Business processing method and device, electronic equipment and storage medium

Abstract

According to the business processing method and device, the electronic equipment and the storage medium provided by the invention, firstly, the business demand data of the target business and the historical business data corresponding to the target business are acquired, and the business demand data at least comprises the simulated user incoming line quantity, the simulated seat quantity and the business simulation termination condition; then, based on the simulated user incoming line quantity and the simulated seat quantity, a simulation implementation scheme of a target service is initialized, simulation implementation is carried out on the simulation implementation scheme until a service simulation termination condition is reached, a simulation result is obtained, the simulation implementation scheme is simulated through a digital twin model, the operation process can be rehearsed before the service is actually developed, and the service simulation efficiency is improved. Potential problems such as seat scheduling and incoming line scheduling are found, and the accuracy of service processing is improved. And finally, determining a to-be-applied implementation scheme of the target service according to a simulation result and historical service data. According to the technical scheme, simulation dynamic adjustment of the implementation scheme is carried out according to the service scene, and the service processing efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of business consultation processing technology, and in particular to a business processing method, apparatus, electronic device and storage medium. Background Technology

[0002] With the development of digital technology, automation has replaced some manual processes in routine customer service within the banking industry. However, a large number of customers still require human customer service to handle more complex issues. Optimizing the operation and management of customer service centers plays a crucial role in improving customer satisfaction and achieving efficient banking operations.

[0003] In existing technologies, statistical reports are used to analyze and process customer demand response and agent scheduling, but they cannot detect and automatically respond to abnormal or sudden situations of customer service center equipment in a timely manner. They rely heavily on manual processing and are not timely.

[0004] Therefore, improving the accuracy and efficiency of business processing has become an urgent technical problem to be solved. Summary of the Invention

[0005] This application provides a business processing method, apparatus, electronic device, and storage medium to solve the problems of low business processing efficiency in the prior art.

[0006] In a first aspect, embodiments of this application provide a business processing method, including:

[0007] Obtain business requirement data for the target business and historical business data corresponding to the target business; the business requirement data should include at least the number of simulated user calls, the number of simulated agents, and the conditions for terminating the business simulation.

[0008] Based on the simulated number of incoming users and the simulated number of agent seats, initialize the simulation implementation plan for the target service;

[0009] The simulation implementation plan is simulated using the digital twin model corresponding to the target customer service center until the business simulation termination conditions are met, and the simulation results are obtained.

[0010] Based on simulation results and historical business data, determine the implementation plan to be applied to the target business.

[0011] In one possible implementation, the simulation results include the simulation device status and simulation agent status generated during the simulation implementation process, as well as the simulation scheduling strategy adopted during the simulation implementation process, and the historical business data includes at least the historical device status and historical agent status.

[0012] Based on simulation results and historical business data, determine the implementation plan for the target business, including:

[0013] The simulated device status is compared with the historical device status to obtain the first comparison result, and the simulated agent status is compared with the historical agent status to obtain the second comparison result.

[0014] If either the first comparison result or the second comparison result is abnormal, adjust the simulation scheduling strategy in the simulation implementation plan and repeat the process of simulating the implementation plan using the digital twin model corresponding to the target customer service center until both the first and second comparison results are normal.

[0015] If both the first and second comparison results are normal, the simulation implementation plan will be determined as the application implementation plan for the target business.

[0016] In one possible implementation, the process of establishing a digital twin model corresponding to the target customer service center includes:

[0017] Obtain equipment layout data, environmental data, and customer service workstation data for the target customer service center;

[0018] Based on equipment layout data, environmental data, and customer service workstation data, a digital twin model corresponding to the target customer service center is established.

[0019] In one possible implementation, real-time operational data of the target customer service center is obtained;

[0020] Real-time operational data is synchronized to the digital twin model so that the digital twin model can simulate operation based on the real-time operational data.

[0021] In one possible implementation, synchronizing real-time operational data to the digital twin model includes:

[0022] When the real-time running data is high-priority data, the first data transmission method is used to synchronize the real-time running data to the digital twin model;

[0023] When the real-time running data is low-priority data, the second data transmission method is used to synchronize the real-time running data to the digital twin model;

[0024] The first data transmission method includes any one of the following: data transmission method based on the User Datagram Protocol (UDP), data transmission method based on the Fast UDP Internet Protocol (QUIC), and data transmission method based on the Transmission Control Protocol (TCP); the second data transmission method is a data transmission method based on an incremental synchronization algorithm or an adaptive adoption algorithm.

[0025] In one possible implementation, after synchronizing real-time operational data to the digital twin model, the method further includes:

[0026] The real-time device status is determined by comparing the deviation between real-time device data and historical device data in the real-time operation data.

[0027] The real-time agent status is determined based on the anomaly detection results of the real-time agent status data and historical agent status data in the real-time operation data.

[0028] In one possible implementation, real-time agent status data includes agent heart rate data, agent facial data, and agent call content; the real-time agent status is determined based on anomaly detection results between the real-time agent status data and historical agent status data in the real-time operation data, including:

[0029] The peak stress level of agents is obtained from agent heart rate data, the agent emotion value is obtained from agent facial data, and the agent call efficiency is obtained from agent call content.

[0030] If any of the peak agent stress, agent mood value, or agent call efficiency exceeds the corresponding preset abnormal threshold, calculate the abnormal combination score of the peak agent stress, agent mood value, and agent call efficiency.

[0031] If the score of the abnormal combination is greater than the preset abnormal combination threshold, the real-time agent status is determined to be abnormal.

