Staff task dynamic scheduling control method and system based on virtual digital identity

Abstract

The invention discloses an employee task dynamic scheduling control method and system based on a virtual digital identity, and relates to the technical field of digital employee task scheduling, and the method comprises the following steps: collecting the task amount of each time period of different dates and the completion condition of each task; extracting a task density dynamic threshold value, and then carrying out weight distribution processing to obtain time cost weight data; obtaining a time weight and a cost weight of the current date based on the time cost weight data, and scheduling a plurality of tasks at the same time based on an auction algorithm mechanism to obtain all selectable scheduling schemes; the optimal scheduling scheme is selected from all the selectable scheduling schemes for task scheduling; the method is used for solving the problems that when the existing digital employee task scheduling technology schedules the work tasks of digital employees based on an auction algorithm mechanism, auction distribution cannot be performed on a plurality of tasks on the current date at the same time according to task distribution data of historical dates, and the task processing time and cost cannot be balanced at the same time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital employee task scheduling, specifically to an employee task dynamic scheduling control method and system based on virtual digital identity. BACKGROUND

[0002] Digital employee task scheduling technology refers to a technical system that uses intelligent algorithms and system architecture to plan, allocate, monitor and optimize the work tasks of digital employees. The core goal is to ensure that digital employees complete tasks efficiently and orderly in complex business scenarios, maximize resource utilization and reduce operating costs.

[0003] The existing digital employee task scheduling technology often uses a task-by-task auction allocation mechanism when scheduling the work tasks of digital employees based on an auction algorithm mechanism. Each time, only a single task auction process is handled. If the total number of tasks to be allocated is too large, the communication and calculation overhead will increase linearly with the number of tasks, and the efficiency of task scheduling will be low. The task-by-task auction results in scattered allocation of digital employee resources, which cannot take advantage of economies of scale. For example, patent application CN119378940A discloses a multi-robot multi-task scheduling method and application based on a dynamic auction algorithm. This scheme uses a task-by-task auction allocation mechanism when scheduling the work tasks of robots based on an auction algorithm mechanism, which is not efficient in task scheduling. Moreover, the allocation rules of existing auction algorithms often use fixed rules, such as always allocating tasks to the digital employee with the lowest cost or the shortest time, which cannot be adjusted dynamically according to real-time task density and system load. For example, during peak business hours, the fixed rule still prioritizes cost, which can result in a large number of tasks being allocated to low-capacity resources and exceeding the time limit. During off-peak hours, the fixed rule still prioritizes time, which can waste computing resources and increase costs. Therefore, the existing digital employee task scheduling technology cannot simultaneously auction multiple tasks on the current date based on historical task allocation data, balance task processing time and cost, and schedule the work tasks of digital employees based on an auction algorithm mechanism. SUMMARY

[0004] The present application aims to at least solve one of the technical problems in the prior art, by collecting the task amount of each time period of different dates and the completion of each task; and extracting the task density dynamic threshold, then performing weight allocation processing to obtain time cost weight data; and obtaining the time weight and cost weight of the current date, and simultaneously scheduling multiple tasks based on the auction algorithm mechanism to obtain all selectable scheduling schemes; selecting the best scheduling scheme from all selectable scheduling schemes according to the time weight and cost weight of the current date to perform task scheduling; to solve the problem that the existing digital employee task scheduling technology cannot simultaneously auction multiple tasks according to the historical date task allocation data in the current date based on the auction algorithm mechanism for scheduling the work tasks of digital employees, while balancing task processing time and cost.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides a virtual digital identity-based employee task dynamic scheduling control method, comprising the following steps: Collecting the task amount of each time period of different dates and the completion of each task to obtain historical task reference data; Extracting the task density dynamic threshold based on the historical task reference data, and performing weight allocation processing according to the task density dynamic threshold to obtain time cost weight data; Obtaining the time weight and cost weight of the current date based on the time cost weight data, and simultaneously scheduling multiple tasks based on the auction algorithm mechanism to obtain all selectable scheduling schemes; Selecting the best scheduling scheme from all selectable scheduling schemes according to the time weight and cost weight of the current date to perform task scheduling.

[0006] Further, collecting the task amount of each time period of different dates and the completion of each task to obtain historical task reference data comprises the following sub-steps: Divide the date type into weekdays and weekends; for any one historical date, denoted as the first date; obtain the date type of the first date; And divide the first date into multiple uniform time periods with a first time interval within the first date, and denote any one time period as the first time period, wherein the first time interval is t1; Obtain the number of tasks received by the scheduling platform in the first time period, denoted as AF; and obtain the limited duration time length of each task, denoted as the limited completion duration, and obtain the actual time length used by each task to complete, denoted as the actual completion duration, to obtain the initial task-related data of the first time period after completion.

[0007] Further, collecting the task amount of each time period of different dates and the completion of each task to obtain historical task reference data further comprises the following sub-steps: According to the initial task-related data of the first time period, a task density MP corresponding to the first time period is calculated, MP = AF / t1; And the number of tasks with an actual completion duration not greater than the limited completion duration among all tasks received by the scheduling platform in the first time period is obtained, denoted as BF; and a task completion rate RE corresponding to the first time period is calculated, RE = BF / AF; The task density MP and the task completion rate RE corresponding to the first time period are recorded as the first task-related data of the first time period. The first task-related data of all time periods in the first date is repeatedly obtained to obtain the first task reference data of the first date; and the first task reference data of each date is continuously repeatedly obtained and recorded as historical task reference data.

[0008] Further, a task density dynamic threshold is extracted based on the historical task reference data, and a weight allocation process is performed according to the task density dynamic threshold to obtain time cost weight data, including the following sub-steps: According to the date type of the first date, k1 historical dates before the first date and having the same date type as the first date are obtained, denoted as reference dates of the first date, and any one of the reference dates is recorded as a first reference date, wherein k1 is a set number; According to the first task reference data of the first reference date, an average value of the task densities of all time periods of the first reference date is calculated, denoted as an average task density of the first reference date; The average task densities of all reference dates of the first date are repeatedly obtained, and are sorted from far to near according to the corresponding date order to obtain an average task density sequence; and a reference task density CP0 of the first date is calculated according to a first formula, the first formula being as follows: Wherein TPi represents the i-th average task density in the average task density sequence.

