Intelligent service linkage queuing and calling method based on AI learning model

Abstract

The invention discloses an intelligent service linkage queuing number calling method based on an AI learning model. The method comprises the following steps: a transacter obtains a unique queuing number through a queuing number taking module; the queuing management module is used for managing the transaction processes of all transactions and dynamically informing the transactors of transaction recommendation sequences of all the transactions, the currently available optimal waiting window and the predicted waiting time in real time; in the transaction handling process, if a transacter cannot arrive at a handling window in time, a queuing management module intelligently adjusts a reservation number calling queue of subsequent transactions according to the actual transaction handling progress and the situation of a queuing resource pool of remaining transactions; and if the transacter needs to cancel the transaction of a certain transaction, the queuing management module intelligently adjusts the waiting window of the subsequent transaction and the required waiting time. The invention relates to the technical field of artificial intelligence scheduling, and can solve the problem that in the prior art, when a transacter transacts multiple transactions at the same time, conflicts or number missing is likely to happen.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence scheduling technology, and in particular to an intelligent business linkage queuing calling method based on an AI learning model. Background Art

[0002] The queuing and calling system currently in common use in government halls can dynamically allocate the order of calling numbers through intelligent queuing and calling scheduling. At the same time, it can display the calling information in real time through large screens, electronic screens or voice broadcasts, so as to guide the public to handle affairs in an orderly manner.

[0003] When the public is handling a matter, they can make an appointment and take a number through the existing queuing system. The system will automatically assign waiting numbers and waiting windows through a queuing algorithm, and automatically call the public to go to the service window in a timely manner to handle the matter.

[0004] However, if the public needs to handle multiple matters at the same time, they need to make an appointment and take a number for each matter they need to handle. Since it is impossible to know the queue of each matter and the busyness of the service window, if all appointments are made at once, there will be situations where different matters will be called at the same time and conflict with each other or some matters will be missed; if the matters are made and handled in sequence, it will take a long time to handle multiple matters and the efficiency will be very low.

[0005] In the case where a transaction includes multiple related transactions, the public needs to go to multiple windows to handle the transactions. The existing queuing and calling system cannot allocate numbers for multiple related matters at the same time, which may also lead to conflicts or missed numbers in the handling of various related transactions. Therefore, there is a need for an intelligent business linkage queuing and calling method based on an AI learning model, which can solve the problem of conflicts or missed numbers when the handlers handle multiple transactions at the same time in the existing technology. Summary of the invention

[0006] The purpose of the present invention is to provide an intelligent business linkage queuing and calling method based on an AI learning model, which can solve the problem in the prior art that conflicts or missed numbers easily occur when the handlers handle multiple matters at the same time.

[0007] The present invention is achieved in that:

[0008] An intelligent business linkage queuing calling method based on an AI learning model comprises the following steps:

[0009] Step 1: The applicant obtains a unique queue number through the queue number collection module;

[0010] Step 2: The queue management module manages the processing progress of all transactions and dynamically informs the transaction person of the recommended processing order of each transaction, the best waiting window currently available and the estimated waiting time;

[0011] Step 3: During the transaction process, if the person who handles the transaction fails to arrive at the processing window in time, the queue management module intelligently adjusts the appointment call queue for subsequent transactions based on the actual progress of the transaction and the queue resource pool of the remaining transactions;

[0012] Step 4: During the transaction processing, if the person handling the transaction needs to cancel the processing of a certain transaction, the queue management module intelligently adjusts the waiting window and the required waiting time for subsequent transactions.

[0013] In step 1, the affairs that the handler needs to handle are one or more affairs. When handling multiple affairs, each affair shares a queue number.

[0014] In the step 1, when the handler takes a number through the queuing number taking module, the queuing number taking module obtains the basic information of the handler, queries the historical records of the handler's previous transactions in the system, recommends and displays a list of possible transactions to the handler, and the handler selects the transactions he needs to handle.

[0015] In the step 2, if the transaction handled by the handler is one item, the queue management module informs the handler of the queue status and estimated waiting time based on the on-site queue situation of the processing window for handling the transaction.

[0016] In step 2, if the person handling multiple matters, the queue management module analyzes the handling process of each matter and the sequence dependency between the matters based on the multiple matters selected by the person handling the matters, and combines the queue situation of the on-site handling windows to predict the optimal transaction processing order through the AI ​​learning model, and recommends it to the person handling the matters as the recommended handling order of the multiple matters, and displays the handling windows corresponding to each matter and the navigation routes of each handling window to the person handling the matters.

