of system resources. At the same time, continuous optimization of the model is achieved through timely resource recovery and data feedback mechanism. This method not only solves the problem of unreasonable resource allocation of traditional call centers during business fluctuations, but also improves the overall service quality through dynamic resource scheduling, and ultimately achieves the technical effect of improving line resource utilization, reducing operating costs and enhancing system reliability.

[0050] Furthermore, the present embodiment also provides a system for concurrently processing multiple lines of a call center system, comprising: an acquisition module, for acquiring real-time signals of multiple communication lines of a call center, fusing the real-time signals and converting them into original line monitoring data, wherein the real-time signals include call load signals, line quality signals and response delay signals; a feature extraction module, for performing wavelet transform on the original line monitoring data, extracting its high-frequency components and low-frequency components, and constructing a line performance feature vector using the high-frequency components and the low-frequency components; a load index prediction module, for inputting the line performance feature vector into a pre-trained long short-term memory neural network model to obtain a load index of the call center line, wherein the load index represents the allocation and processing capacity of line resources; a resource allocation and scheduling optimization module, for establishing a multidimensional line resource allocation model based on the load index, and calculating the resource weight parameters and scheduling strategy parameters of the multidimensional line resource allocation model through the maximum likelihood estimation method, to achieve concurrent processing and optimized scheduling of multiple lines.

[0051] This embodiment also provides a computer device, which is suitable for the case of a method for concurrently processing multiple lines in a call center system, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the method for concurrently processing multiple lines in a call center system as proposed in the above embodiment.

[0052] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0053] Example 2

[0054] Referring to Table 1, which is a second embodiment of the present invention, this embodiment provides a method for concurrently processing multiple lines in a call center system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0055] Specifically, the historical call data of a provincial customer service center in the past six months was collected, including the call volume, call duration, call density, etc. at each time period of the day. The historical data was preprocessed using Python data analysis tools, time series analysis was performed using the pandas library, and the support vector regression algorithm was implemented using the sklearn library to build a concurrent model.

[0056] Furthermore, in the model training phase, the historical data is divided into 48 time segments according to a 24-hour cycle, each time segment is 30 minutes. The call pattern characteristics are extracted by exponential smoothing method, and the capacity constraint matrix is ​​established. The capacity constraint matrix takes into account the maximum number of concurrent calls (range 80-120 concurrent calls), peak load (up to 90% load during peak hours) and available time (normal operation time > 23 hours / day) of each communication line. Markov chain is used to describe the call transfer probability, the time decay factor is set to 0.85, and the observation window size is set to 12 time segments.

[0057] Furthermore, based on the trained model, the 300 lines are initially divided into 200 fixed lines (fixed line group) and 100 dynamic lines (dynamic line group). The load reference value in the line grouping threshold is set to 65%, the upper limit of fluctuation is 85%, and the lower limit of fluctuation is 40%. When the system detects that the load index exceeds 85% for three consecutive sampling cycles (each sampling cycle is 5 minutes), the line allocation mechanism is triggered; when it is lower than 40% for five consecutive sampling cycles, the line recovery mechanism is started.

[0058] Specifically, as shown in Table 1, the dynamic resource allocation method proposed in the present invention has significant advantages over the traditional fixed allocation scheme; in terms of service quality, the average connection rate is increased from 82.5% to more than 95%, and the average response time is reduced from 45.8 seconds to less than 10 seconds. This is due to the system's ability to quickly adjust resource allocation according to real-time load indicators and effectively respond to business peak demand.