[0009] Further, a task density dynamic threshold is extracted based on the historical task reference data, and a weight allocation process is performed according to the task density dynamic threshold to obtain time cost weight data, including the following sub-steps: A time period in which a starting working time of a working day is located is recorded as a starting time period; according to the first task reference data of the first date, the task densities MP and the task completion rates RE of all time periods in the first date are sorted according to the time order starting from the starting time period, and a first task density sequence and a first task completion rate sequence are obtained in sequence; For the starting time period of the first date, a completion rate change trend corresponding to the starting time period of the first date is calculated according to a second formula, the second formula being as follows: , wherein AH represents the completion rate change trend; AS is a set task density sensitivity coefficient, 0 < AS < 1; MPi represents the i-th task density in the first task density sequence; CP(i-1) is the task density bearing value of the first date if i-1 = 0, and CP(i-1) represents the task density bearing value of the i-1-th time period of the first date if i-1 > 0; The task density bearing value corresponding to the starting time period of the first date is calculated according to a third formula, and the third formula is as follows: , wherein AQ is a set density change amplitude coefficient, 0 < AQ < 1; The task density bearing values of all time periods in the first date are repeatedly calculated in sequence, and are arranged in the order of the corresponding time periods, denoted as a first density bearing value sequence; weight coefficients are set for all time periods in the first date, and the sum of the weight coefficients of all time periods in the first date is ensured to be equal to 1; and the task density dynamic threshold YP of the first date is calculated according to a fourth formula, and the fourth formula is as follows: , wherein qi is the weight coefficient of the i-th time period; n is the total number of time periods in the first date; The task density dynamic thresholds of all historical dates are repeatedly obtained according to the historical task reference data.

[0010] Further, based on the historical task reference data, the task density dynamic threshold is extracted, and the weight allocation processing is performed according to the task density dynamic threshold, and the time cost weight data further includes the following sub-steps: For the first time period of the first date, according to the task density dynamic threshold YP of the first date and the task density MP corresponding to the first time period; if MP ≥ YP, the time weight TE of the first time period is set to 1, and the cost weight CE is set to 0; if MP < YP, the time weight TE of the first time period is set to MP / YP, and the cost weight CE is set to 1-MP / YP; The time weight and the cost weight of all time periods in the first date are repeatedly obtained, and the time cost weight information of the first date is obtained; and the time cost weight information of all historical dates is repeatedly obtained, and the time cost weight data is obtained.

[0011] Further, based on the time cost weight data, the time weight and the cost weight of the current date are obtained, and a plurality of tasks are simultaneously scheduled based on an auction algorithm mechanism, and all selectable scheduling schemes include the following sub-steps: For the current date to be controlled, denoted as a first current date; any one time period in the first current date is denoted as a first current time period; obtaining a date type according to the first current date, obtaining k2 historical dates before the first current date and same as the date type of the first current date, denoted as weight reference dates of the first current date, wherein k2 is the number of settings; obtaining the time weight and the cost weight of the time period same as the first current time period in all weight reference dates according to the time cost weight information of the weight reference dates, and respectively sorting the corresponding date order, and respectively denoted as time weight reference sequence and cost weight reference sequence; setting a weight coefficient for each weight reference date according to the time distance of each weight reference date from the first current date, and ensuring that the sum of the weight coefficients of all weight reference dates is equal to 1; Further, according to the weight coefficient of each weight reference date, the weighted average value of the time weight reference sequence and the cost weight reference sequence is calculated respectively, and the average time weight and the average cost weight are obtained in sequence, and are respectively marked as the current time weight and the current cost weight of the first current time period; Repeat to obtain the current time weight and the current cost weight of all time periods in the first current date.

[0012] Further, based on the time cost weight data, the time weight and the cost weight of the current date are obtained, and based on the auction algorithm mechanism, multiple tasks are simultaneously scheduled to obtain all selectable scheduling schemes, which further includes the following sub-steps: For the first current time period, the number of currently idle virtual digital employees is obtained, denoted as LA; and any idle virtual digital employee is denoted as the first idle employee; and the number of currently scheduled tasks is obtained, denoted as LB; Based on the auction algorithm mechanism, and according to LA and LB, LC scheduled tasks are auctioned; wherein, if LA> LB, then LC= LB, otherwise LC= LA; the estimated processing time and the estimated processing cost given by the first idle employee for each auctioned task are obtained, denoted as processing time and processing cost respectively, to obtain the task processing data of the first idle employee; Repeat to obtain the task processing data of all idle virtual digital employees, denoted as task processing cost data; LC scheduled tasks are allocated to LA idle virtual digital employees, and all allocation schemes are obtained, denoted as selectable scheduling schemes; For any one selectable scheduling scheme, denoted as the first scheduling scheme, the sum of the estimated processing costs of all tasks in the first scheduling scheme is calculated, denoted as the processing cost of the first scheduling scheme, and the estimated processing time of all tasks in the first scheduling scheme is sorted in ascending order, and the k3 percentile is obtained, denoted as the processing time of the first scheduling scheme; wherein k3 is the set percentile; Repeat to obtain the processing time and the processing cost of each selectable scheduling scheme.

[0013] Further, the selecting the optimal scheduling scheme from all the optional scheduling schemes according to the time weight and the cost weight of the current date for task scheduling comprises the following sub-steps: sorting the processing time length and the processing cost of all the optional scheduling schemes in descending order respectively to obtain a processing time length sequence and a processing cost sequence; for the first scheduling scheme, obtaining the positions of the processing time length and the processing cost of the first scheduling scheme in the corresponding processing time length sequence and processing cost sequence, denoted as AG1 and BG1 in order respectively; and calculating the processing time length weight QG1 and the processing cost weight QG2 of the first scheduling scheme, wherein QG1= (AG1 / W0) / [(AG1 / W0) + (BG1 / W0)], QG2= (BG1 / W0) / [(AG1 / W0) + (BG1 / W0)], repeatedly obtaining the processing time length weight and the processing cost weight of all the first scheduling schemes, and sorting them in order respectively, denoted as a time length weight sequence and a cost weight sequence; if the current time weight TE=1 and the current cost weight CE=0 of the first current time period, the optional scheduling scheme with the largest processing time length weight is marked as the optimal scheduling scheme; otherwise, obtaining VE and NE, VE=MAX (TE, CE), NE=MIN (TE, CE), and according to the weight type of VE, the time length weight sequence or the cost weight sequence is denoted as the first sequence, and the other is denoted as the second sequence; obtaining the optional scheduling schemes in [k4*VE, k5*VE] in the first sequence, denoted as the optional scheduling schemes; for any one of the optional scheduling schemes, denoted as the first optional scheme, obtaining the processing cost weight VG of the first optional scheme and the processing time length weight corresponding to VE, and the other is denoted as NG; and calculating the weight difference degree XG of the first optional scheme, XG= (|VE-VG|\VE) + (|NE-NG|\NE); repeatedly obtaining the weight difference degree of all the optional scheduling schemes, and the optional scheduling scheme with the smallest weight difference degree is denoted as the optimal scheduling scheme; performing task allocation scheduling according to the optimal scheduling scheme.