[0017] The step 2 comprises the following sub-steps:

[0018] Step 2.1: The queue management module analyzes the flow of people and waiting time at each service window, intelligently adjusts the order of the services that the applicant needs to handle and the best service window that can be visited at the moment, and estimates the waiting time corresponding to each service and the time required to complete all services;

[0019] Step 2.2: The queue management module obtains the processing order of each transaction and its estimated waiting time, as well as the geographical location of each processing window, intelligently plans the walking route for the person handling all transactions, and estimates the completion time of all transactions;

[0020] Step 2.3: During the transaction handling and waiting process, the queue management module provides the handler with information on the handling progress and time nodes.

[0021] The step 2.1 comprises the following sub-steps:

[0022] Step 2.1.1: The queue management module obtains the on-site real-time queue resource pool corresponding to the multiple transactions, the handling process of each transaction, the transaction preferences and historical records of the current queue handlers at each handling window, and the sequence dependency data between the transactions;

[0023] Step 2.1.2: For each service window, based on the historical records of the current queued service personnel and the real-time queue number on site, analyze the flow distribution and change trend of the service window, and predict the current waiting time required for each service window based on the flow distribution and change trend of each service window;

[0024] Step 2.1.3: Dynamically calculate the priority of each handler based on the urgency of each matter, the handler’s VIP level, and waiting time;

[0025] Step 2.1.4: Allocate applicants to different application windows based on the current queue status of each application window and the applicant’s priority;

[0026] Step 2.1.5: Based on the current waiting time and transaction processing time of each service window, as well as the user's current location, recommend the best service window to the service user.

[0027] In step 2.1.4, complex matters or matters requiring special processing are assigned to the processing window where relevant experienced staff are located.

[0028] The step 3 comprises the following sub-steps:

[0029] Step 3.1: The applicant shares the real-time location of the mobile phone with the queue management module, and the queue management module determines the distance between the applicant and the processing window;

[0030] Step 3.2: Update the transaction processing progress of each processing window in real time and dynamically adjust the time for the processing person to arrive at the processing window;

[0031] Step 3.3: Send reminder information to the applicant in advance to remind him / her of the waiting time at the next processing window and the current queue progress;

[0032] Step 3.4: If the applicant fails to arrive at the window on time, the queue management module will schedule the applicant to the next optimal time slot or processing window;

[0033] Step 3.5: Synchronously update the processing window and required waiting time for the remaining unprocessed matters of the handler.

[0034] In step 4, after receiving the request from the handler to cancel the transaction, the queue management module analyzes the processing flow of the transaction required by the handler and warns of the risk of canceling the transaction. After the handler determines that there is no risk impact, the corresponding queue number of the handler's transaction in the queue resource pool is deleted, and the processing window and required waiting time of subsequent transactions are adjusted according to the current queue situation of the queue resource pool.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. The present invention calculates one or more matters that the handler needs to handle based on the AI ​​learning model through the queue management module, and intelligently adjusts the queue order by using the current queue situation of the queue resource pool, the transaction processing progress of the processing window, the handler's priority and transaction list, the handler's habits and needs, etc., and intelligently allocates the processing window, recommends the processing order and navigation route to the handler, provides personalized services, optimizes the service process, improves the handler's satisfaction while improving service efficiency, and reduces operating costs and management difficulty.

[0037] 2. The present invention provides a unique queue number through the queue number taking module, which is convenient for the queue management module to manage the linkage sorting of various affairs of the handler in the queue resource pool. Even if the handler cannot arrive at the processing window in time or the handler temporarily cancels the processing of the transaction, the queue status of the queue resource pool can be dynamically updated in real time to avoid cumulative number overflow and uneven window busyness, thereby improving the utilization rate of the processing window. The handler only needs to make an appointment to take a number once to handle multiple affairs, and there will be no conflict or number overflow between the various affairs, thus reducing the waiting time of the handler, and can effectively improve the handling efficiency of the handler when handling multiple affairs or multiple related affairs at the same time. DETAILED DESCRIPTION

[0038] An intelligent business linkage queuing calling method based on an AI learning model comprises the following steps:

[0039] Step 1: The applicant obtains a unique queue number through the queue number collection module.

[0040] The traditional queuing number-taking system can provide a variety of number-taking methods, including on-site number-taking machines, mobile phone APPs, WeChat applets, etc. The queuing number-taking module can also adopt the above-mentioned existing number-taking methods, which will not be repeated here.