[0032] If the score of the abnormal combination is less than or equal to the preset abnormal combination threshold, the real-time agent status is determined to be normal.

[0033] Secondly, embodiments of this application provide a business processing apparatus, including:

[0034] The acquisition module is used to acquire business requirement data for the target business and historical business data corresponding to the target business; the business requirement data includes at least the number of simulated user calls, the number of simulated agents, and the business simulation termination conditions.

[0035] The processing module is used to initialize the simulation implementation plan of the target service based on the number of simulated user incoming calls and the number of simulated agent seats;

[0036] The implementation module is used to simulate the implementation of the simulation plan through the digital twin model corresponding to the target customer service center until the business simulation termination conditions are met, and the simulation results are obtained.

[0037] The determination module is used to determine the application implementation plan for the target service based on simulation results and historical business data.

[0038] In one possible implementation, the simulation results include the simulation device status and simulation agent status generated during the simulation implementation process, as well as the simulation scheduling strategy adopted during the simulation implementation process. The historical service data includes at least the historical device status and historical agent status. The determination module is specifically used for:

[0039] The simulated device status is compared with the historical device status to obtain the first comparison result, and the simulated agent status is compared with the historical agent status to obtain the second comparison result.

[0040] If either the first comparison result or the second comparison result is abnormal, adjust the simulation scheduling strategy in the simulation implementation plan and repeat the process of simulating the implementation plan using the digital twin model corresponding to the target customer service center until both the first and second comparison results are normal.

[0041] If both the first and second comparison results are normal, the simulation implementation plan will be determined as the application implementation plan for the target business.

[0042] In one possible implementation, the process of establishing a digital twin model corresponding to the target customer service center is specifically implemented by the following module:

[0043] Obtain equipment layout data, environmental data, and customer service workstation data for the target customer service center;

[0044] Based on equipment layout data, environmental data, and customer service workstation data, a digital twin model corresponding to the target customer service center is established.

[0045] In one possible implementation, the acquisition module is also used for;

[0046] Real-time operational data is synchronized to the digital twin model so that the digital twin model can simulate operation based on the real-time operational data.

[0047] In one possible implementation, the real-time runtime data is synchronized to the digital twin model, and the acquisition module is specifically used for:

[0048] When the real-time running data is high-priority data, the first data transmission method is used to synchronize the real-time running data to the digital twin model;

[0049] When the real-time running data is low-priority data, the second data transmission method is used to synchronize the real-time running data to the digital twin model;

[0050] The first data transmission method includes any one of the following: data transmission method based on the User Datagram Protocol (UDP), data transmission method based on the Fast UDP Internet Protocol (QUIC), and data transmission method based on the Transmission Control Protocol (TCP); the second data transmission method is a data transmission method based on an incremental synchronization algorithm or an adaptive adoption algorithm.

[0051] In one possible implementation, after synchronizing real-time operational data to the digital twin model, the determining module is further configured to:

[0052] The real-time device status is determined by comparing the deviation between real-time device data and historical device data in the real-time operation data.

[0053] The real-time agent status is determined based on the anomaly detection results of the real-time agent status data and historical agent status data in the real-time operation data.

[0054] In one possible implementation, real-time agent status data includes agent heart rate data, agent facial data, and agent call content; the determination module is specifically used for:

[0055] The peak stress level of agents is obtained from agent heart rate data, the agent emotion value is obtained from agent facial data, and the agent call efficiency is obtained from agent call content.

[0056] If any of the peak agent stress, agent mood value, or agent call efficiency exceeds the corresponding preset abnormal threshold, calculate the abnormal combination score of the peak agent stress, agent mood value, and agent call efficiency.

[0057] If the score of the abnormal combination is greater than the preset abnormal combination threshold, the real-time agent status is determined to be abnormal.

[0058] If the score of the abnormal combination is less than or equal to the preset abnormal combination threshold, the real-time agent status is determined to be normal.

[0059] Thirdly, embodiments of this application provide an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0060] The memory stores instructions that the computer executes;

[0061] The processor executes computer-executable instructions stored in memory to implement the method as described in the first aspect or any of the above.

[0062] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in the first aspect or any of the above-mentioned methods.

[0063] Fifthly, embodiments of this application provide a computer program. The computer program product includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, it can implement the methods described in the first aspect or any of the above-mentioned methods.

[0064] The business processing method, apparatus, electronic device, and storage medium provided in this application embodiment first acquire business requirement data of the target business and historical business data corresponding to the target business. The business requirement data includes at least simulated user call volume, simulated agent number, and business simulation termination conditions. Then, based on the simulated user call volume and simulated agent number, a simulation implementation plan for the target business is initialized. The simulation implementation plan is then simulated using a digital twin model corresponding to the target customer service center until the business simulation termination conditions are met, yielding simulation results. By simulating the implementation plan using the digital twin model, the operational process can be rehearsed before the actual business is launched, identifying potential problems such as agent scheduling and call scheduling in advance, thus improving the efficiency and accuracy of business processing. Finally, based on the simulation results and historical business data, the implementation plan to be applied for the target business is determined. This technical solution includes the entire process from data acquisition, plan initialization, simulation verification to plan determination, and can dynamically adjust the plan according to the business scenario, improving the efficiency of business processing. Attached Figure Description

[0065] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0066] Figure 1 A flowchart illustrating the business processing method provided in the embodiments of this application. Figure 1 ;

[0067] Figure 2 A flowchart illustrating the business processing method provided in the embodiments of this application. Figure 2 ;

[0068] Figure 3 A flowchart illustrating the business processing method provided in the embodiments of this application. Figure 3 ;

[0069] Figure 4 This is a schematic diagram of the structure of the business processing apparatus provided in the embodiments of this application;

[0070] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0071] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0073] Before introducing the embodiments of this application, the technical terms involved in the embodiments of this application will be explained first:

[0074] Digital twins fully utilize data from physical models, sensors, and operational history to integrate multidisciplinary, multi-physical, multi-scale, and multi-probability simulation processes. This process maps data in a virtual space to reflect the entire lifecycle of the corresponding physical equipment, thereby enabling optimization, modification, or problem identification, and selecting the optimal solution through simulation and prediction.