[0014] In a second aspect, the present application provides an employee task dynamic scheduling control system based on virtual digital identity, comprising a data acquisition module, a threshold weight module, a scheduling preset module and a task scheduling module; The data acquisition module is used to acquire the task quantity of each time period of different dates and the completion of each task to obtain historical task reference data; The threshold weight module comprises a threshold extraction unit and a weight distribution unit, the threshold extraction unit extracts a task density dynamic threshold based on historical task reference data, and the weight distribution unit performs weight distribution processing according to the task density dynamic threshold to obtain time cost weight data; The scheduling preset module obtains time weight and cost weight of the current date based on the time cost weight data, and simultaneously schedules multiple tasks based on an auction algorithm mechanism to obtain all selectable scheduling schemes; The task scheduling module selects a best scheduling scheme from all selectable scheduling schemes according to the time weight and cost weight of the current date to perform task scheduling.

[0015] The present application has the following advantages: the present application obtains historical task reference data by collecting the task amount of each time period and the completion condition of each task of different dates, extracts a task density dynamic threshold based on the historical task reference data, performs weight distribution processing according to the task density dynamic threshold to obtain time cost weight data, obtains time weight and cost weight of the current date based on the time cost weight data, simultaneously schedules multiple tasks based on an auction algorithm mechanism to obtain all selectable scheduling schemes, selects a best scheduling scheme from all selectable scheduling schemes according to the time weight and cost weight of the current date to perform task scheduling, and balances the task processing time and cost when scheduling the work tasks of digital employees based on the auction algorithm mechanism, and simultaneously auctions and allocates multiple tasks according to the task allocation data of historical dates, while balancing the task processing time and cost. The present application calculates the carrying capacity of each time period by the completion rate trend and density carrying value based on the sequence change of task density and completion rate, adapts to real-time load change, makes the obtained threshold more in line with the actual situation, generates the final dynamic threshold by weighted average, highlights the influence of the work period on the overall threshold, avoids the interference of the data of the rest period on the threshold, distributes the time weight and cost weight of each period according to the task density dynamic threshold, makes the weight distribution not only fit the business periodic fluctuation, but also respond to the task quantity change in real time, ensures that the task processing time and cost can be balanced in real time when scheduling tasks, simultaneously auctions and allocates multiple tasks, and selects the best scheduling scheme, which can improve the efficiency of task scheduling and ensure that the task scheduling strategy conforms to the current actual situation. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 It is the principle block diagram of the system of the present application; Figure 2 It is the step flow chart of the method of the present application; Figure 3 It is the task scheduling strategy flow chart of the present application; Figure 4 It is the structure schematic diagram of the electronic equipment of the present application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be apparently and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0018] Embodiment 1, please refer to Figure 1 As shown in the figure, the present application provides an employee task dynamic scheduling control system based on virtual digital identity, including a data acquisition module, a threshold weight module, a scheduling preset module and a task scheduling module; The data acquisition module is configured to acquire the task quantity of each time period of different dates and the completion condition of each task, and obtain historical task reference data; The data acquisition module is configured with a data acquisition strategy. The data acquisition strategy includes: dividing the date type into weekdays and rest days; for example, Monday to Friday are weekdays, and holidays and weekends are rest days; for any historical date, denoted as a first date; obtaining the date type of the first date; And the first date is divided into multiple uniform time periods at a first time interval within the first date, and any one time period is denoted as a first time period, wherein the first time interval is t1; the single-day task quantity fluctuation is decomposed into changes of smaller time units, avoiding that the coarse-grained data in units of "day" masks the intra-day fluctuation rule; the first time interval can be set according to the actual application scenario, and in this embodiment, t1=1 hour; Obtain the number of tasks received by the scheduling platform in the first time period, denoted as AF; and obtain the time length of the limited duration of each task, denoted as the limited completion duration, and obtain the actual time length of each task, denoted as the actual completion duration, and obtain the initial task related data of the first time period after completion; the limited completion duration is the deadline constraint required by the task; the actual completion duration is the actual execution time consumed by the task; the deviation from the limited completion duration directly reflects the scheduling efficiency, providing data support for subsequent evaluation of period carrying capacity; According to the initial task related data of the first time period, calculate the task density MP corresponding to the first time period, MP=AF / t1; quantify the task inflow in the unit time period, which is used to measure the period load pressure, and facilitate the load comparison of different time periods and different dates; BF is obtained, and the task completion rate RE corresponding to the first time period is calculated, RE=BF / AF. The proportion of tasks delivered on time in the quantization period directly reflects the matching degree of system processing capacity and task pressure. For example, when a time period MP=40, RE=70%, it can be determined that the MP has exceeded the system carrying capacity; The task density MP and the task completion rate RE corresponding to the first time period are recorded as the first task related data of the first time period. The first task related data of all time periods in the first date is repeatedly obtained to obtain the first task reference data of the first date, and the first task reference data of each date is continuously repeatedly obtained to obtain the historical task reference data. In the specific implementation process, the date type is divided into weekdays and rest days, in order to lay a reference basis for subsequent threshold calculation, and to ensure that the subsequent threshold calculation conforms to the business cycle law. For example, the threshold calculation of weekdays only depends on the historical weekday data, avoiding the misjudgment of the threshold of the peak scene of weekdays caused by the low load data of rest days. Taking the average task density of rest days as the threshold of weekdays will lead to insufficient scheduling resources of weekdays.

