CN109190929B - Workload allocation method and device, computer equipment and storage medium - Google Patents
Workload allocation method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109190929B CN109190929B CN201810920356.5A CN201810920356A CN109190929B CN 109190929 B CN109190929 B CN 109190929B CN 201810920356 A CN201810920356 A CN 201810920356A CN 109190929 B CN109190929 B CN 109190929B
- Authority
- CN
- China
- Prior art keywords
- workload
- preset
- target object
- work information
- capacity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063112—Skill-based matching of a person or a group to a task
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06398—Performance of employee with respect to a job function
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a workload allocation method, a workload allocation device, computer equipment and a storage medium, wherein the workload allocation method comprises the following steps: and acquiring working information corresponding to each employee corresponding to each target object, inputting the working information corresponding to each employee into a pre-trained deep learning model as input, so as to obtain a working capacity evaluation value corresponding to each employee, counting each object capacity value corresponding to each target object, and distributing the preset work load to be distributed to each target object according to the ratio of each object capacity value corresponding to each target object in the sum of the object capacity values corresponding to all target objects. Because the deep learning model is obtained by collecting a large amount of historical data as samples in advance and training, the output working capacity evaluation value is very close to the current working capacity evaluation value of staff, so that the matching degree of workload allocation is improved.
Description
Technical Field
The present invention relates to the field of finance, and in particular, to a workload distribution method, a workload distribution device, a workload distribution computer device, and a workload distribution storage medium.
Background
In the current era of increasing competition, many finance companies are very important in keeping with the execution of work plans in order to preempt a larger market share.
In order to better execute a work plan, a finance company often needs to distribute a preset work load to be distributed to target objects such as each subordinate department or each subordinate branch company, but due to reasons such as fluctuation of staff of each target object, poor working state of the staff or scale change of each target object, the work load distributed to each target object is not matched with the work capacity of each target object, so that the matching degree of work load distribution is low.
Disclosure of Invention
In view of the above, it is necessary to provide a workload distribution method, a workload distribution device, a computer apparatus, and a storage medium, which can improve the workload distribution accuracy.
A workload distribution method, comprising:
acquiring working information corresponding to each employee subordinate to different target objects to obtain a working information set, wherein the target object subordinate comprises more than one employee, and the employees are in one-to-one correspondence with the working information;
inputting each piece of working information in the working information set as input into a pre-trained deep learning model to obtain each first output result as a working capacity evaluation value of each employee, wherein the deep learning model is obtained by training historical working information and historical working capacity evaluation values of employees corresponding to the historical working information as samples, the working information corresponds to the first output results one by one, and the employees correspond to the working capacity evaluation values one by one;
Counting the working capacity evaluation values of all staff subordinate to each target object to obtain the object capacity value of each target object, wherein the target objects correspond to the object capacity values one by one;
and according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects, the preset workload to be allocated is allocated to each target object.
A workload distribution device, comprising:
the system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring work information corresponding to each employee subordinate to different target objects to obtain a work information set, the target object subordinate comprises more than one employee, and the employees are in one-to-one correspondence with the work information;
the input module is used for inputting each piece of working information in the working information set into a pre-trained deep learning model to obtain each first output result as a working capacity evaluation value of each employee, wherein the deep learning model is obtained by training historical working information and historical working capacity evaluation values of employees corresponding to the historical working information as samples, the working information corresponds to the first output results one by one, and the employees correspond to the working capacity evaluation values one by one;
The statistics module is used for counting the working capacity evaluation values of all staff subordinate to each target object to obtain the object capacity value of each target object, wherein the target objects correspond to the object capacity values one by one;
and the allocation module is used for allocating the preset workload to be allocated to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the workload distribution method described above when the computer program is executed.
A computer readable storage medium storing a computer program which, when executed by a processor, performs the steps of the workload distribution method described above.
