CN112258150A

CN112258150A - Person portrait based controller scheduling system and method

Info

Publication number: CN112258150A
Application number: CN202011109618.3A
Authority: CN
Inventors: 余波; 李巍巍; 李海凉; 郑亦斌; 钱江
Original assignee: EASTERN CHINA AIR TRAFFIC MANAGEMENT BUREAU CAAC
Current assignee: EASTERN CHINA AIR TRAFFIC MANAGEMENT BUREAU CAAC
Priority date: 2020-10-16
Filing date: 2020-10-16
Publication date: 2021-01-22

Abstract

The invention discloses a controller scheduling system and method based on figure portrait, belonging to the field of air traffic control. Aiming at the problem that the intelligent scheduling can not be completely realized in the prior art, the invention provides a basic setting module which configures and checks basic information in a configurable mode; the controller portrait module is used for checking the portrait of the controller; and the automatic shift scheduling module is used for generating and outputting a shift scheduling plan. The method can realize that the system obtains the figure portrait from a plurality of data sources, and the intelligent shift scheduling is carried out through the system based on the method, thereby changing the traditional allocation mode of controlling human resources. Meanwhile, the portrait data and the scheduling data are shared to systems such as performance assessment, administrative management and the like, so that the management level is shortened, and the original management mode is changed.

Description

Person portrait based controller scheduling system and method

Technical Field

The invention relates to the field of air traffic control, in particular to a person portrait based controller scheduling system and method.

Background

The portrait technology is a comprehensive technology, and the underlying technology mainly comprises big data and machine learning. The purpose of the big data is to record information as input for machine learning. The purpose of machine learning is to learn information and gradually optimize a matching degree model.

The scheduling of the existing controller is mainly performed through Excel, the operation and adjustment are complex, and a manager familiar with the corresponding scheduling mode needs to manage. And other personnel are difficult to get on the hands quickly, which causes poor continuity of business management and higher management cost. The schedule formats of all control rooms are not uniform, and personalized symbols are adopted in many cases. This makes it difficult to monitor the scheduling compliance and the cost of exchanging human resources between control rooms is high. Most of the schemes need fixed team shift, and if the shift is not scheduled according to the fixed team, the workload of personnel qualification matching and fatigue management is huge. The scheduling method of fixed groups also results in lower flexibility of human resource allocation. In the original scheduling mode, the personnel duty matching and the personnel fatigue degree verification of a controller are mainly judged by a manager. The method is lack of a system for data verification and cannot always consider the whole situation.

The Chinese patent application: a controller scheduling method based on portrait, which is applied for No. 2019, 08 and 15 months and public for No. 2019, 12 and 20 months, discloses a controller scheduling method based on machine learning, and comprises the following steps: step 1, a plurality of scheduling templates are made according to historical scheduling data of an air traffic control unit; each scheduling template comprises a plurality of scheduling modules; step 2, selecting a scheduling template which is most matched with the actual situation on the scheduling day, and filling a controller participating in scheduling on the scheduling day into the scheduling template; step 3, checking the filled scheduling template according to the scheduling rule, if all the scheduling modules in the scheduling template accord with the scheduling rule, executing step 6, otherwise executing step 4; step 4, setting the scheduling module which does not accord with the scheduling rule in the scheduling template as a blank module; step 5, machine learning is carried out on the historical shift scheduling data, and a proper controller is selected to be filled into a blank module; and 6, generating a schedule of the controllers. However, the method cannot completely realize intelligent shift arrangement, manual modification or adjustment of shift arrangement results is required, the matching performance of the shift arrangement results is not perfect, the acquired personnel data are single, and the visualization degree is poor.

Disclosure of Invention

1. Technical problem to be solved

Aiming at the problem that the intelligent scheduling can not be completely realized in the prior art, the invention provides a system and a method for scheduling a controller based on a character portrait, which can realize that the system obtains data from a plurality of data sources to obtain the character portrait and perform intelligent scheduling through the system, thereby changing the traditional allocation mode of controlling human resources. Meanwhile, the portrait data and the scheduling data are shared to systems such as performance assessment, administrative management and the like, so that the management level is shortened, and the original management mode is changed.