[0019] The threshold weight module includes a threshold extraction unit and a weight allocation unit. The threshold extraction unit extracts a task density dynamic threshold based on the historical task reference data, and the weight allocation unit performs weight allocation processing according to the task density dynamic threshold to obtain time cost weight data. The threshold extraction unit is configured with a threshold extraction strategy, which includes: according to the date type of the first date, obtaining k1 historical dates before the first date and having the same date type as the first date, recording as the reference dates of the first date, and recording any one reference date as the first reference date, wherein k1 is the number of settings. In this embodiment, k1=7 days; using historical similar date data, such as reference weekdays for weekdays and reference rest days for rest days, to avoid confusion of business characteristics of different date types, such as interference of rest day low load data to weekday threshold calculation, to ensure that the subsequent threshold calculation conforms to the business periodicity law. According to the first task reference data of the first reference date, the average value of the task density of all time periods of the first reference date is calculated, recorded as the average task density of the first reference date. The average task density of all reference dates of the first date is repeatedly obtained, and is sorted from far to near according to the corresponding date order, recorded as the average task density sequence. The reference task density CP0 of the first date is calculated according to the first formula, and the first formula is as follows: where TPi represents the ith average task density in the average task density sequence; the dynamic weighting of k1 recent dates is used, and the closer the date, the higher the weight, so that CP0 can reflect the recent business trend and avoid threshold deviation caused by outdated historical data; CP0 is the initial task density carrying value for subsequent; The time period in which the starting working time of the working day is located is recorded as the starting time period; according to the first task reference data of the first date, the task density MP and the task completion rate RE of all time periods in the first date are sorted in time sequence from the starting time period, and the first task density sequence and the first task completion rate sequence are obtained in sequence respectively; For the starting time period of the first date, the completion rate change trend corresponding to the starting time period of the first date is calculated according to the second formula, and the second formula is as follows: where AH represents the completion rate change trend; AS is a set task density sensitivity coefficient, 0<AS<1; MPi represents the ith task density in the first task density sequence; CP(i-1), if i-1=0, then CP(i-1) is the reference task density CP0, and if i-1>0, then CP(i-1) represents the task density carrying value of the i-1th time period of the first date; The second formula is used to quantify the change trend of the completion rate with the task density, when a certain time period MPi exceeds the carrying value CP(i-1) of the previous time period, AH is negative, which drives CPi to decrease, and realizes the linkage adjustment of "time period-density-completion rate", for example, automatically reducing the threshold value of the subsequent time period after the peak time period overload; AS represents the sensitivity of the task completion rate to the change of the task density, the greater AS, the greater the influence of the task completion rate on the task density, which can be set according to the actual application scene, and in this embodiment, AS=0.6 Then, the task density carrying value corresponding to the starting time period of the first date is calculated according to the third formula, and the third formula is as follows: where AQ is a set density change amplitude coefficient, 0<AQ<1; CPi not only considers the current MPi, but also associates the carrying state of the previous time period, forming a dynamic calibration in time sequence, avoiding the problem that the static threshold value cannot adapt to real-time load fluctuation; AQ is used to adjust the dynamic change amplitude of the task density carrying value, the greater AQ, the more intense the change of the threshold task density carrying value, which can be set according to the actual application scene, and in this embodiment, AQ=0.4; The task density carrying values of all time periods in the first date are repeatedly calculated in sequence, and are arranged in the order of the corresponding time periods, which are recorded as the first density carrying value sequence; the weight coefficients of all time periods in the first date are set, and the sum of the weight coefficients of all time periods in the first date is ensured to be equal to 1; and the task density dynamic threshold YP of the first date is calculated according to the fourth formula, and the fourth formula is as follows: wherein qi is the weight coefficient of the ith time period; n is the total number of time periods within the first date; The weighted average mechanism enables YP to reflect the overall level of the full-time period load and focus on the service peak pressure, for example, the working hours are 10 hours, the weight of each hour period is set to 0.6 / 10, the working hours are 14 hours, and the weight of each hour period is set to 0.4 / 14; and ensure that YP is more in line with the actual demand of the high-load scene; According to the historical task reference data, the task density dynamic threshold of all historical dates is repeatedly obtained; The weight allocation unit is configured with a weight allocation strategy, and the weight allocation strategy comprises: for the first time period of the first date, according to the task density dynamic threshold YP of the first date and the task density MP corresponding to the first time period; if MP≥YP, set the time weight TE of the first time period =1, and the cost weight CE =0; if MP<YP, set the time weight TE of the first time period =MP / YP, and the cost weight CE =1-MP / YP; that is, high load is preferentially guaranteed for task timeliness, and sudden drop of completion rate is avoided; low load is proportionally allocated weight, and cost optimization is considered; The time weight and the cost weight of all time periods within the first date are repeatedly obtained, and the time cost weight information of the first date is obtained; and the time cost weight information of all historical dates is repeatedly obtained, and the time cost weight data is obtained; In the specific implementation process, YP highlights the dominant role of the key period on the overall threshold through weight aggregation, avoids the low-density data of the low-peak period to lower the threshold, and causes insufficient scheduling resources in the high-peak period; and YP as an "efficiency-cost" balance point enables the weight allocation to automatically adjust with real-time load, for example, MP of a certain period increases from 30 to 40, YP=35, TE increases from 0.86 to 1, triggering the time priority strategy to prevent task backlog.

[0020] The scheduling preset module obtains the time weight and the cost weight of the current date based on the time cost weight data, and simultaneously schedules multiple tasks based on the auction algorithm mechanism to obtain all selectable scheduling schemes; The scheduling preset module is configured with a scheduling preset strategy, and the scheduling preset strategy comprises: for the current date to be controlled, denoted as the first current date; any one time period in the first current date is denoted as the first current time period; According to the date type of the first current date, the k2 historical dates before the first current date and the same as the date type of the first current date are obtained, denoted as the weight reference date of the first current date, wherein k2 is the set number; in this embodiment, k2=7; According to the time cost weight information of the weight reference date, the time weight and the cost weight of the same time period as the first current period in all weight reference dates are obtained, and are sorted in corresponding date order respectively, and are marked as time weight reference sequence and cost weight reference sequence respectively; The precise data matching of "same time period, same type date" is realized, for example, the weight distribution of the first current date 9 o'clock is analyzed, and the weight distribution of the weight reference date 9 is obtained; Avoid the interference of cross-period data on the current period weight; According to the time distance of each weight reference date from the first current date, the weight coefficient of each weight reference date is set, and the sum of the weight coefficients of all weight reference dates is equal to 1; The closer the time distance, the higher the weight, giving higher weight to recent data to make the current period weight closer to the latest business status; According to the weight coefficient of each weight reference date, the weighted average value of the time weight reference sequence and the cost weight reference sequence is calculated respectively, and the average time weight and the average cost weight are obtained in sequence, and are marked as the current time weight and the current cost weight of the first current period respectively; Through the weighted aggregation of historical weights of the same type and the same time period, the current weight conforming to the business rule is generated, which provides accurate reference for subsequent task scheduling; Repeat the process of obtaining the current time weight and the current cost weight of all time periods in the first current date; For the first current period, the number of currently idle virtual digital employees is obtained, marked as LA; And any idle virtual digital employee is marked as the first idle employee; And the number of tasks to be scheduled is obtained, marked as LB; Based on the auction algorithm mechanism, and according to LA and LB, LC tasks to be scheduled are auctioned; If LA> LB, then LC= LB, otherwise LC= LA; Avoid the number of allocated tasks exceeding the number of idle employees; The estimated processing time and the estimated processing cost of the first idle employee for each auctioned task are obtained, marked as processing time and processing cost respectively, and the task processing data of the first idle employee is obtained; Repeat the process of obtaining the task processing data of all idle virtual digital employees, marked as task processing cost data; For example, there are three tasks, and the first idle employee gives the time and cost required for processing task 1, task 2 and task 3, marked as the task processing data of the first idle employee; LC tasks to be scheduled are allocated to LA idle virtual digital employees, and all allocation schemes are obtained, marked as optional scheduling scheme; For any optional scheduling scheme, denoted as the first scheduling scheme, calculate the sum of the estimated processing costs of all tasks in the first scheduling scheme, denoted as the processing cost of the first scheduling scheme, sort the estimated processing times of all tasks in the first scheduling scheme from smallest to largest, and obtain the k3th percentile, denoted as the processing time of the first scheduling scheme; where k3 is the set percentile; repeatedly obtain the processing time and processing cost of each optional scheduling scheme; in this embodiment, k3 = 90; During the specific implementation process, the processing cost of the plan, that is, the sum of the processing costs of all tasks, is used to measure the overall economic efficiency; and the k3th percentile is selected as the processing time of the first scheduling plan to prevent some small number of tasks from timeout. It is more representative of the time cost of the plan than the average processing time.