According to the workload allocation method, the workload allocation device, the computer equipment and the storage medium, firstly, working information corresponding to each employee subordinate to different target objects is obtained to obtain a working information set, then, each piece of working information in the working information set is input into a pre-trained deep learning model to obtain a working capacity evaluation value of each employee, next, the object capacity value subordinate to each target object is counted, and finally, the preset workload to be allocated is allocated to each target object according to the ratio of the object capacity value of each target object in the sum of the object capacity values corresponding to all target objects. The deep learning model is used for collecting a large amount of work information and the work capacity evaluation value of the employee corresponding to the work information in advance, the work capacity evaluation value is obtained by training samples, and the samples are accurate historical data corresponding to the employee, so that the work capacity evaluation value output by the deep learning model is the accurate work capacity evaluation value of the employee corresponding to the input work information, the work capacity evaluation value corresponding to the employee output by the deep learning model is very close to the current work capacity evaluation value of the employee, the work capacity which can be completed and corresponding to each department is counted according to the current work capacity of each employee, and the accurate real-time counting method can definitely ensure that the work capacity which is allocated to each target object is more close to the work capacity which can be completed by each target object at the current stage, so that the matching degree of work capacity allocation is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic view of an application environment of a workload distribution method according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method of workload distribution in accordance with an embodiment of the present invention;
FIG. 3 is a flow chart of a deep learning model training process in a workload distribution method in accordance with an embodiment of the present invention;
FIG. 4 is a flow chart of the method for workload assessment determination and self-correction of deep learning model in accordance with an embodiment of the present invention;
FIG. 5 is a flow chart of the method for workload distribution in which data sets are de-duplicated for items in accordance with an embodiment of the present invention;
FIG. 6 is a schematic view of a work-quantity apportioning apparatus according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a computer device in accordance with an embodiment of the invention.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The workload distribution method provided by the application can be applied to an application environment as shown in fig. 1, wherein the computer equipment communicates with a server through a network. The server acquires the working information corresponding to each employee subordinate to each target object of the client, the server inputs the working information as input into a pre-trained deep learning model to obtain each first output result as a working capacity evaluation value corresponding to each employee subordinate to each target object, the server respectively counts the working capacity evaluation values corresponding to all employees subordinate to each target object to obtain each object capacity value corresponding to each target object, and the server distributes the preset work load to be distributed to each target object according to the ratio of each object capacity value corresponding to each target object to the sum of the object capacity values corresponding to all target objects. The computer device may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, among others. The server may be implemented by a stand-alone server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, a workload distribution method is provided, and the workload distribution method is applied in the financial industry, and is illustrated by taking a server side of the method in fig. 1 as an example, and includes the following steps:
s10: and acquiring the corresponding working information of each employee subordinate to different target objects to obtain a working information set.
In this embodiment, there are more than one employee in the target object subordinate, and one employee corresponds to one piece of work information.
Specifically, the corresponding work information of each employee subordinate to different target objects is obtained through a preset obtaining mode to obtain a work information set.
It should be noted that, the preset obtaining manner may be manually collected or extracted through a database, the target object may be a company or a department, and the working information includes gender, age, academic and working years, etc., and the specific content of the preset obtaining manner, the target object and the working information may be set according to practical application, which is not limited herein.
For a better understanding of step S10, the following description will be given by way of an example, specifically expressed as follows:
for example, assuming that the preset acquisition mode is manual collection, the target object comprises a research and development department and a marketing department, the research and development department comprises three employees, the first employee is Zhang three, the second employee is Li four, the third employee is King five, the marketing department comprises four employees, the fourth employee is Huang Liu, the fifth employee is Liu Qi, the sixth employee is Luo eight, the seventh employee is Jiu, the working information comprises gender, age, academic and working years, and then the men, 35, doctor and 7-year information corresponding to Zhang three subordinate to the research and development department, the men, 30, shuoshi and 5-year information corresponding to Li four, and the collection of women, 26, university family and 4-year information corresponding to King five are acquired by manual collection; and acquiring information of men, 33, university, family and 10 years corresponding to yellow six subordinate to the marketing department, information of women, 31, specialty and 9 years corresponding to Liu Qi, information of men, 28, specialty and 6 years corresponding to Luo eight, and collection of information of women, 25, specialty and 3 years corresponding to nine women of the game through manual collection.
S20: and inputting each piece of work information in the work information set as input into a pre-trained deep learning model to obtain each first output result as a work capacity evaluation value of each employee.
In this embodiment, the working ability evaluation value refers to an evaluation value, which is obtained after the working ability evaluation is performed on an employee and is close to the employee, and the working information corresponds to the first output result one by one.
Specifically, each working information set in the acquired working information is input into a pre-trained deep learning model to obtain each first output result, and each first output result is used as a working capacity evaluation value of each employee, namely, the input working information corresponds to the obtained working capacity evaluation value one by one.
It should be noted that the deep learning model is trained by taking historical work capability evaluation values of historical data including historical work information and employees corresponding to the historical work information as samples.
S30: and counting the working capacity evaluation values of all staff subordinate to each target object to obtain the object capacity value of each target object.
In this embodiment, the target object has more than one object capability value, which is the sum of the work capability evaluation values corresponding to all employees subordinate to the target object, and the object capability values corresponding to each target object are different, and the target objects correspond to the object capability values one by one.
Specifically, the work ability evaluation values of all staff subordinate to each target object are counted to obtain the object ability value of each target object.
It should be noted that, the specific statistics process may be to count the target objects one by one in a sequential order, or to separately and concurrently execute statistics on each target object.
S40: and according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects, the preset workload to be allocated is allocated to each target object.
Specifically, first, object capability values corresponding to all target objects are added to obtain a sum of object capability values of all target objects, then, the duty ratio of the object capability value corresponding to each target object in the sum of the object capability values of all target objects is calculated, and then, the preset workload to be apportioned is apportioned to each target object according to the duty ratio.
It should be noted that, the preset workload to be allocated may be sales or development project tasks, and the specific content of the preset workload to be allocated may be set according to practical applications, which is not limited herein.