[0075] Discrete Event Simulation (DES) is a method for predicting system changes based on the patterns of change of events at discrete points in time. DES abstracts the changes of a system over time into a series of events at discrete points in time, and evolves the system by processing these events in chronological order; it represents an event-driven simulation worldview.

[0076] Agent-based modeling (ABM) views a system as composed of multiple interacting agents, each with its own attributes and behavioral rules. It studies the dynamics and behavior of complex systems by simulating the interactions between these agents. It is a computational model used to simulate the actions and interactions of intelligent agents (independent organizations or groups) with autonomous consciousness.

[0077] The application background of the embodiments of this application will then be explained:

[0078] With the development of digital technology, automation has replaced some manual processes in routine customer service within the banking industry. However, a large number of customers still require human customer service to handle more complex issues. Therefore, optimizing the operation and management of customer service centers plays a crucial role in improving customer satisfaction and achieving efficient operation of banking services.

[0079] One example of a related technology is a branch operation management device based on digital twins, suitable for predicting the operation and customer behavior of bank branches over a future time period. Using the latest acquired branch operation data, prediction results are obtained through a target model. The prediction accuracy is calculated based on the prediction results and actual results, and the model is fine-tuned. Simulation and verification are then performed using the constructed branch digital twin model, saving operational costs and time.

[0080] Another example of related technology involves a method for visually monitoring the status of call center agents. A status data model is established based on the agent's status, and a scenario model is established based on the distribution of information such as the agent's location and the call center's corridors. The agent's status is obtained at regular intervals, and managers can select instructions to conduct quality inspections or skill queue allocation to monitor the operation of the call center.

[0081] Another example of related technology is a call service detection system, which uses a voice acquisition module to capture real-time call audio streams from mobile agents, a voice recognition module to analyze the voice to obtain the call content, and an intelligent quality inspection module to achieve automated real-time quality inspection during the call. The system provides timely reminders to agents to improve service quality by providing feedback on the quality inspection results.

[0082] In existing bank customer service center operations, there is often a heavy reliance on human experience. Statistical reports are used to analyze customer response times and agent scheduling. However, automatic responses to unexpected equipment malfunctions are not readily available, leading to a high dependence on manual handling and a lack of timeliness. With the development of digital technology in various banks, monitoring systems have been gradually established, improving real-time monitoring capabilities. However, because these monitoring systems are decentralized, they can only monitor specific areas and are difficult to integrate effectively. Therefore, there is still significant room for improvement in monitoring-based business processes.

[0083] In conclusion, improving the efficiency of business processing has become an urgent technical problem to be solved.

[0084] To address the technical problems existing in the prior art, the inventors of this application propose the following solution: For the issue of low efficiency in business processing, a 3D modeler is used to establish a digital twin model corresponding to the target customer service center. This data twin model includes hardware data, environmental data, agent data, and business data, providing multi-dimensional support for subsequent business processing. Based on the business requirements data of the target business, a simulation implementation plan for the target business is initialized. Then, the digital twin model is used to simulate the implementation plan, obtaining simulation results. Through simulation, dynamic simulation of user and agent services is achieved, improving the efficiency of business processing. Finally, based on the simulation results and historical business data, the proposed implementation plan for the target business is determined.

[0085] The parts not described in detail are disclosed in the following embodiments.

[0086] The technical solution of this application will now be described in detail through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0087] Figure 1 A flowchart illustrating the business processing method provided in the embodiments of this application. Figure 1 ,like Figure 1 As shown, the method may include the following steps:

[0088] Step 11: Obtain the business requirements data of the target business and the historical business data corresponding to the target business.

[0089] The business requirement data should include at least the number of simulated user calls, the number of simulated agents, and the conditions for terminating the business simulation.

[0090] In this step, by acquiring business demand data that is strongly related to customer service center operations in the target business, and by acquiring historical business data corresponding to the target business, we can provide data support for the validity verification of subsequent simulation results.

[0091] The simulated user call volume is determined based on historical call peaks and time period distribution patterns, while the simulated agent number is determined by combining agent configuration and shift saturation data in the agent information data. The business simulation termination condition can be set to a specific duration (such as a marketing campaign cycle) or a key indicator threshold (such as call processing completion rate).

[0092] Step 12: Based on the simulated user inbound volume and the simulated number of agent seats, initialize the simulation implementation plan for the target service.

[0093] In this step, based on the number of simulated user calls and the number of simulated agents, ABM technology is used to build a group of customer intelligent agents according to the number of simulated user calls. At the same time, user attributes and user behavior rules are configured. A group of agent agents is built according to the number of simulated agents, and attributes such as agent skill level and service duration rules are configured. The simulation implementation plan is initialized, and the business requirements are transformed into simulateable digital model parameters.

[0094] Step 13: Simulate the implementation of the simulation scheme using the digital twin model corresponding to the target customer service center until the business simulation termination conditions are met, and obtain the simulation results.