[0059] Table 1 Comparison of performance indicators of call center dynamic resource allocation system

[0060]

[0061] Furthermore, in terms of resource utilization efficiency, through the dynamic allocation mechanism, the overall resource utilization of the system has been increased from 58.6% to more than 85%, especially on weekdays through the intelligent prediction allocation scheme, the resource utilization rate can reach up to 91.2%. At the same time, the peak concurrent processing capacity of the system has been significantly improved, from the original 850 concurrency to 1350 concurrency, an increase of 58.8%; in terms of system stability, by introducing the line group switching buffer and the progressive load distribution mechanism, the system stability index has been increased from 0.82 to more than 0.95. In particular, after adopting the intelligent prediction allocation scheme, the system stability index reached 0.99, indicating that the present invention can maintain the stable operation of the system while ensuring high efficiency.

[0062] Furthermore, in terms of economic benefits, through dynamic resource allocation and intelligent prediction mechanisms, the system operating cost reduction rate reaches 15.8%-25.3% in different scenarios. This is mainly due to the system's ability to automatically recycle idle resources during off-peak periods to avoid resource waste, while reducing the need for additional resource investment through accurate resource prediction and allocation during peak periods.

[0063] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for concurrently processing multiple lines in a call center system, characterized in that: include, Establish a call concurrency model based on line capacity data and historical call record data, and calculate call load indicators for each time period; Based on the call load indicator, the call center lines are divided into dynamic line groups and fixed line groups; When the call load index is greater than a preset threshold, allocate idle lines from the fixed line group to the dynamic line group, and perform load balancing on the lines of the dynamic line group; When the call load index is less than or equal to the preset threshold, the idle lines in the dynamic line group are recycled to the fixed line group, and the line allocation data is recorded to update the call concurrency model to achieve concurrent processing and optimized scheduling of multiple lines.

2. The method for concurrently processing multiple lines in a call center system according to claim 1, characterized in that: The method for calculating the call load index is: Collecting line capacity data of the call center system and obtaining historical call record data, wherein the line capacity data includes the maximum number of concurrent connections, peak load capacity, and available time of each communication line; Establishing a capacity constraint matrix according to the line capacity data, performing time series analysis on the historical call record data, and extracting call pattern features using an exponential smoothing method; Based on the capacity constraint matrix and the call pattern characteristics, a call concurrency model is constructed using a support vector regression algorithm to calculate a call load index.

3. The method for concurrently processing multiple lines in a call center system according to claim 2, characterized in that: The historical call record data includes call time distribution, call duration distribution and call density distribution; the capacity constraint matrix is ​​divided into 48 time segments with a 24-hour cycle; the call concurrency model uses a Markov chain to describe the call transfer probability; the specific formula of the call load indicator is as follows: Where L(t) is the call load index at time t, ω i is the weight coefficient of the ith line, λ is the time decay factor, t i is the activation time of the ith line, C i is the maximum number of concurrent connections on the ith line, Δ is the time window size, and P ij is the call transfer probability matrix element, H j (τ) is the historical call density function, k is the number of observation time segments, D s is the call density deviation in the sth time segment, m is the total number of lines in the system, and n is the number of call states.

4. The method for concurrently processing multiple lines in a call center system according to claim 3, characterized in that: Based on the call load index, the call center lines are divided into dynamic line groups and fixed line groups, including: According to the fluctuation law of the call load index, a line grouping threshold is set to analyze the call load index of each time segment, wherein the line grouping threshold includes a load reference value, a fluctuation upper limit value and a fluctuation lower limit value; The time segment in which the call load index is greater than the load reference value and the fluctuation range exceeds 30% is classified as a dynamic control interval; At the same time, the time segments in which the call load index is less than or equal to the load reference value and the fluctuation range is within 15% are classified as fixed protection intervals; Based on the dynamic control interval and the fixed guarantee interval, a dynamic line group and a fixed line group are set; A coordination mechanism between the dynamic line group and the fixed line group is established. When the dynamic line group has insufficient capacity, the idle resources of the fixed line group can be temporarily called, and a line group switching buffer is set, wherein the line group switching buffer is used to smooth load fluctuations during the switching process.