2. Technical scheme

A controller scheduling system based on figure portrait comprises a basic setting module, a basic information configuration module and a scheduling module, wherein the basic setting module is used for configuring basic information in a configurable mode; the controller portrait module is used for establishing portrait of the controller for viewing and selection based on the configuration information of the basic setting module; and the automatic scheduling module is used for generating a scheduling logic and outputting a scheduling plan according to the configuration information of the basic setting module and the selection information of the controller portrait module.

Further, the basic setting module comprises a class maintenance submodule for maintaining class; the new shift subsystem is used for creating detailed contents of the shift; setting a seat opening time sub-module to realize the maintenance of the seat opening time; configuring a shift personnel submodule for maintaining available shift personnel and shift types; and the unavailable time maintenance submodule is used for maintaining the unavailable time of the shift arrangement personnel.

Further, the controller representation module comprises an initial interface sub-module used for selecting the personnel needing to be checked; and the detailed interface sub-module is used for checking the details of the controller portrait.

Furthermore, the automatic scheduling module comprises a daily scheduling submodule for realizing automatic daily scheduling; the monthly shift submodule is used for realizing automatic monthly shift; and the release shift table submodule is used for releasing daily shift and monthly shift.

A controller scheduling method based on portrait comprises the following steps:

step 1: creating a controller portrait;

step 2: establishing a personnel matching model through the controller portrait, and calculating the matching degree scores of the controller on different posts;

and step 3: establishing automatic scheduling logic, determining seats to be arranged according to the scheduling setting, determining optional personnel from the scheduling personnel, and pre-scheduling and adjusting according to the principle of force balance of all the working personnel at each time interval according to the limiting conditions and the personnel matching degree;

and 4, step 4: and generating and outputting a shift arrangement result.

Further, the step 2 comprises:

step 201, selecting a sample: selecting different employees as samples according to the types of the posts;

step 202, label selection: selecting a corresponding controller portrait label according to the scoring basis of the sample;

step 203, acquiring data: acquiring selected label data;

step 204, data cleaning: cleaning the basic data by using a data processing method;

step 205, feature engineering conversion: performing characteristic engineering conversion on the existing data;

step 206, model building: inputting the cleaned data subjected to the characteristic work conversion into a model, and predicting the matching degree scores of the controller and different posts;

step 207, evaluation optimization: performing iterative optimization on the model according to the result output by the model;

step 208, data output: and outputting the model prediction result or the score to a database.

Further, in step 204, the data processing method adopts an abnormal value processing method, analyzes the image of the value, and modifies the value after finding the abnormal value.

Further, in step 206, a linear regression model is selected, which is expressed as:

y＝β₀+β₁x₁+β₂x₂+…+β_kx_k+ε

wherein y is the score of the degree of matching between the human posts, beta₀Is a regression constant, beta_1…kIs a partial regression coefficient, x_1…kFor the controller tag value, ε is the normal distribution with an error that follows a mean of 0.

Further, in step 207, whether the residual value of the model conforms to the normal distribution is observed by comparing the model with the artificial score. The residual expression is:

in the formula

Is the residual, y is the artificial score value,

and scoring the post matching in the linear regression model formula of the last step. If it is

The residual error is in accordance with the normal distribution with the average value of 0, which indicates that the human-sentry matching model is reasonable. If it is

If the residual error does not conform to the normal distribution, the coefficient beta is returned to the deviation_1…kAnd (4) carrying out manual adjustment until the residual value accords with the normal distribution with the average value of 0.

Further, the step 3 comprises:

step 301: removing the unlicensed staff and selecting all the staff needing to be scheduled;

step 302: selecting a date and a shift to be ranked;

step 303: selecting employees with high post matching degree score and low post matching degree score according to the post requirement to carry out capability average pairing, and selecting the paired employees to enter a step 304;

step 304: selecting the employee whose accumulated duration does not exceed the working duration limit in a specific time period by calculating the occurrence frequency of the employee in the specific time period, and entering step 305;

step 305: according to the post requirements, selecting employees with post continuous working time of a specific duration by calculating the time difference between the starting time of the current shift and the last shift, and entering step 306;

step 306: and 4, scheduling the selected staff, and returning to the step three to select the rest staff for scheduling. If the staff does not meet the conditions of the step 304 or the step 305, the staff returns to the step 303 to select the staff again for the rest steps until the time period required by the shift of the post is completely exhausted.