[0095] In this step, a digital twin model corresponding to the target customer service center is used to simulate the implementation of the simulation scheme. During the simulation, DES is used to manage global events according to the timeline, such as user call in, user abandonment, service termination, etc. ABM updates the agent status in real time according to the progress of global events, such as updating agent pressure value and customer waiting time, until the business simulation termination condition is reached. Finally, the simulation results are output, including equipment operation status, agent service indicators, customer experience data and agent scheduling strategy execution status.

[0096] Furthermore, the process of establishing the digital twin model corresponding to the target customer service center in step 13 includes the following implementation methods:

[0097] Step 1: Obtain the equipment layout data, environmental data, and customer service workstation data of the target customer service center.

[0098] In this implementation, terminal device collectors are used to acquire equipment layout data, environmental data, and customer service workstation data of the target customer service center, providing a data foundation for the subsequent establishment of a data twin model.

[0099] The equipment layout data includes equipment data and the location data of the equipment in space.

[0100] The data acquisition process for the aforementioned devices focuses on collecting communication data from routers and switches. Based on communication protocols, it collects CPU utilization, memory usage, and port traffic of core switches every 5 seconds, triggering DDoS attack alarms if abnormal usage occurs. Current sensors are used to collect real-time power consumption of voice switches and server racks, triggering equipment aging warnings if abnormal power consumption fluctuations occur.

[0101] The acquisition of the aforementioned environmental data focuses on collecting ambient noise levels in decibels, employing a microphone array for directional sound pickup, and assessing the impact of ambient noise on call quality. Humidity and temperature sensors are used to collect ambient humidity and temperature data in real time, intelligently adjusting the humidity and temperature of the customer service center to a comfortable range.

[0102] The process of acquiring data from the aforementioned customer service workstations focuses on collecting service information from agents. Cameras are used to capture facial information, microphones to acquire voice information, and heart rate monitoring wristbands to obtain real-time heart rate data. This allows for the real-time collection of agents' physical condition information and timely detection of abnormal states.

[0103] In addition, a system data collector is used to acquire customer service-related data from various channels, providing richer data for the subsequent establishment of the digital twin model. This data includes the following:

[0104] 1. Telephone Inbound Data Collection: The telephone inbound system mainly collects real-time telephone inbound volume, transfer to human operator, and current waiting volume. The main purpose is to obtain real-time call status.

[0105] 2. Customer tag data collection: After a customer calls, in order to refine operations, the customer's tag information will be queried based on the customer number. It is necessary to collect the customer tag information for each call.

[0106] 3. Agent Information Data Collection: Collect basic information about agents, such as skill group information, job information, and shift schedule information. Collect agent operation log information, and obtain the operation log information of the current call in real time, such as clicked knowledge base items.

[0107] 4. Work order information data collection: Collect work order information generated after the agent provides service, and pay attention to the customer's needs during the call for subsequent customer intent identification.

[0108] Step 2: Based on equipment layout data, environmental data, and customer service workstation data, establish a digital twin model corresponding to the target customer service center.

[0109] In this implementation, a 3D modeler is used to process equipment layout data and environmental data, and a behavior modeler is used to process customer service workstation data to obtain a digital twin model of the target customer service center.

[0110] The 3D modeler uses high-precision 3D visualization modeling technology to reproduce the physical entities of the customer service center in a virtual space, such as customer service workstations, equipment layout, floor environment, etc., and dynamically maps real-time status, such as call queues, equipment operating parameters, etc. The core objective of the behavior modeler is to simulate relevant personnel (such as agents or customers) and system status (such as incoming calls, manual transfers, queuing, etc.) within the customer service center.

[0111] The processing steps of the 3D modeler are as follows:

[0112] First, a LiDAR scanner is used to scan the central physical space, acquiring millimeter-level point cloud data. Interactive visualization modeling is then performed using Unity3D, and building information modeling (BIM) algorithms are used to model infrastructure such as power cabling. Next, topology mapping is used to bind specific physical devices like voice switches and agent terminals to virtual model components with unique IDs. Real-time data stream visualization mapping is set up, for example, mapping temperature and humidity to colors and adding relevant particle effects. Level of detail optimization (LMD) techniques are used to dynamically adjust model precision based on the user's perspective. A simplified model is used in the global view, focusing on high-precision loading of details in individual agent areas, thereby reducing the rendering load on the graphics processing unit (GPU) and alleviating modeling performance pressure.

[0113] The processing steps of the behavior modeler are as follows:

[0114] Using ABM (Abstract Aspect-Oriented Modeling) technology, customer agents are randomly modeled with defined attributes (e.g., easily complained customers), and customer behavior rules are set (e.g., triggering abandonment if waiting time exceeds a patience threshold). Simultaneously, agent agents are modeled with attributes, and agent behavior rules are set (e.g., agent proficiency levels, setting service durations based on proficiency levels, etc.). System agents are also configured (e.g., agent scheduling strategies simulating service users in the system).

[0115] Step 14: Based on the simulation results and historical business data, determine the implementation plan for the target business.

[0116] In this step, the intelligent decision-making unit compares the device status (such as peak load of virtual devices) in the simulation results with the device status in historical business data to determine whether the device status is normal. It also compares the agent status (such as peak pressure and service duration) in the simulation results with the agent status in historical business data to determine whether the agent status is normal. The simulation implementation schemes with normal device status and normal agent status are determined as the application implementation schemes for the target business.