[0041] In step 1, the affairs that the handler needs to handle can be one or more affairs. When handling multiple affairs, each affair shares a queue number.

[0042] When the person handling multiple matters handles multiple matters, the same queue number is used to facilitate the sorting and management of the queue numbers.

[0043] In the step 1, when the handler takes a number through the queuing number taking module, the queuing number taking module obtains the basic information of the handler, queries the historical records of the handler's previous transactions in the system, recommends and displays a list of possible transactions to the handler, and the handler selects the transactions he needs to handle.

[0044] The person handling the matter can select the matters he needs to handle through the matter list, which makes the operation more convenient and quick. The person handling the matter can also select the matters he needs to handle through the total matter list.

[0045] Step 2: The queue management module manages the processing progress of all transactions, and dynamically informs the person handling the transactions in real time of the recommended processing order of each transaction, the best waiting window currently available and the estimated waiting time.

[0046] In the step 2, if the transaction handled by the handler is one item, the queue management module informs the handler of the queue status and estimated waiting time based on the on-site queue situation of the processing window for handling the transaction.

[0047] Specifically, the queue management module calculates the estimated waiting time based on the number of people in the queue and the average processing time of the transaction, and displays the number of people in the queue and the estimated waiting time to the person handling the transaction.

[0048] In the above step 2, if the person handling multiple transactions (i.e. two or more transactions, or one transaction contains two or more related transactions), the queue management module analyzes the handling process of each transaction and the sequence dependency between the transactions based on the multiple transactions selected by the person handling the transaction, and predicts the optimal transaction processing order through the AI ​​learning model in combination with the queue situation of the on-site processing windows. The module recommends the optimal transaction processing order to the person handling the transaction as the recommended processing order for the multiple transactions, and displays the processing windows corresponding to each transaction and the navigation routes of each processing window to the person handling the transaction.

[0049] Preferably, an AI learning model can be established in the queue management module based on a deep neural network, the input items of the AI ​​learning model are the various affairs that the handler needs to handle, and the output items of the AI ​​learning model are the recommended order for handling the various affairs.

[0050] After predicting the recommended order for handling multiple matters through the AI ​​learning model, a navigation route is formed according to the locations of each handling window corresponding to the recommended order, so as to guide the person to handle the corresponding matters at each handling window in turn according to the navigation route.

[0051] The step 2 comprises the following sub-steps:

[0052] Step 2.1: The queue management module analyzes the flow of people and waiting time at each service window, intelligently adjusts the order of the services that the applicant needs to handle and the best service window that can be visited at the moment, and estimates the waiting time corresponding to each service and the time required to complete all services.

[0053] The step 2.1 comprises the following sub-steps:

[0054] Step 2.1.1: The queue management module obtains the resource pool of the on-site real-time queues corresponding to the multiple transactions (i.e., obtains the number of people queuing at the processing window corresponding to each transaction), the processing flow of each transaction (i.e., obtains the complexity of processing each transaction, which is used to estimate the processing time of the transaction), the processing preferences and historical records of the current queued persons at each processing window (used to improve the accuracy of the estimated processing time of the transaction), the sequence dependency between the transactions (used to adjust the processing order of the transactions), and other data.

[0055] Step 2.1.2: For each service window, analyze the flow of people and changing trends at the service window based on the historical records of the current queued service personnel and the real-time queue number on site, and predict the current waiting time required for each service window based on the flow of people and changing trends at each service window.

[0056] Step 2.1.3: Dynamically calculate the priority of each handler based on factors such as the urgency of each matter, the handler’s VIP level, and waiting time.

[0057] Step 2.1.4: Assign applicants to different processing windows based on their current queue status and priority.

[0058] In step 2.1.4, complex matters or matters requiring special processing are assigned to processing windows where relevant experienced staff are located to improve the efficiency of transaction processing.

[0059] Step 2.1.5: Based on the current waiting time and transaction processing time of each service window, as well as the user's current location, recommend the best service window to the service user, that is, the service window that is currently accessible and has the shortest waiting time.

[0060] Step 2.2: The queue management module obtains the processing order of each transaction and its estimated waiting time, as well as the geographical location of each processing window, intelligently plans the walking route for the person handling all transactions, and estimates the completion time of all transactions.

[0061] In step 2.2, the queue management module plans the route according to the sequence of the multiple matters that the applicant needs to handle and the locations of the corresponding handling windows, and recommends the shortest route to the applicant.