For a better understanding of step S40, the following description will be given by way of an example, specifically expressed as follows:
For example, assuming that the target object includes a first marketing department, a second marketing department and a third marketing department, the object capability value corresponding to the first marketing department is 2, the object capability value corresponding to the second marketing department is 3 and the object capability value corresponding to the third marketing department is 5, the preset workload to be shared is 1000 tens of thousands of car insurance policy, firstly, adding 2, 3 and 5 to obtain 10, and then, calculating the ratio of the object capability value corresponding to the first marketing department to 10 to be 10The first marketing department corresponds to an object ability value of a duty ratio of +.10>The object capability value corresponding to the first marketing department has a duty ratio of 10 ofNext, according to the duty ratio +.>And->The 1000 ten thousand car insurance premium is allocated to the first marketing department 200 ten thousand car insurance premium, the second marketing department 300 ten thousand car insurance premium and the third marketing department 500 ten thousand car insurance premium
In the embodiment corresponding to fig. 2, first, working information corresponding to each employee subordinate to different target objects is obtained to obtain a working information set, then each working information in the working information set is input into a pre-trained deep learning model to obtain a working capacity evaluation value of each employee, next, the object capacity value subordinate to each target object is counted, and finally, a preset workload to be apportioned is apportioned to each target object according to the ratio of the object capacity value of each target object in the sum of the object capacity values corresponding to all target objects. The deep learning model is used for collecting a large amount of work information and the work capacity evaluation value of the employee corresponding to the work information in advance, the work capacity evaluation value is obtained by training samples, and the samples are accurate historical data corresponding to the employee, so that the work capacity evaluation value output by the deep learning model is the accurate work capacity evaluation value of the employee corresponding to the input work information, the work capacity evaluation value corresponding to the employee output by the deep learning model is very close to the current work capacity evaluation value of the employee, the work capacity which can be completed and corresponding to each department is counted according to the current work capacity of each employee, and the accurate real-time counting method can definitely ensure that the work capacity which is allocated to each target object is more close to the work capacity which can be completed by each target object at the current stage, so that the matching degree of work capacity allocation is improved.
In an embodiment, the workload distribution method is applied in the financial industry, as shown in fig. 3, in step S20, that is, the deep learning model is trained in advance, which specifically includes the following steps:
s201: and acquiring historical work information and historical work capacity evaluation values of staff corresponding to the historical work information as samples.
In this embodiment, the historical work information refers to the historical data of the work information corresponding to the employee, and the historical work ability evaluation value refers to the historical data of the work ability evaluation value of the employee evaluated according to the historical work information.
Specifically, the acquired history data includes history work information and a history work ability evaluation value of an employee corresponding to the history work information as a sample.
S202: and inputting the historical work information in the sample as input into a deep learning model to obtain a second output result.
Specifically, the historical work information in the acquired sample is input into a deep learning model, and after analysis of the deep learning model, a second output result is obtained.
S203: and taking the second output result as a target, and adjusting hidden layer parameters of the deep learning model to minimize errors of historical work capacity evaluation values of staff corresponding to the historical work information in the sample by the second output result.
In this embodiment, the hidden layer parameters include the number of neural nodes, the step constant per improvement, the target accuracy, the maximum number of iterations, and the cost function.
Specifically, the second output result is taken as an output target, and hidden layer parameters of the deep learning model are continuously adjusted, so that the error minimization of the historical work ability evaluation value of the employee, corresponding to the historical work information in the sample, of the second output result is achieved.
It should be noted that, in the adjustment parameter, the step constant is corrected by first adjusting the step constant, that is, observing the rate of decrease of the cost function, so as to quickly decrease on one hand, and prevent the misconvergence on the other hand, then adjusting the number of hidden layer nodes after adjusting appropriately, gradually increasing, the accuracy rate should theoretically be first increased and then decreased, and after finding the appropriate number of nodes, finally gradually increasing the target accuracy rate.
Further, the preset condition may be 0.01% or 0.015%, and the specific content of the preset condition may be set according to the actual application, which is not limited herein.
Further, it is determined whether the error of the historical work ability evaluation value of the employee corresponding to the historical work information in the sample and the second output result meets the preset condition, if yes, step S204 is executed, and if not, steps S201 to S203 are executed again until the error meets the preset condition.
S204: and determining the current deep learning model as a trained deep learning model.
Specifically, if the error of the historical work capacity evaluation value of the employee corresponding to the historical work information in the sample and the second output result meet the preset condition, determining that the current deep learning model is a trained deep learning model; if the error of the historical work capacity evaluation value of the employee corresponding to the historical work information in the sample and the second output result does not meet the preset condition.