[0117] The business processing method provided in this application first acquires business requirement data for the target business and corresponding historical business data. The business requirement data includes at least simulated user call volume, simulated agent number, and business simulation termination conditions. Then, based on the simulated user call volume and simulated agent number, a simulation implementation plan for the target business is initialized. The implementation plan is simulated using a digital twin model corresponding to the target customer service center until the business simulation termination conditions are met, yielding simulation results. Simulating the implementation plan using the digital twin model allows for a preview of the operational process before actual business implementation, identifying potential problems such as agent scheduling and call volume control in advance, thus improving the efficiency and accuracy of business processing. Finally, based on the simulation results and historical business data, the implementation plan to be applied to the target business is determined. This technical solution includes the entire process from data acquisition, plan initialization, simulation verification to plan determination, and can dynamically adjust the simulation plan according to the business scenario, improving the efficiency of business processing.

[0118] Based on the above embodiments, the simulation results include the simulation device status and simulation agent status generated during the simulation implementation process, as well as the simulation scheduling strategy adopted during the simulation implementation process. The historical business data includes at least the historical device status and historical agent status. Figure 2 A flowchart illustrating the business processing method provided in the embodiments of this application. Figure 2 ,like Figure 2 As shown, step 14 may include the following steps:

[0119] Step 21: Compare the simulated device status with the historical device status to obtain the first comparison result, and compare the simulated agent status with the historical agent status to obtain the second comparison result.

[0120] In this step, the status of the simulated equipment is compared with the status of the historical equipment. By calculating the deviation between the simulated equipment status data and the historical equipment status data, the first comparison result is obtained, which is used to determine whether the status of the simulated equipment is normal.

[0121] Furthermore, the simulated agent status (such as agent service duration, peak pressure, and heart rate fluctuation) is compared with the historical agent status to obtain a second comparison result, which is used to determine the rationality of the agent load and provide a precise basis for subsequent scheme optimization.

[0122] The simulation equipment status includes data such as CPU utilization, memory usage, port traffic of virtual network devices in the digital twin model, and power consumption fluctuations of server racks. The historical equipment status is the corresponding equipment operation data accumulated by the data acquisition unit under similar business scenarios.

[0123] Step 22: If the first comparison result is abnormal or the second comparison result is abnormal, adjust the simulation scheduling strategy in the simulation implementation plan and repeat the process of simulating the implementation plan by using the digital twin model corresponding to the target customer service center until both the first comparison result and the second comparison result are normal.

[0124] In this step, when the first comparison result is abnormal, optimize the equipment resource allocation strategy (e.g., increase the number of virtual servers, adjust network bandwidth allocation, etc.). When the second comparison result is abnormal, adjust the agent scheduling rules (e.g., add additional agents) or the incoming call allocation logic (e.g., prioritize assigning complex services to highly skilled agents). Repeat the process of simulating the implementation of the simulation scheme using a digital twin model corresponding to the target customer service center until both the first and second comparison results are normal.

[0125] Step 23: When both the first comparison result and the second comparison result are normal, the simulation implementation plan is determined as the implementation plan to be applied to the target business.

[0126] In this step, when both the first and second comparison results are normal, it indicates that the simulation implementation plan has passed multi-dimensional verification. At this time, the deviation between the simulation equipment status and the historical equipment operation benchmark is within a reasonable range, the simulation agent status meets the historical high-quality service standards (such as pressure value and service duration being within the normal range), and the simulation implementation plan can meet the resource requirements and service quality requirements of the target business. Therefore, the simulation implementation plan is identified as the implementation plan to be applied and can be used to guide the actual operation of the customer service center.

[0127] The business processing method provided in this application compares the simulated device status with the historical device status to obtain a first comparison result, and compares the simulated agent status with the historical agent status to obtain a second comparison result. This method can be used to verify the device status and agent status, preventing device overload or substandard service quality during actual implementation. When either the first or second comparison result is abnormal, the simulation scheduling strategy in the simulation implementation plan is adjusted, and the process of simulating the implementation plan using a digital twin model corresponding to the target customer service center is repeated until both the first and second comparison results are normal. When both the first and second comparison results are normal, the simulation implementation plan is determined as the application implementation plan for the target business. This mechanism of dynamically adjusting the simulation implementation plan eliminates the reliance on static historical experience, making the final implementation plan more closely match the actual business scenario and significantly improving the accuracy and reliability of business processing.

[0128] Figure 3 A flowchart illustrating the business processing method provided in the embodiments of this application. Figure 3 ,like Figure 3As shown, the method may also include the following steps:

[0129] Step 31: Obtain real-time operational data of the target customer service center.

[0130] In this step, real-time operational data such as hardware operating parameters, environmental data, business dynamic data, and agent status data are captured by the terminal device collector. This provides real-time updated data for the digital twin model, ensuring that the digital twin model can accurately reflect the current status of the physical customer service center.

[0131] Step 32: Synchronize the real-time running data to the digital twin model so that the digital twin model can simulate the operation based on the real-time running data.

[0132] In this step, different transmission strategies are used to synchronize real-time running data to the digital twin model, so that the digital twin model can simulate operation based on the real-time running data.

[0133] Furthermore, step 32 can be implemented as follows:

[0134] Step 1: When the real-time running data is high-priority data, the first data transmission method is used to synchronize the real-time running data to the digital twin model.

[0135] The first data transmission method includes any one of the following: a data transmission method based on the User Datagram Protocol (UDP), a data transmission method based on the Quick UDP Internet Connections (QUIC) protocol, and a data transmission method based on the Transmission Control Protocol (TCP).