5. The method for concurrently processing multiple lines in a call center system according to claim 4, characterized in that: When the call load index is greater than a preset threshold, idle lines are allocated from the fixed line group to the dynamic line group, and load balancing is performed on the lines of the dynamic line group, including: Monitor the changing trend of the call load indicator. When it is detected that the call load indicator exceeds the preset threshold for three consecutive sampling periods, the line allocation mechanism is triggered to calculate the real-time load rate of the dynamic line group; if the real-time load rate exceeds 85% and still has an increasing trend, the cross-group line allocation procedure is started; Detecting the idle line status in the fixed line group, and selecting lines that are continuously idle within an expected time and have good performance indicators as resources to be allocated; Implement line allocation operations, and the number of lines allocated is determined according to the load excess ratio, and a single allocation does not exceed 20% of the total capacity of the fixed line group; The performance of the newly deployed lines is preheated, and call tasks are allocated to the new lines by gradual loading. At the same time, a load balancing algorithm is executed on all lines of the dynamic line group.

6. The method for concurrently processing multiple lines in a call center system according to claim 5, characterized in that: The line allocation mechanism includes the following steps: Collecting the value change of the call load indicator and calculating the real-time load rate of the dynamic line group, wherein the real-time load rate is calculated by the ratio of the number of active calls in the current period to the total number of available lines; Performing trend analysis on the real-time load rate, if the real-time load rate exceeds 85% and shows an upward trend, automatically detecting the idle line status in the fixed line group, selecting lines that are continuously idle and have good performance as deployment resources, and determining the specific number of lines to be deployed according to the current load excess ratio; The performance of the selected new lines is preheated, and call tasks are allocated in a progressive manner. At the same time, a load balancing algorithm based on the principle of minimum number of connections is executed on all lines of the dynamic line group to achieve optimal allocation of resources.

7. The method for concurrently processing multiple lines in a call center system according to claim 6, characterized in that: When the call load index is less than or equal to the preset threshold, the idle lines in the dynamic line group are recycled to the fixed line group, and the line allocation data is recorded to update the call concurrency model, so as to realize concurrent processing and optimized scheduling of multiple lines, including: Monitor the change of the call load indicator, and when it is detected that the call load indicator is lower than the preset threshold for 5 consecutive sampling periods, trigger the line recovery mechanism, analyze the resource occupancy of each line in the dynamic line group, and identify the lines whose call volume is continuously lower than 40% within the expected time as the objects to be recovered; Perform line recovery preprocessing and orderly migrate existing call tasks on the recovered lines; Implement line recovery operations in batches, and the number of lines recovered in each batch shall not exceed 15% of the current capacity of the dynamic line group, and an observation period shall be set between adjacent batches; The recovered lines are re-integrated into the fixed line group, status reset and performance test are performed, and detailed data of this line recovery is recorded, wherein the detailed data includes the recovery time point, the number of recovered lines, the load change trend and the service quality index; The detailed data is input into the call concurrency model for online learning to optimize the model's ability to predict load changes.

8. A system for concurrently processing multiple lines in a call center system, based on the method for concurrently processing multiple lines in a call center system according to any one of claims 1 to 7, characterized in that: include, A call concurrency model building module is used to build a call concurrency model based on line capacity data and historical call record data, and calculate call load indicators for each time period; A division module, based on the call load index, divides the call center lines into dynamic line groups and fixed line groups; A load allocation and balancing module, configured to allocate idle lines from the fixed line group to the dynamic line group when the call load index is greater than a preset threshold, and perform load balancing on the lines of the dynamic line group; The line recovery and model update module is used to recover the idle lines in the dynamic line group to the fixed line group when the call load index is less than or equal to the preset threshold, and record the line allocation data for updating the call concurrency model to realize concurrent processing and optimized scheduling of multiple lines.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for concurrently processing multiple lines in a call center system according to any one of claims 1 to 7 are implemented.

10. A 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 the method for concurrently processing multiple lines in a call center system according to any one of claims 1 to 7 are implemented.

Images