[0062] Step 2.3: During the transaction handling and waiting process, the queue management module provides the handler with information on the handling progress and time nodes.

[0063] Preferably, the processing progress and time node information can be provided to the applicant through voice calls, text messages, WeChat public accounts, WeChat mini-programs, etc. The applicant can choose the method of obtaining information according to actual needs.

[0064] Step 3: During the transaction process, if the person handling the transaction fails to arrive at the processing window in time, the queue management module will intelligently adjust the appointment call queue for subsequent transactions based on the actual progress of the transaction and the queue resource pool for the remaining transactions to avoid the situation of cumulative number overflow.

[0065] The step 3 comprises the following sub-steps:

[0066] Step 3.1: The handler shares the real-time location of the mobile phone with the queue management module, and the queue management module determines the distance between the handler and the handling window.

[0067] Step 3.2: Update the transaction progress of each processing window in real time, and dynamically adjust the time for the applicant to arrive at the processing window. This can prevent the applicant from arriving at the processing window too early and increasing the waiting time, and can also prevent the applicant from missing the processing window due to failure to arrive at the processing window in time.

[0068] Step 3.3: Send reminder information to the applicant in advance to remind him / her of the waiting time for the next processing window and the current queue progress.

[0069] Preferably, a reminder message can be sent to the applicant in advance based on the distance between the applicant and the processing window and the time required to reach the processing window, so as to ensure that the applicant can arrive at the processing window in time to handle the transaction. The reminder message can be sent via WeChat applet, SMS, phone call, etc.

[0070] Step 3.4: If the applicant fails to arrive at the window on time, the queue management module will schedule him / her to the next optimal time period or processing window.

[0071] For example: if the distance between the handler and the service window is close, the handler's queue number can be moved back one place, or the handler can be scheduled to other service windows that will be free and can handle related matters; if the distance between the handler and the service window is far, the handler's position in the queue resource pool can be readjusted according to the handler's distance and arrival time.

[0072] Step 3.5: Synchronously update the processing window and required waiting time for the remaining unprocessed matters of the handler.

[0073] The calculation process of the remaining unprocessed transaction processing window and the required waiting time is the same as the calculation process in step 2, which will not be repeated here. After the synchronous update is completed, the new processing recommendation order and navigation route are resent to the handler so that the handler can adjust the route in time to avoid affecting the transaction processing order. The AI ​​learning model is used to intelligently re-sort the remaining unprocessed transactions, and the queue order of the queue resource pool is dynamically adjusted to ensure the maximization of service efficiency and processing efficiency.

[0074] Step 4: During the transaction processing, if the person handling the transaction needs to cancel the processing of a certain transaction, the queue management module intelligently adjusts the waiting window and the required waiting time for subsequent transactions.

[0075] In step 4, after receiving the request from the handler to cancel the transaction, the queue management module analyzes the processing flow of the transaction required by the handler and warns of the risk of canceling the transaction. After the handler determines that there is no risk impact, the corresponding queue number of the handler's transaction in the queue resource pool is deleted, and the processing window and required waiting time of subsequent transactions are adjusted according to the current queue situation of the queue resource pool.

[0076] The calculation process of the waiting window and the required waiting time for subsequent transactions is the same as that in step 2, and will not be repeated here. The AI ​​learning model is used to intelligently re-sort subsequent transactions and dynamically adjust the queue order of the queue resource pool to ensure maximum service efficiency and processing efficiency.

[0077] Preferably, the person handling the transaction can submit a request to cancel a transaction through WeChat mini-programs, phone calls, etc., and at the same time confirm whether there is any risk in canceling the transaction.

[0078] The above are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent business linkage queuing method based on an AI learning model, characterized by: The following steps are involved: Step 1: The applicant obtains a unique queue number through the queue number collection module; Step 2: The queue management module manages the processing progress of all transactions and dynamically informs the transaction person of the recommended processing order of each transaction, the best waiting window currently available and the estimated waiting time; Step 3: During the transaction process, if the person who handles the transaction fails to arrive at the processing window in time, the queue management module intelligently adjusts the appointment call queue for subsequent transactions based on the actual progress of the transaction and the queue resource pool of the remaining transactions; Step 4: During the transaction processing, if the person handling the transaction needs to cancel the processing of a certain transaction, the queue management module intelligently adjusts the waiting window and the required waiting time for subsequent transactions.