[0021] The task scheduling module selects the best scheduling scheme from all available scheduling schemes according to the time weight and cost weight of the current date to schedule the task; The task scheduling module is equipped with a task scheduling strategy. The task scheduling strategy includes: sorting the processing time and processing cost of all optional scheduling schemes in descending order to obtain the processing time sequence and processing cost sequence; clarifying the relative performance of each scheme in terms of time and cost through sorting See also Figure 3 As shown, for the first scheduling scheme, the processing time and processing cost of the first scheduling scheme are obtained in the corresponding processing time sequence and processing cost sequence, and are recorded as AG1 and BG1 respectively in order; and the processing time weight QG1 and processing cost weight QG2 of the first scheduling scheme are calculated, where QG1=(AG1 / W0) / [(AG1 / W0)+(BG1 / W0)], QG2=(BG1 / W0) / [(AG1 / W0)+(BG1 / W0)], and the processing time weights and processing cost weights of all first scheduling schemes are repeatedly obtained and sorted according to size, and recorded as the time weight sequence and cost weight sequence respectively; the relative position of the scheme is converted into a quantitative weight to reflect its comprehensive performance in the time and cost dimensions. For example, if a scheme has a high time position and a low cost position, then QG1 is low and QG2 is high, indicating that it has a significant cost advantage but takes a long time; If the current time weight TE of the first current period is 1 and the current cost weight CE is 0, the optional scheduling scheme with the largest processing time weight is marked as the optimal scheduling scheme. In high-load scenarios, the mandatory time priority strategy is used to ensure task timeliness and avoid task timeouts due to cost considerations. Otherwise, obtain VE and NE, VE = MAX (TE, CE), NE = MIN (TE, CE), determine the dominant weight of the current time period; and according to the weight type of VE, record the time length weight sequence or the cost weight sequence as the first sequence, and record the other as the second sequence; for example, VE is the current time weight, record the time length weight sequence as the first sequence, and record the cost weight sequence as the second sequence; Obtain the optional scheduling scheme in [k4*VE, k5*VE] in the first sequence sequence, record it as a candidate scheduling scheme; for any one candidate scheduling scheme, record it as a first candidate scheme, obtain the processing time length weight VG corresponding to VE and the processing cost weight NG, and record the other as NG; and calculate the weight difference degree XG of the first candidate scheme, XG = (|VE-VG|VE) + (|NE-NG|NE); Repeat to obtain the weight difference degree of all candidate scheduling schemes, and record the candidate scheduling scheme with the minimum weight difference degree as the best scheduling scheme; in this embodiment, k4 = 0.8, k5 = 1.2; [k4*VE, k5*VE] ensures that the candidate scheme meets the basic requirements in the dominant dimension, for example, only consider the upstream scheme in the time length weight when time priority is considered, and improve the screening efficiency; According to the best scheduling scheme, perform task allocation scheduling; In the specific implementation process, XG quantifies the matching degree of the scheme and the current weight through standardized absolute deviation, for example, a scheme VE = 0.7 corresponds to VG = 0.65, NE = 0.3 corresponds to NG = 0.35, XG = (0.05 / 0.7) + (0.05 / 0.3) ≈ 0.24, the smaller the difference degree, the more suitable for the scheduling target, and the excessive deviation from the secondary dimension is avoided when only the dominant dimension is pursued; for example, when time priority is considered, the cost rationality is still considered.

[0022] Embodiment 2, please refer to Figure 2 As shown in the figure, the present application provides a virtual digital identity-based employee task dynamic scheduling control method, which comprises the following steps: Step S1, collect the task quantity of each time period and the completion of each task on different dates to obtain historical task reference data; step S1 comprises the following substeps: Step S101, divide the date type into weekdays and weekends; for any one historical date, record it as a first date; obtain the date type of the first date; Step S102, divide the first date into multiple uniform time periods at a first time interval within the first date, and record any one time period as a first time period, wherein the first time interval is t1; Step S103, obtaining the number of tasks received by the scheduling platform in the first time period, denoted as AF; and obtaining the time length of the limited duration of each task, denoted as the limited completion duration, and obtaining the actual time length of each task, denoted as the actual completion duration, to obtain the initial task-related data of the first time period after completion; Step S104, calculating the task density MP corresponding to the first time period according to the initial task-related data of the first time period, MP = AF / t1; Step S105, obtaining the number of tasks in which the actual completion duration is not greater than the limited completion duration among all tasks received by the scheduling platform in the first time period, denoted as BF; and calculating the task completion rate RE corresponding to the first time period, RE = BF / AF; Step S106, recording the task density MP and the task completion rate RE corresponding to the first time period as the first task-related data of the first time period; Step S107, repeatedly obtaining the first task-related data of all time periods in the first date to obtain the first task reference data of the first date; and continuously repeatedly obtaining the first task reference data of each date, denoted as the historical task reference data.