In the embodiment corresponding to fig. 3, by acquiring the historical work information and the historical work capability evaluation value of the employee corresponding to the historical work information as a sample, inputting the historical work information in the sample as input into the deep learning model, obtaining a second output result as a target, adjusting the hidden layer parameters of the deep learning model to minimize the error of the historical work capability evaluation value of the employee corresponding to the second output result and the historical work information in the sample, and if the error meets the preset condition, determining that the current deep learning model is a trained deep learning model. Because a large amount of historical data is adopted as a sample, the historical data comprise historical work information and historical work capability evaluation values of the staff corresponding to the historical work information, meanwhile, the historical data are obtained according to the actual situation of the staff, the historical work information in the sample is taken as input to a deep learning model to obtain an output result, then the output result is used for adjusting the output target of the deep learning model most, and the hidden layer parameters of the deep learning model are continuously adjusted by adopting the actual and effective historical work information and the historical work capability evaluation values of the staff corresponding to the historical work information, so that the error of the output result of the deep learning model and the historical work capability evaluation values of the staff corresponding to the historical work information in the sample is minimized, and the accuracy of the work capability evaluation values of the staff output by the deep learning model is ensured.
In an embodiment, the workload distribution method is applied in the financial industry, as shown in fig. 4, in step S20, each piece of work information in the work information set is input into a pre-trained deep learning model, and after each piece of first output result is obtained, the workload distribution method specifically further includes the following steps:
s50: and judging whether the first output result meets a preset threshold condition.
Specifically, whether the first output result of the deep learning model meets a preset threshold condition is judged.
It should be noted that, the preset threshold condition defines the standard range of the working ability evaluation value corresponding to the employee, the preset threshold condition may be from the 9 working ability evaluation value to the 12 working ability evaluation value, and the specific content of the preset threshold condition may be set according to the actual application, which is not limited herein.
S60: and if the first output result does not meet the preset threshold condition, determining the input corresponding to each piece of work information in the work information set and the first output result as negative samples.
Specifically, if the first output result of the deep learning model does not meet the preset threshold condition, taking the input and the first output result of the deep learning model corresponding to the work information of each employee in the work information set as negative samples.
S70: and if the first output result meets the preset threshold condition, determining the first output result as a work capacity evaluation value corresponding to the employee, and taking the input corresponding to each piece of work information in the work information set and the first output result as positive samples.
Specifically, if the first output result of the deep learning model meets a preset threshold condition, determining the first output result as a work capacity evaluation value corresponding to the employee, and taking the input of the deep learning model corresponding to the work information of each employee in the work information set and the first output result as positive samples.
In the embodiment corresponding to fig. 4, by determining whether the first output result meets the preset threshold condition, if not, taking the input corresponding to each piece of work information in the work information set and the first output result as negative samples, if so, determining the first output result as a work capacity evaluation value corresponding to the employee, and taking the input corresponding to each piece of work information in the work information set and the first output result as positive samples. Because the working capacity evaluation value of the staff has a standard range, whether the output result of the deep learning model is within the standard range or not needs to be judged, when the output result of the deep learning model is not within the standard range, the working information corresponding to the staff and the output result are used as negative samples, and when the output result of the deep learning model is within the standard range, the output result is used as the working capacity evaluation value corresponding to the staff and the working information corresponding to the staff and the output result are used as positive samples, so that the effect of training and correcting errors can be achieved while the analysis of the deep learning model to obtain the output result can be ensured, and the accuracy of the analysis capacity of the deep learning model is improved.
In an embodiment, the workload allocation method is applied in the financial industry, the preset workload to be allocated includes a data type, the data type includes a number of copies, a balance and a year premium, and step S40 specifically includes the steps of allocating the preset workload to be allocated to each target object according to a ratio of an object capability value of each target object to a sum of object capability values of all target objects, including the following steps:
s401: if the preset data type of the workload to be allocated is a combination of the number of copies, the amount of the premium and the annual premium, allocating the preset number of copies of the workload to be allocated to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects.
In this embodiment, the data type may be one of the three of the number of copies, the amount of the premium and the annual premium, or may be a combination of two of the three, or a combination of the number of copies, the amount of the premium and the annual premium.
Specifically, if the data type of the preset workload to be allocated is a combination of the number of copies, the amount of insurance and the annual premium, the number of copies is taken as a priority allocation field, and the number of copies of the preset workload to be allocated to each target object is allocated to the target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects, wherein the amount of insurance and the annual premium are taken as standby allocation fields.
S402: if the data type of the preset workload to be allocated is a combination of the amount of the deposit and the annual fee, allocating the amount of the deposit of the preset workload to be allocated to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects.
Specifically, if the data type of the preset workload to be allocated is a combination of a premium and a annual premium, the premium is taken as a priority allocation field, and the premium of the preset workload to be allocated is allocated to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects, wherein the annual premium is taken as a standby allocation field.
Further, if the data type of the preset workload to be shared is a combination of the number of copies and the guard, the number of copies is used as a priority sharing field, and the number of copies of the preset workload to be shared is shared to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects, wherein the guard is used as a standby sharing field.
S403: if the preset data type of the workload to be shared is annual premium, sharing the annual premium of the workload to be shared to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects.
Specifically, if the preset data type of the workload to be allocated is annual premium, allocating the annual premium of the preset workload to be allocated to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects.
Further, if the data type of the preset workload to be allocated is a combination of the number of copies and the annual premium, the number of copies is used as a priority allocation field, and the number of copies of the preset workload to be allocated is allocated to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects, wherein the annual premium is used as a standby allocation field.