2. According to the intelligent business linkage queuing calling method based on AI learning model in claim 1, it is characterized by: In step 1, the affairs that the handler needs to handle are one or more affairs. When handling multiple affairs, each affair shares a queue number.

3. The intelligent service linkage queuing calling method based on AI learning model according to claim 1 or 2 is characterized in that: In the step 1, when the handler takes a number through the queuing number taking module, the queuing number taking module obtains the basic information of the handler, queries the historical records of the handler's previous transactions in the system, recommends and displays a list of possible transactions to the handler, and the handler selects the transactions he needs to handle.

4. The intelligent business linkage queuing calling method based on AI learning model according to claim 1 is characterized by: In the step 2, if the transaction handled by the handler is one item, the queue management module informs the handler of the queue status and estimated waiting time based on the on-site queue situation of the processing window for handling the transaction.

5. The intelligent business linkage queuing calling method based on AI learning model according to claim 1 is characterized by: In step 2, if the person handling multiple matters, the queue management module analyzes the handling process of each matter and the sequence dependency between the matters based on the multiple matters selected by the person handling the matters, and combines the queue situation of the on-site handling windows to predict the optimal transaction processing order through the AI ​​learning model, and recommends it to the person handling the matters as the recommended handling order of the multiple matters, and displays the handling windows corresponding to each matter and the navigation routes of each handling window to the person handling the matters.

6. The intelligent service linkage queuing calling method based on AI learning model according to claim 1 or 5 is characterized in that: The step 2 comprises the following sub-steps: Step 2.1: The queue management module analyzes the flow of people and waiting time at each service window, intelligently adjusts the order of the services that the applicant needs to handle and the best service window that can be visited at the moment, and estimates the waiting time corresponding to each service and the time required to complete all services; Step 2.2: The queue management module obtains the processing order of each transaction and its estimated waiting time, as well as the geographical location of each processing window, intelligently plans the walking route for the person handling all transactions, and estimates the completion time of all transactions; Step 2.3: During the transaction handling and waiting process, the queue management module provides the handler with information on the handling progress and time nodes.

7. The intelligent service linkage queuing calling method based on AI learning model according to claim 6 is characterized by: The step 2.1 comprises the following sub-steps: Step 2.1.1: The queue management module obtains the on-site real-time queue resource pool corresponding to the multiple transactions, the handling process of each transaction, the transaction preferences and historical records of the current queue handlers at each handling window, and the sequence dependency data between the transactions; Step 2.1.2: For each service window, based on the historical records of the current queued service personnel and the real-time queue number on site, analyze the flow distribution and change trend of the service window, and predict the current waiting time required for each service window based on the flow distribution and change trend of each service window; Step 2.1.3: Dynamically calculate the priority of each handler based on the urgency of each matter, the handler’s VIP level, and waiting time; Step 2.1.4: Allocate applicants to different application windows based on the current queue status of each application window and the applicant’s priority; Step 2.1.5: Based on the current waiting time and transaction processing time of each service window, as well as the user's current location, recommend the best service window to the service user.

8. The intelligent service linkage queuing calling method based on AI learning model according to claim 7 is characterized by: In step 2.1.4, complex matters or matters requiring special processing are assigned to the processing window where relevant experienced staff are located.

9. The intelligent business linkage queuing calling method based on AI learning model according to claim 1 is characterized by: The step 3 comprises the following sub-steps: Step 3.1: The applicant shares the real-time location of the mobile phone with the queue management module, and the queue management module determines the distance between the applicant and the processing window; Step 3.2: Update the transaction processing progress of each processing window in real time and dynamically adjust the time for the processing person to arrive at the processing window; Step 3.3: Send reminder information to the applicant in advance to remind him / her of the waiting time at the next processing window and the current queue progress; Step 3.4: If the applicant fails to arrive at the window on time, the queue management module will schedule the applicant to the next optimal time slot or processing window; Step 3.5: Synchronously update the processing window and required waiting time for the remaining unprocessed matters of the handler.

10. The intelligent service linkage queuing and calling method based on AI learning model according to claim 1 is characterized by: In step 4, after receiving the request from the handler to cancel the transaction, the queue management module analyzes the processing flow of the transaction required by the handler and warns of the risk of canceling the transaction. After the handler determines that there is no risk impact, the corresponding queue number of the handler's transaction in the queue resource pool is deleted, and the processing window and required waiting time of subsequent transactions are adjusted according to the current queue situation of the queue resource pool.