[0023] Step S2, extracting a task density dynamic threshold based on the historical task reference data, and performing weight allocation processing according to the task density dynamic threshold to obtain time cost weight data; Step S2 includes the following sub-steps: Step S201, according to the date type of the first date, obtaining k1 historical dates before the first date which have the same date type as the first date and are the closest to the first date, denoted as the reference dates of the first date, and recording any one reference date as the first reference date, wherein k1 is the number of settings; Step S202, calculating the average value of the task density of all time periods of the first reference date according to the first task reference data of the first reference date, denoted as the average task density of the first reference date; Step S203, repeatedly obtaining the average task density of all reference dates of the first date, and sorting from far to near according to the corresponding date order, denoted as the average task density sequence; and calculating the reference task density CP0 of the first date according to the first formula, the first formula being as follows: Wherein TPi represents the i-th average task density in the average task density sequence; Step S204, recording the time period in which the starting working time of the working day is located as the starting time period; according to the first task reference data of the first date, sorting the task density MP and the task completion rate RE of all time periods in the first date from the starting time period according to the time order, respectively, to obtain the first task density sequence and the first task completion rate sequence in order; Step S205, for the starting time period of the first date, a completion rate change trend corresponding to the starting time period of the first date is calculated according to a second formula, the second formula being as follows: wherein AH represents the completion rate change trend; AS is a set task density sensitivity coefficient, 0 < AS < 1; MPi represents the ith task density in the first task density sequence; CP(i-1) is the task density carrying value of the i-1th time period of the first date, if i-1 = 0, CP(i-1) is the reference task density CP0, if i-1 > 0, CP(i-1) represents the task density carrying value of the i-1th time period of the first date; Step S206, the task density carrying value corresponding to the starting time period of the first date is calculated according to a third formula, the third formula being as follows: wherein AQ is a set density change amplitude coefficient, 0 < AQ < 1; Step S207, the task density carrying values of all time periods in the first date are repeatedly calculated in sequence, and are arranged in the order of the corresponding time periods, and are recorded as a first density carrying value sequence; weight coefficients are set for all time periods in the first date, and the sum of the weight coefficients of all time periods in the first date is ensured to be equal to 1; and a task density dynamic threshold YP of the first date is calculated according to a fourth formula, the fourth formula being as follows: wherein qi is the weight coefficient of the ith time period; n is the total number of time periods in the first date; Step S208, the task density dynamic thresholds of all historical dates are repeatedly obtained according to historical task reference data; Step S209, for the first time period of the first date, the task density dynamic threshold YP of the first date and the task density MP corresponding to the first time period are used; if MP ≥ YP, the time weight TE of the first time period is set to 1, and the cost weight CE is set to 0; if MP < YP, the time weight TE of the first time period is set to MP / YP, and the cost weight CE is set to 1-MP / YP; Step S210, the time weight and the cost weight of all time periods in the first date are repeatedly obtained, to obtain the time cost weight information of the first date; and the time cost weight information of all historical dates is repeatedly obtained, to obtain the time cost weight data.

[0024] Step S3, the time weight and the cost weight of the current date are obtained based on the time cost weight data, and a plurality of tasks are simultaneously scheduled based on an auction algorithm mechanism, to obtain all selectable scheduling schemes; step S3 includes the following substeps: Step S301, for the current date to be controlled, recorded as a first current date; any one time period in the first current date is recorded as a first current time period; Step S302, acquiring the date type according to the first current date, acquiring k2 historical dates before the first current date and the same as the date type of the first current date, denoted as the weight reference date of the first current date, wherein k2 is the number set; Step S303, according to the time cost weight information of the weight reference date, acquiring the time weight and cost weight of the time period same as the first current time period in all weight reference dates, and respectively sorting according to the corresponding date order, and respectively denoted as the time weight reference sequence and the cost weight reference sequence; Step S304, according to the time distance of each weight reference date from the first current date, setting the weight coefficient for each weight reference date, and ensuring that the sum of the weight coefficients of all weight reference dates is equal to 1; Step S305, further according to the weight coefficient of each weight reference date, respectively calculating the weighted average value of the time weight reference sequence and the cost weight reference sequence, and respectively obtaining the average time weight and the average cost weight, and respectively denoted as the current time weight and the current cost weight of the first current time period; Step S306, repeating the acquisition of the current time weight and the current cost weight of all time periods in the first current date; Step S307, for the first current time period, acquiring the number of currently idle virtual digital employees, denoted as LA; and any idle virtual digital employee is denoted as the first idle employee; and acquiring the number of currently to-be-scheduled tasks, denoted as LB; Step S308, based on the auction algorithm mechanism, and according to LA and LB, simultaneously auctioning LC to-be-scheduled tasks; wherein, if LA>LB, then LC=LB, otherwise LC=LA; acquiring the estimated processing time and the estimated processing cost given by the first idle employee for each auctioned task, respectively denoted as the processing time and the processing cost, and obtaining the task processing data of the first idle employee; repeating the acquisition of the task processing data of all idle virtual digital employees, denoted as the task processing cost data; Step S309, distributing the LC to-be-scheduled tasks to the LA idle virtual digital employees, acquiring all distribution schemes, denoted as the optional scheduling scheme; Step S310, for any optional scheduling scheme, denoted as the first scheduling scheme, calculating the sum of the estimated processing costs of all tasks in the first scheduling scheme, denoted as the processing cost of the first scheduling scheme, and sorting the estimated processing time of all tasks in the first scheduling scheme according to the order from small to large, and acquiring the k3 percentile, denoted as the processing time of the first scheduling scheme; wherein k3 is the set percentile; repeating the acquisition of the processing time and the processing cost of each optional scheduling scheme.

[0025] Step S4, selecting the best scheduling scheme from all optional scheduling schemes according to the time weight and the cost weight of the current date to schedule the task; Step S4 includes the following sub-steps: Step S401, sorting the processing time length and the processing cost of all optional scheduling schemes in descending order respectively to obtain a processing time length sequence and a processing cost sequence; Step S402, for the first scheduling scheme, obtaining the position of the processing time length and the processing cost of the first scheduling scheme in the corresponding processing time length sequence and the processing cost sequence, denoted as AG1 and BG1 in order respectively; and calculating the processing time length weight QG1 and the processing cost weight QG2 of the first scheduling scheme, wherein QG1= (AG1 / W0) / [(AG1 / W0) + (BG1 / W0)], QG2= (BG1 / W0) / [(AG1 / W0) + (BG1 / W0)], repeatedly obtaining the processing time length weight and the processing cost weight of all first scheduling schemes, and sorting them in size respectively, denoted as a time length weight sequence and a cost weight sequence respectively; Step S403, if the current time weight TE=1 and the current cost weight CE=0 of the first current time period, the optional scheduling scheme with the largest processing time length weight is marked as the best scheduling scheme; Step S404, otherwise, obtaining VE and NE, VE=MAX (TE, CE), NE=MIN (TE, CE), and according to the weight type of VE, the time length weight sequence or the cost weight sequence is denoted as the first sequence, and the other is denoted as the second sequence; Step S405, obtaining the optional scheduling scheme located in [k4*VE, k5*VE] in the first sequence, denoted as a candidate scheduling scheme; for any one candidate scheduling scheme, denoted as a first candidate scheme, obtaining the processing cost weight of the first candidate scheme and the processing time length weight corresponding to VE, denoted as VG, and the other is denoted as NG; and calculating the weight difference degree XG of the first candidate scheme, XG= (|VE-VG|\VE) + (|NE-NG|\NE); repeatedly obtaining the weight difference degree of all candidate scheduling schemes, and the candidate scheduling scheme with the smallest weight difference degree is denoted as the best scheduling scheme; Step S406, performing task allocation scheduling according to the best scheduling scheme.