3. Advantageous effects

Compared with the prior art, the invention has the advantages that:

the intelligent scheduling system and the intelligent scheduling method are used for uniformly scheduling, so that the scheduling taking a team as a unit is broken, uniform coordination of resources is realized, and the utilization rate of a controller is improved on the whole; the matching degree of the controller at different seats is calculated by comprehensively considering six dimensions of command capability, safety record, control experience, health state, qualification and character characteristics through the controller figure, and the scientificity of scheduling is improved based on the matching degree of the personnel figure measuring and calculating personnel and the seats; by the aid of the preset limiting conditions, the system checks the scheduling limiting conditions and gives an alarm for the condition of violation of the limiting conditions, so that scheduling personnel are reminded intuitively, and the conditions of non-compliance and non-manageability are avoided; the efficiency of artifical scheduling is improved, the personnel's of having saved the scheduling time.

Drawings

FIG. 1 is a block diagram of the system of the present invention;

FIG. 2 is a schematic representation of the steps of the present invention;

FIG. 3 is a schematic diagram of a controller representation interface according to the present invention;

FIG. 4 is a schematic diagram of a human-job matching model according to the present invention;

FIG. 5 is a schematic diagram of the shift logic of the present invention;

FIG. 6 is a schematic view of a shift schedule interface according to the present invention.

Detailed Description

The invention is described in detail below with reference to the drawings and specific examples.

Example 1

As shown in FIG. 1, a system for scheduling controllers based on portrait images is mainly composed of a basic setting module, a controller portrait module and an automatic scheduling module.

A basic setting module: basic information related to scheduling of each unit is configured and checked in a configurable mode, and the basic information comprises a scheduling system (the scheduling mode of each unit defines the included sectors, the number of seats, the open time of the seats, the schedulable personnel and the classes of the personnel, effective time and ineffective time, and is convenient for a control unit to adjust according to actual conditions) and unavailable time maintenance (such as vacation, meeting, training and the like) of the personnel.

The basic setting module comprises the following sub-modules:

(1) a class maintenance submodule: the system can be maintained by manual addition and introduction.

(2) A new shift sub-module: the detailed contents for creating the shift system comprise shift system codes, shift system names, service sectors, number of people with shift, number of control seats, number of monitoring seats, number of coordination seats, night shift starting time, effective date, expiration date, shift system description and the like.

(3) Setting a seat opening time submodule: the number of seats is read according to the maintenance shift system, and the maintenance of the seat opening time is realized on the interface (night shift after the default 21: 00). The method can be maintained by dragging the gantt bar, and can also directly set the opening and closing time.

(4) Configuring a shift operator submodule: personnel available for maintenance shift and their shift categories can support mixed shift collocations and personal day-night shift preferences.

(5) Personnel unavailable time maintenance submodule: maintenance personnel unavailability time including vacation, meetings, training, etc. The problems that the link between the shift scheduling system and the intranet cannot be opened, the OA approval time is too long, or manual data updating is needed in other situations are solved.

The basic setting module adopts a modular design, so that the system can acquire data from a plurality of data sources, the relative independence can be realized, the modules can be independently designed, manufactured, debugged, modified and stored, and the intercommunication and interconnection among the modules are easy to realize.

Controller portrait module: the system is used for viewing personnel portraits of the controllers (essentially, various attributes of the controllers are labeled), and comprehensively knowing the conditions of the controllers. Besides basic information such as name, age, sex, political appearance, level, affiliated department, post and the like, the controller portrait comprises the following six dimensions: qualification dimension: including whether there is a license, whether the license is valid, the examination score of the license, whether there is a shift/inspection qualification, etc.; safe record dimension: the number of historical errors, the number of violations, the number of unsafe events, the number of accident signs, the number of accidents and the like are included; command capability dimension: the method comprises the highest single-hour control rack, the average single-hour command rack and the highest simultaneous command rack; fatigue degree dimension: including the working time of 1 night shift, the working time of nearly 7 days/15 days and the like; empirical dimension: the method comprises the steps of setting a single year, setting a single sector and setting a sector service year; character dimension: and (5) character characteristics obtained by the capability test. Based on the dimensions, the system can automatically calculate the matching degree of the controller to different seats, and provide basis for scheduling.