[0136] Under this implementation, for high-priority data, it is necessary to ensure that the data is synchronized as soon as possible, and the first data transmission method is used to synchronize the real-time running data to the digital twin model.

[0137] For device information with high real-time requirements but relatively low packet loss rate requirements (such as device monitoring information), UDP protocol is used for data transmission. For data with relatively low real-time requirements but relatively high accuracy requirements (such as real-time work order data), QUIC protocol is used for transmission. This is a UDP-based transport layer protocol that improves data transmission accuracy and balances real-time performance and security. For data with low real-time requirements but high accuracy requirements (such as agent operation logs), TCP protocol is used directly for data transmission to ensure data accuracy.

[0138] Step 2: When the real-time running data is low-priority data, the second data transmission method is used to synchronize the real-time running data to the digital twin model.

[0139] The second data transmission method is a data transmission method based on incremental synchronization algorithm or adaptive algorithm.

[0140] In this implementation, for low-priority data, an incremental synchronization algorithm is used to detect significant changes in the data before data synchronization is performed, thereby reducing the amount of invalid data transmitted.

[0141] In one example, for low-priority data such as environmental temperature and humidity, an incremental synchronization algorithm is used to detect significant changes in temperature and humidity data. When significant changes are found, data is transmitted.

[0142] Simultaneously, an adaptive sampling algorithm is employed to adjust the sampling frequency, reducing the transmission of invalid data while ensuring security. For example, when the ambient temperature fluctuation is less than ±1℃, the sampling frequency is reduced, increasing the sampling interval to 30 seconds or 1 minute. When a sudden temperature change is detected, a 1-second high-frequency sampling mode is immediately switched.

[0143] Furthermore, after step 32, the business processing method may also include the following implementation steps:

[0144] S1. Determine the real-time device status based on the deviation comparison between real-time device data and historical device data in the real-time operation data.

[0145] Under this implementation, real-time device data in the real-time operation data is compared with historical device data to determine whether the calculated deviation exceeds a threshold. If the deviation is within the threshold range, the device status is determined to be normal. If it exceeds the threshold, it is marked as abnormal and an alarm is triggered. This enables real-time monitoring and risk prediction of device operation status, ensuring the stable operation of the customer service center hardware system.

[0146] Real-time device data includes network device CPU utilization, memory usage, port traffic, and server rack power consumption.

[0147] S2. Determine the real-time agent status based on the anomaly detection results of the real-time agent status data and historical agent status data in the real-time operation data.

[0148] Under this implementation, the real-time agent status data in the real-time operation data is compared with the historical agent status data to determine whether the calculated deviation exceeds the threshold. If the deviation is within the threshold range, the agent status is determined to be normal. If it exceeds the threshold, it is marked as abnormal and an alarm is issued. This enables accurate perception and timely intervention of agent service status, ensuring service quality.

[0149] The aforementioned real-time agent status data includes agent heart rate data, agent facial data, and agent call content. Step S2 may include the following implementation methods:

[0150] Step 1: Obtain the peak stress level of agents based on their heart rate data, obtain the agent's emotional value based on their facial data, and obtain the agent's call efficiency based on the content of their calls.

[0151] In this implementation, a Long Short-Term Memory Attention Mechanism is used to process time-series heart rate data for agents. A sliding window (30-second window + 5-second step) is used to extract time-domain and frequency-domain features. A dynamic baseline calibration algorithm is introduced to automatically adjust the abnormal threshold based on the individual agent's resting heart rate, reducing the false alarm rate due to individual differences, and ultimately obtaining the peak agent stress level. A facial landmark tracking system is constructed based on a 50-layer ResNet-50 network. Optical flow is used to calculate dynamic features such as eyebrow spacing and mouth corner curvature. A temporal convolutional network is used to analyze facial expression trends to obtain agent emotion values. Real-time text-to-text transcription of agent calls is performed, and a Transformer-based Bidirectional Encoder Representations from Transformers (BERT) is used to analyze the call content and obtain agent call efficiency.

[0152] Step 2: If any of the agent stress peak, agent mood value, or agent call efficiency exceeds the corresponding preset abnormal threshold, calculate the abnormal combination score of agent stress peak, agent mood value, and agent call efficiency.

[0153] Under this implementation, millisecond-level data synchronization of agent heart rate data, agent facial data, and agent call content is achieved through the Precise Time Protocol (PTP). A spatiotemporal alignment matrix is ​​constructed to unify the dimensions of the three-modal data (i.e., agent heart rate data, agent facial data, and agent call content) to a unified standard. When any of the corresponding agent stress peak, agent emotion value, or agent call efficiency exceeds the corresponding preset abnormal threshold, the three-modal collaborative cross-validation is triggered, and the abnormal combination score is calculated according to the preset weight rules.

[0154] Step 3: If the score of the abnormal combination is greater than the preset abnormal combination threshold, the real-time agent status is determined to be abnormal.

[0155] In this implementation, when the score of an abnormal combination is greater than the preset abnormal combination threshold (e.g., reaching the agent fatigue threshold or the agent passive service threshold), the real-time agent status is determined to be abnormal, and an agent status abnormality alarm is generated on the agent end and the management end to make auxiliary decisions (e.g., adding agents).

[0156] Step 4: If the score of the abnormal combination is less than or equal to the preset abnormal combination threshold, the real-time agent status is determined to be normal.

[0157] Under this implementation, when the score of abnormal combination is less than or equal to the preset abnormal combination threshold, it indicates that the actual status of the agent is within a reasonable range in the multi-dimensional monitoring, and the real-time agent status is determined to be normal.