[0026] Embodiment 3, please refer to Figure 4 as shown, Figure 4An example is shown in a structural diagram of an electronic device, which can include a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete the communication among each other through the communication bus. The memory stores computer readable instructions, and the processor can call the instructions in the memory, and when the computer readable instructions are executed by the processor, the steps in the employee task dynamic scheduling control method based on virtual digital identity are run to realize the following functions: collecting the task amount of each time period and the completion of each task on different dates to obtain historical task reference data; extracting a task density dynamic threshold based on the historical task reference data, and performing weight allocation processing according to the task density dynamic threshold to obtain time cost weight data; obtaining time weight and cost weight of the current date based on the time cost weight data, and simultaneously scheduling multiple tasks based on an auction algorithm mechanism to obtain all optional scheduling schemes; and selecting the best scheduling scheme from all optional scheduling schemes according to the time weight and cost weight of the current date to perform task scheduling.

[0027] In addition, the logical instructions in the memory described above can be implemented in the form of a software functional unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0028] In embodiment 4, the present application further provides a computer readable storage medium, and the present application provides a storage medium having a computer program stored thereon, and the computer program is executed by a processor to run the steps in the employee task dynamic scheduling control method based on virtual digital identity as described above to realize the following functions: collecting the task amount of each time period and the completion of each task on different dates to obtain historical task reference data; extracting a task density dynamic threshold based on the historical task reference data, and performing weight allocation processing according to the task density dynamic threshold to obtain time cost weight data; obtaining time weight and cost weight of the current date based on the time cost weight data, and simultaneously scheduling multiple tasks based on an auction algorithm mechanism to obtain all optional scheduling schemes; and selecting the best scheduling scheme from all optional scheduling schemes according to the time weight and cost weight of the current date to perform task scheduling.

[0029] Through the description of the above embodiments, the embodiments of the present application can be provided as a method, a system or a computer program product. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.

[0030] In the embodiments provided by the present application, it should be understood that the disclosed system or method can be implemented in other manners. The above described embodiments are merely exemplary, for example, the division of modules or units can be different from the above, the implementation can be combined or integrated in other manners, and some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the modules or units can be electrical, mechanical or other forms.

[0031] Finally, it should be noted that the above embodiments are merely used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still make modifications to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for dynamic scheduling and control of employee tasks based on virtual digital identities, characterized in that: The steps include: Collect the task volume and completion status of each task in different time periods on different dates to obtain historical task reference data; Extract the dynamic threshold of task density based on historical task reference data, and perform weight distribution processing according to the dynamic threshold of task density to obtain time cost weight data; Based on the time-cost weight data, the time weight and cost weight of the current date are obtained, and multiple tasks are scheduled simultaneously based on the auction algorithm mechanism to obtain all optional scheduling solutions; The best scheduling scheme is selected from all optional scheduling schemes according to the time weight and cost weight of the current date for task scheduling.

2. The employee task dynamic scheduling control method based on virtual digital identity according to claim 1 is characterized in that: Collecting the task volume and completion status of each task in different time periods on different dates to obtain historical task reference data includes the following sub-steps: Divide the date type into working days and rest days; for any historical date, record it as the first date; obtain the date type of the first date; and dividing the first date into a plurality of even time periods at a first time interval within the first date, and recording any time period as a first time period, wherein the first time interval is t1; Obtain the number of tasks received by the scheduling platform in the first time period, recorded as AF; obtain the time limit of each task, recorded as the limited completion time, and obtain the actual time length used to complete each task, recorded as the actual completion time. After completion, obtain the initial task-related data of the first time period.

3. The employee task dynamic scheduling control method based on virtual digital identity according to claim 2 is characterized in that: Collecting the task volume and completion status of each task in different time periods on different dates to obtain historical task reference data also includes the following sub-steps: According to the initial task-related data of the first time period, calculate the task density MP corresponding to the first time period, MP=AF / t1; And obtain the number of tasks whose actual completion time is not greater than the specified completion time among all tasks received by the scheduling platform in the first time period, recorded as BF; and calculate the task completion rate RE corresponding to the first time period, RE = BF / AF; The task density MP and the task completion rate RE corresponding to the first time period are recorded as the first task related data of the first time period; Repeatedly obtain the first task related data of all time periods within the first date to obtain the first task reference data of the first date; And continuously and repeatedly obtain the first task reference data of each date and record it as historical task reference data.

4. The employee task dynamic scheduling control method based on virtual digital identity according to claim 3 is characterized in that: Extracting the dynamic threshold of task density based on historical task reference data and performing weight distribution processing according to the dynamic threshold of task density to obtain time cost weight data includes the following sub-steps: According to the date type of the first date, obtain the k1 most recent historical dates with the same date type as the first date before the first date and record them as the base reference date of the first date, and record any base reference date as the first base reference date, where k1 is the set number; Calculate the average value of the task density of all time periods of the first base reference date based on the first task reference data of the first base reference date, and record it as the average task density of the first base reference date; Repeatedly obtain the average task density of all benchmark reference dates for the first date, and sort them from far to near according to the corresponding date sequence, recording them as the average task density sequence; and calculate the benchmark task density CP0 of the first date according to the first formula, which is as follows: , where TPi represents the i-th average task density in the average task density sequence.

5. The employee task dynamic scheduling control method based on virtual digital identity according to claim 4 is characterized in that: Extracting the dynamic threshold of task density based on historical task reference data and performing weight distribution processing according to the dynamic threshold of task density to obtain time cost weight data also includes the following sub-steps: The time period in which the starting working time of a working day is located is denoted as the starting time period; according to the first task reference data of the first date, the task density MP and the task completion rate RE of all time periods within the first date are sorted in chronological order starting from the starting time period, and the first task density sequence and the first task completion rate sequence are obtained respectively in order; For the starting time period of the first date, calculate the completion rate change trend corresponding to the starting time period of the first date according to the second formula, and the second formula is as follows: , where AH represents the completion rate change trend; AS is the set task density sensitivity coefficient, 0 < AS < 1; MPi represents the i-th task density in the first task density sequence; CP(i - 1), if i - 1 = 0, then CP(i - 1) is the reference task density CP0, if i - 1 > 0, then CP(i - 1) represents the task density carrying value of the (i - 1)-th time period of the first date; Then calculate the task density carrying value corresponding to the starting time period of the first date according to the third formula. The third formula is as follows: , where AQ is the set density variation coefficient, 0 <AQ<1; Repeat the calculation of the task density carrying values ​​for all time periods within the first date, and arrange them in the order of the corresponding time periods, recording them as the first density carrying value sequence; set weight coefficients for all time periods within the first date, and ensure that the sum of the weight coefficients of all time periods within the first date is equal to 1; and calculate the task density dynamic threshold YP for the first date according to the fourth formula, which is as follows: , where qi is the weight coefficient of the i-th time period; n is the total number of time periods in the first day; Repeat to obtain the task density dynamic thresholds of all historical dates according to the historical task reference data.

6. The employee task dynamic scheduling control method based on virtual digital identity according to claim 5 is characterized in that: Extracting the task density dynamic threshold based on the historical task reference data and performing weight allocation processing according to the task density dynamic threshold to obtain the time cost weight data further includes the following sub-steps: For the first time period of the first date, according to the task density dynamic threshold YP of the first date and the task density MP corresponding to the first time period; if MP ≥ YP, then set the time weight TE of the first time period to 1 and the cost weight CE to 0; if MP < YP, then set the time weight TE of the first time period to MP / YP and the cost weight CE to 1 - MP / YP; Repeat to obtain the time weight and cost weight of all time periods within the first date to obtain the time cost weight information of the first date; and repeat to obtain the time cost weight information of all historical dates to obtain the time cost weight data.

7. The employee task dynamic scheduling control method based on virtual digital identity according to claim 6 is characterized in that: Obtaining the time weight and cost weight of the current date based on the time cost weight data and simultaneously scheduling multiple tasks based on the auction algorithm mechanism to obtain all optional scheduling schemes further includes the following sub-steps: For the current date to be regulated, denoted as the first current date; any time period in the first current date is denoted as the first current time period; Obtain, according to the date type of the first current date, the k2 historical dates that are the same as and the closest to the date type of the first current date before the first current date, denoted as the weight reference dates of the first current date, where k2 is the set number; According to the time cost weight information of the weight reference dates, obtain the time weight and cost weight of the time periods that are the same as the first current time period among all the weight reference dates, and sort them in the corresponding date order respectively, and denote them as the time weight reference sequence and the cost weight reference sequence respectively in order; Set weight coefficients for each weight reference date according to the time distance between each weight reference date and the first current date, and ensure that the sum of the weight coefficients of all weight reference dates is equal to 1; Then calculate the weighted average values of the time weight reference sequence and the cost weight reference sequence respectively according to the weight coefficients of each weight reference date, and obtain the average time weight and the average cost weight respectively in order, and mark them as the current time weight and the current cost weight of the first current time period respectively in order; ​ 8. The employee task dynamic scheduling control method based on virtual digital identity according to claim 7 is characterized in that: Based on the time-cost weight data, the time weight and cost weight of the current date are obtained, and multiple tasks are scheduled simultaneously based on the auction algorithm mechanism. Obtaining all optional scheduling solutions also includes the following sub-steps: For the first current period, obtain the number of currently idle virtual digital employees, recorded as LA; and record any idle virtual digital employee as the first idle employee; And get the current number of tasks to be scheduled, recorded as LB; Based on the auction algorithm mechanism, LC tasks to be scheduled are auctioned simultaneously according to LA and LB. If LA>LB, then LC=LB, otherwise LC=LA. The estimated processing time and estimated processing cost given by the first idle employee for each auctioned task are obtained, recorded as processing time and processing cost respectively, to obtain the task processing data of the first idle employee. The task processing data of all idle virtual digital employees are repeatedly obtained, recorded as task processing cost data. Assign LC tasks to be scheduled to LA idle virtual digital employees, obtain all allocation plans, and record them as optional scheduling plans; For any optional scheduling scheme, record it as the first scheduling scheme, calculate the sum of the estimated processing costs of all tasks in the first scheduling scheme, record it as the processing cost of the first scheduling scheme, sort the estimated processing time of all tasks in the first scheduling scheme from small to large, and obtain the k3th percentile, record it as the processing time of the first scheduling scheme; where k3 is the set percentile; repeatedly obtain the processing time and processing cost of each optional scheduling scheme.

9. The employee task dynamic scheduling control method based on virtual digital identity according to claim 8 is characterized in that: Selecting the best scheduling scheme from all available scheduling schemes based on the time weight and cost weight of the current date to schedule tasks includes the following sub-steps: Sort the processing time and processing cost of all optional scheduling solutions in descending order to obtain the processing time sequence and processing cost sequence; For the first scheduling scheme, obtain the position of the processing time and processing cost of the first scheduling scheme in the corresponding processing time sequence and processing cost sequence, and record them as AG1 and BG1 respectively in order; and calculate the processing time weight QG1 and processing cost weight QG2 of the first scheduling scheme, where QG1 = (AG1 / W0) / [(AG1 / W0)+(BG1 / W0)], QG2 = (BG1 / W0) / [(AG1 / W0)+(BG1 / W0)], repeatedly obtain the processing time weights and processing cost weights of all first scheduling schemes, and sort them according to size, and record them as the time weight sequence and cost weight sequence respectively; If the current time weight TE of the first current period is 1 and the current cost weight CE is 0, then the optional scheduling scheme with the largest processing time weight is marked as the best scheduling scheme; Otherwise, obtain VE and NE, VE=MAX(TE, CE), NE=MIN(TE, CE), and record the duration weight sequence or the cost weight sequence as the first sequence according to the weight type of VE, and record the other as the second sequence; Obtain the optional scheduling scheme located at [k4*VE, k5*VE] in the first sequence, and record it as the candidate scheduling scheme; for any candidate scheduling scheme, record it as the first candidate scheduling scheme, obtain the processing cost weight by the processing time weight corresponding to the first candidate scheduling scheme and VE, record it as VG, and record the other one as NG; and calculate the weight difference XG of the first candidate scheduling scheme, XG = (|VE-VG|\VE) + (|NE-NG|\NE); repeatedly obtain the weight difference of all candidate scheduling schemes, and record the candidate scheduling scheme with the smallest weight difference as the best scheduling scheme; Perform task allocation and scheduling according to the optimal scheduling plan.

10. A virtual digital identity-based employee task dynamic scheduling control system, used to implement the virtual digital identity-based employee task dynamic scheduling control method according to any one of claims 1 to 9, characterized in that: It includes data acquisition module, threshold weight module, scheduling preset module and task scheduling module; The data collection module is used to collect the task volume of each time period on different dates and the completion status of each task to obtain historical task reference data; The threshold weight module includes a threshold extraction unit and a weight allocation unit. The threshold extraction unit extracts the task density dynamic threshold based on the historical task reference data, and the weight allocation unit performs weight allocation processing according to the task density dynamic threshold to obtain time cost weight data. The scheduling preset module obtains the time weight and cost weight of the current date based on the time cost weight data, and schedules multiple tasks simultaneously based on the auction algorithm mechanism to obtain all optional scheduling solutions; The task scheduling module selects the best scheduling scheme from all optional scheduling schemes according to the time weight and cost weight of the current date to perform task scheduling.

Images