The controller representation module further comprises the following sub-modules:

(1) an initial interface submodule: for selecting the person to be viewed.

(2) The detailed interface sub-module of the controller portrait: for viewing controller representation details, including the degree of match to different seats, the value of each tag, and the details.

An automatic shift arrangement module: and determining seats required to be arranged according to the shift setting, determining optional personnel from the shift personnel, and pre-arranging according to the principle of force balance of all the working personnel at each time interval according to the limiting conditions and the personnel matching degree. The system supports both import and manual adjustment functions.

The matching degree of the controller at different seats is calculated by comprehensively considering six dimensions of command capability, safety record, control experience, health state, qualification and character feature through the controller figure, and the scientificity of scheduling is improved based on the matching degree of the personnel figure measuring and calculating personnel and the seats; through the preset limit conditions, the system checks the scheduling limit conditions and gives an alarm for the condition of violation of the limit, intuitively reminds scheduling personnel, and avoids the conditions of non-compliance and non-rationality

The automatic shift arrangement module further comprises:

(1) daily shift submodule: the system is used for realizing daily scheduling by clicking automatic scheduling, giving priority to strong limiting conditions during automatic scheduling, deleting non-conforming personnel, then giving priority to weak limiting conditions, giving priority to arranging personnel conforming to the limiting conditions, and arranging according to the capability balance principle. Three data of the controller name, the seat matching degree and the starting time and the ending time are displayed on each small Gantt bar. Or, the shift schedule plan can be imported according to the template, and the system can check the imported result and prompt the condition of violation of the limit. Version management may be performed (only one version of the daily schedule validation for each shift). The number of hours scheduled, the shift schedule for a single controller can be viewed at this module.

(2) A monthly shift submodule: the method is used for realizing monthly shift scheduling by clicking automatic shift scheduling. And (3) giving priority to strong limiting conditions and scheduled daily scheduling plans during automatic scheduling, deleting non-conforming personnel, considering weak limiting conditions, giving priority to personnel conforming to the limiting conditions, and performing automatic scheduling according to working time limitation and a scheduling mode. And displaying and checking the shift scheduling condition.

(3) Release class table submodule: the system is used for issuing daily scheduling and monthly scheduling in a mobile phone app or WeChat applet or picture storage mode.

The system for scheduling the controllers combines the characteristics of the air traffic control service, is easily accepted by an air control unit, can be manually and finely adjusted for the part of the input and output of the actual scheduling result, greatly improves the scheduling efficiency, and can also obtain scheduling information in time and plan the work and rest time in advance based on the big data and information network established in the field of the air traffic control.

Example 2

As shown in FIG. 2, the invention discloses a controller scheduling method based on a portrait, which specifically comprises the following steps:

step 1: a controller portrait is created.

The controller figure is used for labeling various attributes of the controller, so that the person figure of the controller can be checked, and the condition of the controller can be comprehensively known. Besides basic information such as name, age, gender, political appearance, level, affiliated department, post and the like, the controller portrait comprises six dimensions: qualification dimension, safety record dimension, command ability dimension, fatigue degree dimension, experience dimension and character dimension. Based on the dimensions, the system can automatically calculate the matching degree of the controller to different seats, and provide basis for scheduling. As shown in fig. 3, the details of the controller representation, including the degree of match to different seats, the value of each tag, and the details, can be viewed through a controller representation details interface.

Step 2: a personnel matching model is established through the controller portrait, the matching degree scores of the controller on different posts are calculated, support is provided for scheduling, and the establishment method of the personnel matching model is shown in figure 4.