The specific contents of the priority allocation field and the spare allocation field may be set according to the actual application, and are not limited herein.
In an embodiment, if the data type of the preset workload to be shared is a combination of the number of copies, the balance and the annual payment premium, the number of copies of the preset workload to be shared is shared to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all target objects, if the data type of the preset workload to be shared is a combination of the balance and the annual payment premium, the balance of the preset workload to be shared is shared to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all target objects, and if the data type of the preset workload to be shared is the annual payment premium, the annual payment premium of the preset workload to be shared is shared to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all target objects. Because the content of the data type is a plurality of fields, when the workload to be shared is shared to each target object, the problem that the workload to be shared lacks the shared field to cause the sharing failure can be avoided, namely when the workload to be shared is shared to each target object, the workload to be shared does not have the shared field to cause the sharing failure, or the workload to be shared does not have the shared field or the shared field to cause the sharing failure, thereby improving the success rate of workload sharing.
In an embodiment, the workload distribution method is applied in the financial industry, as shown in fig. 5, in step S10, before the work information corresponding to each employee subordinate to different target objects is obtained to obtain the work information set, the workload distribution method specifically further includes the following steps:
s80: a first data set and a second data set are acquired.
In one embodiment, the first data set and the second data set are data tables, the data tables include more than one column and more than one row, and the items refer to rows in the first data set and the second data set, that is, the first data set and the second data set are combined by more than one item.
Specifically, the first data set and the second data set are acquired through a preset acquisition mode.
It should be noted that, the preset obtaining mode may be obtained from a financial system, or may be obtained by manually filling a table, and the specific content of the preset obtaining mode may be set according to the actual application, which is not limited herein.
S90: classifying the acquired first data set according to a preset category to obtain a first sub-data set.
In this embodiment, the preset category refers to a category that is divided according to a service function. The acquired first data set refers to a data combination related to the service.
Specifically, the obtained first data set is classified according to a preset category to obtain a first sub-data set.
It should be noted that, the preset category may be Ind, goat, guar, UV, ul, short, pua or wb, where Ind is a traditional risk, goat is a goat risk, guar is a guarantee continuation, UV is a universal risk, ul is a throw continuous risk, short is a short risk, pua is a payoff and increase, and wb is a foreign currency risk. The obtained first data set may be an insurance data information table or a house source information table, etc., and specific content of the preset category may be set according to practical application, which is not limited herein.
S100: and classifying the acquired second data set according to the preset category to obtain a second sub-data set.
In this embodiment, the obtained second data set is an insurance data information table or a room source information table, etc.
Specifically, the acquired second data set is classified according to a preset category to obtain a second sub-data set.
Further, the acquired first data set and the acquired second data set may be stored in a database, or may be stored in other storage devices such as a disk.
S110: and taking each item in the subarray after the items are de-duplicated by the first subarray and the second subarray as a preset workload to be apportioned.
Specifically, each item in the subarray after the items are subjected to the duplicate removal of the first subarray and the second subarray is used as a preset workload to be allocated.
For a better understanding of step S80, step S90 and step S100, the following description will be given by way of an example, specifically expressed as follows:
for example, assuming that the preset categories are car insurance, accident insurance and major-illness hospitalization insurance, the preset first data set is an insurance combination information table, and the preset second data set is an insurance combination information table, classifying the insurance combination information table according to car insurance, accident insurance and major-illness hospitalization insurance to obtain a first sub-data set as a first table, a second sub-data set as a second table, and performing item de-duplication on the first table and the second table to obtain items in a third table, a fourth table and a fifth table as preset workload to be shared, wherein "car insurance implementation part chinese 100000", "accident insurance implementation part chinese 200000" and "major-illness hospitalization insurance implementation part chinese 300000" are items, and the first table, the second table, the third table, the fourth table and the fifth table are specifically as follows:
first form
Second form
| Product(s) | Department(s) | Region(s) | The current occurrence amount |
| Vehicle danger | Vehicle insurance implementation part | China | 1000000 |
| Risk of serious illness in hospital | Life insurance implementing part | China | 300000 |
Third table
| Product(s) | Department(s) | Region(s) | Sales amount |
| Vehicle danger | Vehicle insurance implementation part | China | 1000000 |
| Accident risk | Accident risk implementation part | China | 200000 |
Fourth table
| Product(s) | Department(s) | Region(s) | Sales amount |
| Vehicle danger | Vehicle insurance implementation part | China | 1000000 |
Fifth table
| Product(s) | Department(s) | Region(s) | The current occurrence amount |
| Risk of serious illness in hospital | Life insurance implementing part | China | 300000 |
Is a table of five tables of (a).