[0158] The business processing method provided in this application first acquires real-time operational data of the target customer service center and synchronizes the real-time operational data to a digital twin model, enabling the digital twin model to simulate operation based on the real-time operational data. This technical solution realizes the linkage between real-time operational data and the virtual model, forming a real-time feedback chain of "physical acquisition—synchronization to digital twin model—simulation analysis—actual adjustment," enabling the customer service center to dynamically optimize resource allocation and scheduling strategies based on the simulation results of the virtual model, thereby improving the efficiency of business processing.

[0159] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0160] Figure 4 This is a schematic diagram of the structure of the service processing apparatus provided in an embodiment of this application. Figure 4 As shown, the device includes:

[0161] Module 41 is used to acquire business requirement data of the target business and historical business data corresponding to the target business; the business requirement data includes at least the number of simulated user calls, the number of simulated agents, and the business simulation termination conditions.

[0162] Processing module 42 is used to initialize the simulation implementation scheme of the target service based on the number of simulated user incoming calls and the number of simulated agents;

[0163] Implementation module 43 is used to simulate the implementation of the simulation implementation plan through the digital twin model corresponding to the target customer service center until the business simulation termination condition is reached and the simulation results are obtained.

[0164] The determination module 44 is used to determine the implementation plan to be applied to the target service based on simulation results and historical business data.

[0165] In one possible implementation, the simulation results include the simulation device status and simulation agent status generated during the simulation implementation process, as well as the simulation scheduling strategy adopted during the simulation implementation process, and the historical business data includes at least the historical device status and historical agent status.

[0166] Module 44 is specifically used for:

[0167] The simulated device status is compared with the historical device status to obtain the first comparison result, and the simulated agent status is compared with the historical agent status to obtain the second comparison result.

[0168] If either the first comparison result or the second comparison result is abnormal, adjust the simulation scheduling strategy in the simulation implementation plan and repeat the process of simulating the implementation plan using the digital twin model corresponding to the target customer service center until both the first and second comparison results are normal.

[0169] If both the first and second comparison results are normal, the simulation implementation plan will be determined as the application implementation plan for the target business.

[0170] In one possible implementation, the process of establishing a digital twin model corresponding to the target customer service center is specifically implemented by module 43 as follows:

[0171] Obtain equipment layout data, environmental data, and customer service workstation data for the target customer service center;

[0172] Based on equipment layout data, environmental data, and customer service workstation data, a digital twin model corresponding to the target customer service center is established.

[0173] In one possible implementation, the acquisition module 41 is further configured to;

[0174] Real-time operational data is synchronized to the digital twin model so that the digital twin model can simulate operation based on the real-time operational data.

[0175] In one possible implementation, real-time runtime data is synchronized to the digital twin model, and module 41 is specifically used for:

[0176] When the real-time running data is high-priority data, the first data transmission method is used to synchronize the real-time running data to the digital twin model;

[0177] When the real-time running data is low-priority data, the second data transmission method is used to synchronize the real-time running data to the digital twin model;

[0178] The first data transmission method includes any one of the following: data transmission method based on the User Datagram Protocol (UDP), data transmission method based on the Fast UDP Internet Protocol (QUIC), and data transmission method based on the Transmission Control Protocol (TCP); the second data transmission method is a data transmission method based on an incremental synchronization algorithm or an adaptive adoption algorithm.

[0179] In one possible implementation, after synchronizing the real-time runtime data to the digital twin model, the determination module 44 is further configured to:

[0180] The real-time device status is determined by comparing the deviation between real-time device data and historical device data in the real-time operation data.

[0181] The real-time agent status is determined based on the anomaly detection results of the real-time agent status data and historical agent status data in the real-time operation data.

[0182] In one possible implementation, the real-time agent status data includes agent heart rate data, agent facial data, and agent call content. The determination module 44 is specifically used for:

[0183] The peak stress level of agents is obtained from agent heart rate data, the agent emotion value is obtained from agent facial data, and the agent call efficiency is obtained from agent call content.

[0184] If any of the peak agent stress, agent mood value, or agent call efficiency exceeds the corresponding preset abnormal threshold, calculate the abnormal combination score of the peak agent stress, agent mood value, and agent call efficiency.

[0185] If the score of the abnormal combination is greater than the preset abnormal combination threshold, the real-time agent status is determined to be abnormal.

[0186] If the score of the abnormal combination is less than or equal to the preset abnormal combination threshold, the real-time agent status is determined to be normal.

[0187] The apparatus provided in this application embodiment can be used to execute the determination method in any of the above embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0188] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing element calls; they can be fully implemented in hardware; or some modules can be implemented in software via processing element calls, while others are implemented in hardware. Additionally, these modules can be fully or partially integrated together, or implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. During implementation, each step of the above method or each of the above modules can be completed through the integrated logic circuits in the hardware of the processor element or through software instructions.

[0189] Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 5 As shown, the electronic device may include: a processor 51, a memory 52, and computer program instructions stored in the memory 52 and executable on the processor 51. When the processor 51 executes the computer program instructions, it implements the method provided in any of the foregoing embodiments.

[0190] Optionally, the various components of the electronic device can be connected via a system bus.

[0191] The memory 52 can be a separate storage unit or a storage unit integrated into the processor 51. The number of processors 51 can be one or more.

[0192] It should be understood that the processor 51 can be a Central Processing Unit (CPU), or other general-purpose processors 51, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor 51 can be a microprocessor 51, or any conventional processor 51. The steps of the method disclosed in this application can be directly manifested as being executed by the hardware processor 51, or being executed by a combination of hardware and software modules within the processor 51.