Specifically, the method comprises the following steps:

step 201, selecting a sample: according to the types of the posts, different employees are selected as samples, a certain number of high, middle and low segments are selected, the matching degree scores of different posts are scored for the samples, and the scoring basis is given.

Step 202, label selection: and selecting a corresponding controller portrait label according to the sample scoring basis.

Step 203, acquiring data: and acquiring the selected label data.

Step 204, data cleaning: because the representation forms of the controller portrait labels are different, some are numerical type, some are non-numerical type, and the coverage rate of the labels is different, the basic data cleaning work is needed before modeling. Common data processing methods are: missing value filling, abnormal value processing and the like; because the security accuracy requirement of the control post is high and the label data volume is not large, the missing value filling adopts a manual filling method, and the filling method is more practical and can be accepted by the controller. Abnormal values are found through image analysis of the values, and the abnormal values are also manually modified as much as possible, so that the abnormal values are easily accepted by a controller.

Step 205, feature engineering conversion: before data enters a model, feature engineering conversion needs to be performed on existing data, such as: the labels "security awareness (data form high, medium, low)", which are non-numerical type labels, require conversion of label type to numerical type before modeling, and the common methods are to use unique thermal coding to convert the labels "security awareness to" whether high security awareness is present (0/1) "," whether medium security awareness is present (0/1) ", and" whether low security awareness is present (0/1) ".

Step 206, model building: and inputting the cleaned data subjected to the characteristic work conversion into a model, and predicting the matching degree scores of the controller and different posts. The embodiment selects a linear regression model, and the model has simple principle, strong theoretical basis and strong explanatory property (label weight is determined according to the model); the linear regression model is expressed in the form:

in the formula (I), the compound is shown in the specification,

scoring the degree of matching of the human hills, beta₀Is a regression constant, beta_1…kIs a partial regression coefficient, x_1…kFor the controller tag value, ε is the normal distribution with an error that follows a mean of 0.

Step 207, evaluation optimization: and (4) performing iterative optimization on the model according to the output result of the model, mainly focusing on the distribution of the residual error, so as to ensure that the model has predictability and interpretability. The method used here is to compare the model with the artificial score and observe whether the residual value of the model conforms to the normal distribution. The residual expression is:

in the formula

Is the residual, y is the artificial score value,

And step 3: and (3) formulating automatic scheduling logic, determining seats to be arranged according to the scheduling setting, determining optional personnel from the scheduling personnel, and pre-scheduling and adjusting according to the principle of force balance of all the working personnel at each time interval and the personnel matching degree according to the limiting conditions.

During the shift arrangement, the considered conditions are five in total: whether a post license exists; the accumulated working time can not exceed 10 hours within 24 hours; the radar control post cannot continuously work for more than 2 hours; the overall strength of controllers in different time periods is balanced; the overall strength of the control mat and the monitoring mat is balanced.

Taking 2 hours as one shift and 12 shifts in 24 hours all day as an example, as shown in fig. 5, the specific steps are as follows:

step 301: unlicensed staff is excluded and all staff requiring a shift is selected.

Step 302: the date and shift to be ranked is selected.

Step 303: and selecting the employees with high post matching degree score and low post matching degree score according to the post requirement to carry out capability average pairing. And selecting the matched staff to enter the next step.

Step 304: and selecting the employee with the accumulated time length within 24 hours not exceeding the 10 hour limit to enter the next step by calculating the occurrence frequency of the employee within 24 hours.

Step 305: and (4) according to the post requirements, selecting the staff with the radar post continuous working time not more than 2 hours by calculating the time difference between the starting time of the current shift and the last shift, and entering the step 6.

Step 306: and 4, scheduling the selected staff, and returning to the step three to select the rest staff for scheduling. And if the staff does not meet the conditions of the step 304 or the step 305, returning to the step 303 to reselect the staff for the rest steps until the time period required by the shift of the post is completely exhausted.

And 4, step 4: and outputting a shift arrangement result.

As shown in fig. 6, is a schedule interface diagram through which the scheduled hours, the schedule of a single controller, the daily schedule, and the restriction violation prompt list display may be viewed. The scheduling is realized by clicking the automatic scheduling, strong limiting conditions are preferably considered during the automatic scheduling, non-conforming personnel are deleted, then weak limiting conditions are considered, personnel conforming to the limiting conditions are preferably arranged, and the arrangement is carried out according to the capability balance principle.

Example 3

Based on the technical scheme of the invention, the shift scheduling verification process and effect of the embodiment are as follows.

The related data of the following three record tables are taken as the data input of the shift arrangement:

(1) employee historical work record table: the field information that the record needs to contain is as follows: employee name, start time, job post, etc. Data samples are as follows:

TABLE 1 working record table

(2) Staff basic information table: the fields to be included include the name of the employee, the employee's job number, whether the employee has a license, whether the accumulated working time in the last 24 hours is more than 10 hours, the last working seat, whether the employee continuously works for 2 hours, the matching degree with the shift, the matching degree with the control, the matching degree with the monitoring, the matching degree with the coordination, the qualification with the shift and the like. Data samples are as follows:

TABLE 2 basic information Table

(3) Scheduling schedule: the field information that needs to be included is the time period, number of stations, station name, start date, start time, etc. Data samples are as follows:

TABLE 3 watch for scheduling

Other data used are seat open time data:

TABLE 4 seat open time

Class code	Seat coding	Name of seat	Open time	Closing time
					HZ_APP_2020_03	WP1	Bed with executive function	8:00:00	1:59:59
HZ_APP_2020_03	WP2	Control seat	8:00:00	7:59:59
					HZ_APP_2020_03	WP3	Monitoring mat	8:00:00	7:59:59
HZ_APP_2020_03	WP4	Coordination mat	8:00:00	1:59:59
					HZ_APP_2020_03	WP5	Bed with executive function	6:00:00	21:59:59
HZ_APP_2020_03	WP6	Control seat	8:00:00	21:59:59
					HZ_APP_2020_03	WP7	Monitoring mat	8:00:00	21:59:59
HZ_APP_2020_03	WP8	Coordination mat	8:00:00	21:59:59
					HZ_APP_2020_03	WP9	Backup of	8:00:00	7:59:59

All the related data are input into the shift scheduling system. According to the method, the data output by the system can be arranged to obtain a shift table and the score of each time interval under the equilibrium principle, for example, 7 months and 1 day in 2020 as shown in table 5.

The scheduling result meets the scheduling limitation requirement, and the manpower resource of the controller is distributed from the aspect of force balance. The feasibility of the method is verified sufficiently in the scene. The method comprehensively considers that: the method comprises the steps of maintenance in a shift mode, adding a new shift mode, setting seat open time, configuring shift personnel, maintaining personnel unavailable time, and independently setting the name of a seat and the attributes of the seat, such as whether the seat is a radar control post or not and the longest operation time of a single person. The seat opening time is set by reading the number of each seat according to a maintenance system, and the maintenance of the seat opening time is default to 21: after 00 is night shift. The method can be maintained by dragging the gantt bar, and can also directly set the opening and closing time.

The object of the invention is positioned as a shift manager and a controller, and is also suitable for a manager.

TABLE 5 results of shift scheduling

For the scheduling personnel, the comprehensive ability and the ability change trend of the controller can be scientifically known by checking the figure image of the controller, so that the scheduling personnel can conveniently perform scientific team management. In addition, the system considers the corresponding restriction conditions of the scheduling, combines the matching degree of the controller and the post, and comprehensively considers six dimensions of command capability, safety record, control experience, qualification, fatigue state and character characteristics to automatically schedule the shifts, thereby realizing scientific scheduling. The system also supports manual shift scheduling, and can remind and record the condition of violation of the limit in the shift scheduling result.

For the controller, the comprehensive ability and the tendency of the change of the ability of the controller can be scientifically known by checking the figure image of the controller, so that the direction of promotion can be conveniently found. Meanwhile, the daily and monthly scheduling conditions of the user and the scheduling conditions of other personnel can be conveniently checked, and the reasonable time arrangement is facilitated.

The method can also configure and check basic information related to each unit shift in a configurable mode.

The invention and its embodiments have been described above schematically, without limitation, and the invention can be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The representation in the drawings is only one of the embodiments of the invention, the actual construction is not limited thereto, and any reference signs in the claims shall not limit the claims concerned. Therefore, if a person skilled in the art receives the teachings of the present invention, without inventive design, a similar structure and an embodiment to the above technical solution should be covered by the protection scope of the present patent. Furthermore, the word "comprising" does not exclude other elements or steps, and the word "a" or "an" preceding an element does not exclude the inclusion of a plurality of such elements. Several of the elements recited in the product claims may also be implemented by one element in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Claims

1. A controller scheduling system based on portrait, comprising:

the basic setting module is used for configuring basic information in a configurable mode;

the controller portrait module is used for establishing portrait of the controller based on the configuration information of the basic setting module for checking and selecting;

and the automatic scheduling module generates scheduling logic according to the configuration information of the basic setting module and the selection information of the controller portrait module and outputs a scheduling plan.

2. The system of claim 1, wherein the basic setup module comprises:

the class maintenance submodule is used for maintaining class;

the new shift subsystem is used for creating detailed contents of the shift;

setting a seat opening time sub-module to realize the maintenance of the seat opening time;

configuring a shift personnel submodule for maintaining available shift personnel and shift types;

and the unavailable time maintenance submodule is used for maintaining the unavailable time of the shift arrangement personnel.

3. A person representation based supervisor scheduling system as claimed in claim 2 wherein said supervisor representation module comprises:

the initial interface submodule is used for selecting the personnel needing to be checked;

and the detailed interface sub-module is used for checking the details of the controller portrait.

4. A person representation based supervisor scheduling system as claimed in claim 3 wherein said automatic scheduling module comprises:

the daily scheduling submodule is used for realizing automatic daily scheduling;

the monthly shift submodule is used for realizing automatic monthly shift;

and the release shift table submodule is used for releasing daily shift and monthly shift.

5. A person representation-based controller scheduling method applied to the person representation-based controller scheduling system according to any one of claims 1 to 4, comprising the following steps:

step 1: creating a controller portrait;

and 4, step 4: and generating and outputting a shift arrangement result.

6. The person representation-based controller scheduling method of claim 5, wherein the step 2 comprises:

step 203, acquiring data: acquiring selected label data;

7. The personnel scheduling method of claim 6, wherein in step 204, the data processing method uses an abnormal value processing method, and the image analysis of the value is performed to modify the abnormal value after the abnormal value is found.

8. The method as claimed in claim 7, wherein in step 206, a linear regression model is selected, the model is expressed as:

y＝β₀+β₁x₁+β₂x₂+…+β_kx_k+ε

wherein y is the score of the degree of matching between the human posts, beta₀Is a regression constant, beta_1…kIs a partial regression coefficient, x_1...kFor the controller label values, ε is the normal distribution with an error that follows a mean of 0.

9. The person portrait based controller scheduling method of claim 8, wherein in step 207, the residual value of the model is observed to be in accordance with a normal distribution by comparing the artificially scored value with the model, and the residual expression is as follows:

in the formula

Is the residual, y is the artificial score value,

scoring the post matching in the linear regression model formula of the last step; if it is

The residual error accords with the normal distribution with the average value of 0, which indicates that the human-sentry matching model is reasonable; if it is

If the residual error does not conform to the normal distribution, the partial regression coefficient beta is determined_1…kAnd (4) carrying out manual adjustment until the residual value accords with the normal distribution with the average value of 0.

10. The person representation-based controller scheduling method of claim 5, wherein the step 3 comprises:

step 302: selecting a date and a shift to be ranked;

step 304: selecting the employee with the accumulated duration not exceeding the working duration limit in the specific time period by calculating the occurrence frequency of the employee in the specific time period, and entering step 305;

step 305: according to the requirement of the post, selecting the employees with the post continuous working time with the specific duration by calculating the time difference between the starting time of the current shift and the last shift, and entering step 306;

step 306: the selected staff are arranged, and the step three is returned to select the rest staff for arrangement; if the staff does not meet the conditions of the step 304 or the step 305, the staff returns to the step 303 to select the staff again for the rest steps until the time period required by the shift of the post is completely exhausted.