In the embodiment corresponding to fig. 5, first, a preset first data set is classified according to a preset category to obtain a first sub-data set, then a preset second data set is classified according to a preset category to obtain a second sub-data set, and finally, a work task of completing a sub-array after the first sub-data set and the second sub-data set are de-duplicated is used as a preset work load to be allocated. Because the preset first data set and the preset second data set are grouped according to the preset categories, and the sub-data sets after grouping are subjected to de-duplication processing, repeated workload can be prevented from being shared in time, and the workload sharing efficiency is improved.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
In one embodiment, a workload distribution device is provided, and the workload distribution device corresponds to the workload distribution method in the embodiment one by one. As shown in fig. 6, the workload distribution device includes an acquisition module 71, an input module 72, a statistics module 73, and a distribution module 74. The functional modules are described in detail as follows:
the first obtaining module 71 is configured to obtain work information corresponding to each employee subordinate to different target objects to obtain a work information set, where the target objects subordinate to each employee include more than one employee, and the employees are in one-to-one correspondence with the work information;
the input module 72 is configured to input each piece of work information in the work information set into a pre-trained deep learning model, and obtain each first output result as a work ability evaluation value of each employee, where the work information corresponds to the first output result one to one, and the employee corresponds to the work ability evaluation value one to one;
a statistics module 73, configured to count all employee performance evaluation values subordinate to each target object, and obtain an object performance value of each target object, where the target objects are in one-to-one correspondence with the object performance values;
An allocation module 74, configured to allocate the preset workload to be allocated to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all target objects.
Optionally, the input module 72 includes:
an extraction sub-module 721, configured to obtain historical work information and a historical work capability evaluation value of an employee corresponding to the historical work information as a sample;
a adding sub-module 722, configured to input the historical working information in the sample as input to the deep learning model, so as to obtain a second output result;
an adjustment sub-module 723 for adjusting hidden layer parameters of the deep learning model with the second output result as a target to minimize an error of a historical work ability evaluation value of the employee, the second output result corresponding to the historical work information in the sample;
and the determining submodule 724 is configured to determine that the current deep learning model is a trained deep learning model if the error meets a preset condition.
Optionally, the workload distribution device further comprises:
a judging module 75, configured to judge whether the first output result meets a preset threshold condition;
a first module 76, configured to determine, as a negative example, an input corresponding to each piece of work information in the work information set and the first output result if the first output result does not meet a preset threshold condition;
And the judging module 77 is configured to determine the first output result as a work ability evaluation value corresponding to the employee if the first output result meets a preset threshold condition, and determine the input corresponding to each piece of work information in the work information set and the first output result as positive samples.
Optionally, the apportionment module 74 includes:
a first amortization sub-module 741, configured to, if the data type of the preset workload to be amortized is a combination of the number of copies, the amount of insurance and the annual premium, amortize the number of copies of the preset workload to be amortized to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects;
the second dispatch sub-module 742 is configured to, if the data type of the preset workload to be apportioned is a combination of a premium and a annual premium, apportion the premium of the preset workload to each target object according to a ratio of the object capability value of each target object to a sum of the object capability values of all the target objects;
and the third amortization sub-module 743 is configured to, if the preset data type of the workload to be amortized is annual premium, amortize the annual premium of the workload to be amortized to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects.
Optionally, the workload distribution device further comprises:
a second acquisition module 78 for acquiring the first data set and the second data set;
a first obtaining module 79, configured to classify the obtained first data set according to a preset category to obtain a first sub-data set;
a second obtaining module 710, configured to classify the obtained second data set according to a preset category to obtain a second sub-data set;
and a second module 711, configured to use each item in the sub-array obtained by de-duplicating the items in the first sub-data set and the second sub-data set as a preset workload to be apportioned.
For specific limitations of the workload distribution device, reference may be made to the above limitations of the workload distribution method, and no further description is given here. The various modules in the workload distribution device described above may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for data related to the workload distribution method. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a workload distribution method.
In one embodiment, a computer device is provided that includes a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the workload distribution method of the above embodiments when executing the computer program, such as steps S10 to S40 shown in fig. 2. Alternatively, the processor, when executing the computer program, implements the functions of the modules/units of the workload distribution device in the above embodiment, such as the functions of the acquisition module 71 to the distribution module 74 shown in fig. 6. In order to avoid repetition, a description thereof is omitted.
In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the method for allocating work amounts in the above-described method embodiment, or which when executed by a processor implements the functions of the modules/units in the work amount allocating apparatus in the above-described apparatus embodiment. In order to avoid repetition, a description thereof is omitted. Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.