[0193] The system bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the figure, but this does not indicate that there is only one bus or one type of bus. Memory 52 may include random access memory (RAM) 52, and may also include non-volatile memory (NVM) 52, such as at least one disk storage device 52.

[0194] All or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable memory 52. ​​When the program is executed, it performs the steps of the above method embodiments; and the aforementioned memory 52 (storage medium) includes: read-only memory 52 (ROM), RAM, flash memory 52, hard disk, solid-state hard disk, magnetic tape, floppy disk, optical disk, and any combination thereof.

[0195] The electronic device provided in this application embodiment can be used to execute the method provided in any of the above method embodiments. Its implementation principle and technical effect are similar, and will not be repeated here.

[0196] This application provides a computer-readable storage medium storing computer instructions that, when executed on a computer, cause the computer to perform the above-described method.

[0197] The aforementioned computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0198] Optionally, a readable storage medium can be coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Alternatively, the readable storage medium can be an integral part of the processor. Both the processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components within the device.

[0199] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and the at least one processor can implement the above-described method when executing the computer program.

[0200] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A business processing method, characterized in that, include: Obtain business requirement data for the target business and historical business data corresponding to the target business; the business requirement data shall include at least the number of simulated user calls, the number of simulated agents, and the conditions for terminating the simulated business. Based on the simulated user inbound traffic and the simulated agent number, initialize the simulation implementation scheme for the target service; The simulation implementation scheme is simulated using a digital twin model corresponding to the target customer service center until the business simulation termination condition is met, and the simulation results are obtained. Based on the simulation results and the historical business data, the proposed implementation scheme for the target business is determined.

2. The method according to claim 1, characterized in that, The simulation results include the simulation device status and simulation agent status generated during the simulation implementation process, as well as the simulation scheduling strategy adopted during the simulation implementation process. The historical business data includes at least the historical device status and historical agent status. The step of determining the application implementation scheme for the target service based on the simulation results and the historical service data includes: The simulated device state is compared with the historical device state to obtain a first comparison result, and the simulated agent state is compared with the historical agent state to obtain a second comparison result. If the first comparison result is abnormal or the second comparison result is abnormal, adjust the simulation scheduling strategy in the simulation implementation scheme and repeat the process of simulating the implementation scheme using a digital twin model corresponding to the target customer service center until both the first comparison result and the second comparison result are normal. When both the first comparison result and the second comparison result are normal, the simulation implementation scheme is determined as the application implementation scheme for the target service.

3. The method according to claim 1, characterized in that, The process of establishing a digital twin model corresponding to the target customer service center includes: Obtain the equipment layout data, environmental data, and customer service workstation data of the target customer service center; Based on the equipment layout data, the environmental data, and the customer service workstation data, a digital twin model corresponding to the target customer service center is established.

4. The method according to claim 1, characterized in that, The method further includes: Obtain the real-time operational data of the target customer service center; The real-time operating data is synchronized to the digital twin model so that the digital twin model can perform simulated operation based on the real-time operating data.

5. The method according to claim 4, characterized in that, The step of synchronizing the real-time operational data to the digital twin model includes: When the real-time running data is high-priority data, the first data transmission method is used to synchronize the real-time running data to the digital twin model; When the real-time running data is low-priority data, a second data transmission method is used to synchronize the real-time running data to the digital twin model; The first data transmission method includes any one of the following: a data transmission method based on the User Datagram Protocol (UDP), a data transmission method based on the Fast UDP Internet Protocol (QUIC), and a data transmission method based on the Transmission Control Protocol (TCP); the second data transmission method is a data transmission method based on an incremental synchronization algorithm or an adaptive adoption algorithm.

6. The method according to claim 4, characterized in that, After synchronizing the real-time operational data to the digital twin model, the method further includes: The real-time device status is determined based on the comparison results of the deviation between the real-time device data and the historical device data in the real-time operation data; The real-time agent status is determined based on the anomaly detection results of the real-time agent status data and the historical agent status data in the real-time operation data.

7. The method according to claim 4, characterized in that, The real-time agent status data includes agent heart rate data, agent facial data, and agent call content; determining the real-time agent status based on the anomaly detection results of the real-time agent status data and historical agent status data in the real-time operation data includes: The peak pressure of the agent is obtained based on the agent's heart rate data, the agent's emotional value is obtained based on the agent's facial data, and the agent's call efficiency is obtained based on the agent's call content. If any one of the agent stress peak, agent emotion value, or agent call efficiency exceeds the corresponding preset abnormal threshold, calculate the abnormal combination score of the agent stress peak, agent emotion value, and agent call efficiency. If the score of the abnormal combination is greater than the preset abnormal combination threshold, the real-time agent status is determined to be an abnormal status. If the score of the abnormal combination is less than or equal to the preset abnormal combination threshold, the real-time agent status is determined to be normal.

8. A business processing apparatus, characterized in that, include: The acquisition module is used to acquire business requirement data of the target business and historical business data corresponding to the target business; the business requirement data includes at least the number of simulated user calls, the number of simulated agents, and the business simulation termination conditions. The processing module is used to initialize the simulation implementation scheme of the target service based on the number of simulated user calls and the number of simulated agents; The implementation module is used to simulate the implementation of the simulation scheme through the digital twin model corresponding to the target customer service center until the business simulation termination condition is reached, and obtain the simulation result. The determination module is used to determine the application implementation scheme of the target service based on the simulation results and the historical service data.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 7.

Images