Claims (7)
1. A method of workload distribution, the method comprising:
acquiring working information corresponding to each employee subordinate to different target objects to obtain a working information set, wherein the target object subordinate comprises more than one employee, and the employees are in one-to-one correspondence with the working information;
Inputting each piece of work information in the work information set into a pre-trained deep learning model to obtain an output set, wherein the output set is a work capacity evaluation value set of each employee, the deep learning model is obtained by training historical work information and historical work capacity evaluation values of employees corresponding to the historical work information as samples, and historical work information and historical work capacity evaluation values of employees corresponding to the historical work information are obtained as samples; inputting the historical work information in the sample into the deep learning model to obtain a second output result; taking the second output result as a target, and adjusting hidden layer parameters of the deep learning model to minimize errors of historical work capacity evaluation values of staff corresponding to the historical work information in the sample by the second output result;
if the error meets a preset condition, determining that the current deep learning model is a trained deep learning model;
counting the working capacity evaluation values of all staff subordinate to each target object to obtain an object capacity value of each target object, wherein the target objects correspond to the object capacity values one by one;
According to the ratio of the object capability value of each target object in the sum of the object capability values of all target objects, distributing the preset workload to be distributed to each target object, wherein the preset workload to be distributed comprises data types, the data types comprise the number of copies, the amount of insurance and the annual fee, and distributing the preset workload to be distributed to each target object according to the ratio of the object capability value of each target object in the sum of the object capability values of all target objects comprises the following steps: if the data type of the preset workload to be allocated is a combination of the number of copies, the amount of insurance and the annual premium, allocating the number of copies of the preset workload to be allocated to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects; if the data type of the preset workload to be allocated is a combination of the amount of the deposit and the annual fee, allocating the amount of the deposit of the preset workload to be allocated to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects; and if the data type of the preset workload to be allocated is annual premium, allocating the annual premium of the preset workload to be allocated to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects.
2. The workload distribution method according to claim 1, wherein after inputting each of the set of work information to a pre-trained deep learning model to obtain an output set, the workload distribution method further comprises:
judging whether each output value in the output set meets a preset threshold condition, wherein the threshold condition limits the standard range of the working capacity evaluation value of the staff;
if each output value in the output set does not meet a preset threshold condition, determining the working information set and the output set as negative samples;
and if each output value in the output set meets a preset threshold condition, determining a working capacity evaluation value set for all staff in the output set, and determining the working information set and the output set as positive samples.
3. A workload distribution method according to any one of claims 1 to 2, wherein before said obtaining the work information corresponding to each employee subordinate to the different target objects to obtain the work information set, the workload distribution method further comprises:
acquiring a first data set and a second data set;
Classifying the acquired first data set according to a preset category to obtain a first sub-data set, wherein the first data set comprises more than one item;
classifying the acquired second data set according to a preset category to obtain a second sub-data set, wherein the second data set comprises more than one item;
and determining each item in the sub-arrays after the items are de-duplicated by the first sub-data set and the second sub-data set as the preset workload to be apportioned.
4. A workload distribution device, characterized in that the workload distribution device comprises:
the system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring work information corresponding to each employee subordinate to different target objects to obtain a work information set, the target object subordinate comprises more than one employee, and the employees are in one-to-one correspondence with the work information;
the input module is configured to input each piece of work information in the work information set to a pre-trained deep learning model, and obtain each first output result as a work ability evaluation value of each employee, where the deep learning model is obtained by training historical work information and a historical work ability evaluation value of an employee corresponding to the historical work information as a sample, the work information corresponds to the first output result one to one, and the employee corresponds to the work ability evaluation value one to one, and the input module includes: the extraction sub-module is used for acquiring historical work information and historical work capacity evaluation values of staff corresponding to the historical work information as samples; the adding submodule is used for inputting the historical work information in the sample as input into the deep learning model to obtain a second output result; the adjusting sub-module is used for taking the second output result as a target and adjusting hidden layer parameters of the deep learning model so as to minimize errors of historical work capacity evaluation values of staff corresponding to the second output result and the historical work information in the sample; the determining submodule is used for determining that the current deep learning model is a trained deep learning model if the error meets a preset condition;
The statistics module is used for counting the working capacity evaluation values of all staff subordinate to each target object to obtain the object capacity value of each target object, wherein the target objects correspond to the object capacity values one by one;
an allocation module, configured to allocate a preset workload to be allocated to each target object according to a ratio of an object capability value of each target object to a sum of object capability values of all target objects, where the preset workload to be allocated includes a data type, and the data type includes a number of copies, a balance, and a year premium, and allocate the preset workload to be allocated to each target object according to a ratio of an object capability value of each target object to a sum of object capability values of all target objects, where the allocation module includes: if the data type of the preset workload to be allocated is a combination of the number of copies, the amount of insurance and the annual premium, allocating the number of copies of the preset workload to be allocated to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects; if the data type of the preset workload to be allocated is a combination of the amount of the deposit and the annual fee, allocating the amount of the deposit of the preset workload to be allocated to each target object according to the ratio of the object capacity value of each target object to the sum of the object capacity values of all the target objects; and if the data type of the preset workload to be allocated is annual premium, allocating the annual premium of the preset workload to be allocated to each target object according to the ratio of the object capability value of each target object to the sum of the object capability values of all the target objects.
5. The workload distribution device of claim 4, wherein after the input module, the workload distribution device further comprises:
the judging module is used for judging whether the first output result meets a preset threshold condition, wherein the threshold condition limits the standard range of the working capacity evaluation value corresponding to the staff;
the first module is configured to determine, if the first output result does not meet a preset threshold condition, an input corresponding to each piece of work information in the work information set and the first output result as negative samples;
and the first module is used for determining the first output result as a working capacity evaluation value corresponding to the staff if the first output result meets a preset threshold condition, and determining the input corresponding to each piece of working information in the working information set and the first output result as positive samples.
6. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the workload distribution method according to any of the claims 1 to 3 when the computer program is executed.
7. A computer-readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the workload distribution method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810920356.5A CN109190929B (en) | 2018-08-14 | 2018-08-14 | Workload allocation method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810920356.5A CN109190929B (en) | 2018-08-14 | 2018-08-14 | Workload allocation method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109190929A CN109190929A (en) | 2019-01-11 |
| CN109190929B true CN109190929B (en) | 2023-09-19 |
Family
ID=64921322
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810920356.5A Active CN109190929B (en) | 2018-08-14 | 2018-08-14 | Workload allocation method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109190929B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909915B (en) * | 2019-10-16 | 2022-08-12 | 厦门美柚股份有限公司 | Method and device for predicting work plan |
| CN111126860B (en) * | 2019-12-26 | 2023-12-26 | 科大讯飞华南人工智能研究院（广州）有限公司 | Task allocation method, task allocation device and electronic equipment |
| CN117541021A (en) * | 2024-01-04 | 2024-02-09 | 北京宜通华瑞科技有限公司 | Pipeline task management system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107301497A (en) * | 2017-05-27 | 2017-10-27 | 西南交通大学 | The method and system of working memory ability evaluation |
| CN107578183A (en) * | 2017-09-18 | 2018-01-12 | 上海量明科技发展有限公司 | Method for managing resource and device based on capability evaluation |
| WO2018051179A1 (en) * | 2016-09-15 | 2018-03-22 | Omron Corporation | Evaluation system, evaluation device, computer-implemented evaluation method, computer program and ai system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005276168A (en) * | 2004-02-27 | 2005-10-06 | Vitalnot:Kk | Personnel evaluation method, personnel evaluation system, information processor, and personnel evaluation program |
| US20080126285A1 (en) * | 2006-11-02 | 2008-05-29 | International Business Machines Corporation | Method, Computer Program Product, And System For Automatic Software Provisioning Based On Learning History And Competency Level |
| US20080243575A1 (en) * | 2007-03-30 | 2008-10-02 | Keith Weinberger | System and Method for Dynamically Allocating Human Resources to a Project Plan |
-
2018
- 2018-08-14 CN CN201810920356.5A patent/CN109190929B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018051179A1 (en) * | 2016-09-15 | 2018-03-22 | Omron Corporation | Evaluation system, evaluation device, computer-implemented evaluation method, computer program and ai system |
| CN107301497A (en) * | 2017-05-27 | 2017-10-27 | 西南交通大学 | The method and system of working memory ability evaluation |
| CN107578183A (en) * | 2017-09-18 | 2018-01-12 | 上海量明科技发展有限公司 | Method for managing resource and device based on capability evaluation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109190929A (en) | 2019-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020248843A1 (en) | Big data-based profile analysis method and apparatus, computer device, and storage medium | |
| CN109858740B (en) | Enterprise risk assessment method and device, computer equipment and storage medium | |
| CN109190929B (en) | Workload allocation method and device, computer equipment and storage medium | |
| Shi et al. | Cross-sample entropy statistic as a measure of synchronism and cross-correlation of stock markets | |
| EP3825861A2 (en) | Method and apparatus of user clustering, computer device | |
| US10565022B2 (en) | Systems for parallel processing of datasets with dynamic skew compensation | |
| US11074516B2 (en) | Load balancing for distributed processing of deterministically assigned data using statistical analysis of block data | |
| CN107622326B (en) | User classification and available resource prediction method, device and equipment | |
| WO2020015143A1 (en) | Case reminding method and apparatus, computer device and storage medium | |
| CN109325796B (en) | Potential user screening method, device, computer equipment and storage medium | |
| CN111209929A (en) | Access data processing method and device, computer equipment and storage medium | |
| CN108764553B (en) | User scale prediction method and device and computer equipment | |
| WO2019019753A1 (en) | Judgement method and apparatus for providing health report, computer device and storage medium | |
| CN112016765A (en) | Asset allocation method and device based on quantitative investment risk preference and terminal | |
| CN112464059A (en) | Power distribution network user classification method and device, computer equipment and storage medium | |
| Fu et al. | Logistics enterprise evaluation model based on fuzzy clustering analysis | |
| CN109542947B (en) | Data statistical method, device, computer equipment and storage medium | |
| CN110705816B (en) | Task allocation method and device based on big data | |
| CN114372681A (en) | Enterprise classification method, device, equipment, medium and product based on pipeline data | |
| CN114881761A (en) | Determination method of similar sample and determination method of credit limit | |
| CN106156470B (en) | Time series abnormity detection and labeling method and system | |
| US20210350468A1 (en) | Network graph outlier detection for identifying suspicious behavior | |
| Anfuso et al. | A sound basel iii compliant framework for backtesting credit exposure models | |
| CN112667754A (en) | Big data processing method and device, computer equipment and storage medium | |
| Li et al. | High-yield versus investment-grade bonds: less risk and greater returns? |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |