CN117952564B - Scheduling simulation optimization method and system based on progress prediction - Google Patents
Scheduling simulation optimization method and system based on progress prediction Download PDFInfo
- Publication number
- CN117952564B CN117952564B CN202410332683.4A CN202410332683A CN117952564B CN 117952564 B CN117952564 B CN 117952564B CN 202410332683 A CN202410332683 A CN 202410332683A CN 117952564 B CN117952564 B CN 117952564B
- Authority
- CN
- China
- Prior art keywords
- time sequence
- workload
- data set
- characteristic
- equipment state
- 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
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000004088 simulation Methods 0.000 title claims abstract description 19
- 238000005457 optimization Methods 0.000 title claims abstract description 17
- 230000002159 abnormal effect Effects 0.000 claims abstract description 110
- 238000004458 analytical method Methods 0.000 claims abstract description 48
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 11
- 230000005856 abnormality Effects 0.000 claims description 12
- 238000012937 correction Methods 0.000 claims description 5
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims description 5
- 125000004122 cyclic group Chemical group 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000013439 planning Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 235000003642 hunger Nutrition 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 230000037351 starvation Effects 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
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
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- 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
- G06Q10/0633—Workflow analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a scheduling simulation optimization method and system based on progress prediction, and relates to the technical field of big data processing. The method comprises the following steps: collecting historical data, constructing a sample data set and training to obtain an anomaly analysis model; generating and processing a first reference characteristic of each abnormal data point, generating a third time sequence characteristic of each abnormal data point, carrying out data reconstruction on a workload time sequence data set in the time sequence data set, and generating a target workload time sequence data set; extracting a workload time sequence and a device state time sequence, and processing the workload time sequence and the device state time sequence through a prediction model to generate a first workload prediction result and a device state prediction result; and processing the equipment state prediction result through the anomaly analysis model, generating a second workload prediction result, and correcting the first workload prediction result to obtain a target workload prediction result. The invention realizes accurate prediction of workload and is convenient for enterprises to adjust work scheduling.
Description
Technical Field
The invention relates to the technical field of big data processing, in particular to a scheduling simulation optimization method and system based on progress prediction.
Background
Along with the development of economy, small-scale diversified project management gradually becomes a research topic for improving the self competitiveness, flexibility and sustainability of each project enterprise, and due to the fact that the mode needs the enterprise to have a higher project management level, reasonable planning of scheduling, reasonable supply of resource materials and the like, the enterprise is guaranteed to complete project delivery on time, and meanwhile market demands can be flexibly met.
In the process of implementing engineering projects, material supply and equipment stability of enterprises are key factors influencing the engineering progress of the enterprises, accurate prediction of engineering states and workload of the enterprises is one of effective means for improving the competitiveness of the enterprises, however, factors such as resource supply change, equipment faults and the like often interfere with prediction results, compared with a large-scale standard engineering mode, engineering data in a small-scale non-standard mode are more disordered, engineering progress is predicted by analyzing trends or stepwise workload means of historical data, and the accuracy of the prediction results is lower.
Disclosure of Invention
In order to overcome the defects, the invention provides a scheduling simulation optimization method and a scheduling simulation optimization system based on progress prediction.
As a first aspect of an embodiment of the present invention, there is provided a scheduling simulation optimization method based on progress prediction, including:
collecting historical data and constructing a time sequence data set, wherein the time sequence data set comprises an equipment state time sequence data set, a material supply time sequence data set and a workload time sequence data set;
determining a plurality of abnormal data points in the equipment state time sequence data set and the material supply time sequence data set, extracting a first time sequence characteristic of each abnormal data point and a second time sequence characteristic of an engineering project period to which the abnormal data point belongs, wherein the time sequence characteristic P= (e, m, y, t), e is the equipment state characteristic, m is the material supply characteristic, y is the workload characteristic, and t is the time characteristic;
Constructing a sample data set based on the first time sequence features and the second time sequence features, inputting the sample data set into an anomaly analysis model, and training to obtain the anomaly analysis model;
performing feature stitching on the first time sequence feature and the second time sequence feature of each abnormal data point to generate a first reference feature of each abnormal data point;
Processing the plurality of first reference features through an anomaly analysis model to generate a third time sequence feature of each anomaly data point, and carrying out data reconstruction on a workload time sequence data set in the time sequence data set based on the third time sequence feature of each anomaly data point to generate a target workload time sequence data set;
Respectively extracting a workload time sequence and an equipment state time sequence from a target workload time sequence data set and an equipment state time sequence data set;
Respectively processing the workload time sequence and the equipment state time sequence through a prediction model to generate a first workload prediction result and an equipment state prediction result in a time range to be predicted;
and generating a second reference characteristic based on the equipment state prediction result, inputting the second reference characteristic into the anomaly analysis model, generating a second workload prediction result, correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result, and adjusting the work schedule according to the target workload prediction result.
Preferably, feature stitching is performed on the first timing feature and the second timing feature of each abnormal data point, so as to generate a first reference feature of each abnormal data point, including:
a time signature of the first timing signature, an equipment status signature of the second timing signature, and a material supply signature are selected to construct a first reference signature of the outlier data point.
Preferably, the data reconstruction is performed on the workload time sequence data set in the time sequence data set based on the third time sequence characteristic of each abnormal data point, and the target workload time sequence data set is generated, which comprises the following steps:
and determining the workload data corresponding to each abnormal data point in the workload time sequence data set according to the time characteristics in the third time sequence characteristics of each abnormal data point, and replacing the workload data corresponding to each abnormal data point in the workload time sequence data set by the workload characteristics of each abnormal data point to obtain the target workload time sequence data set.
Preferably, generating a second reference feature based on the device state prediction result and inputting the second reference feature into the anomaly analysis model, generating a second workload prediction result includes:
determining a plurality of equipment state abnormal points in a time range to be predicted according to equipment state prediction results, obtaining equipment state characteristics of each equipment state abnormal point, determining time characteristics of the equipment state abnormal points according to the time of the equipment state abnormal points, determining material supply characteristics of the equipment state abnormal points according to material supply data, and generating second reference characteristics of each equipment state abnormal point;
And processing the second reference characteristic of each equipment state abnormal point through the abnormality analysis model to generate a predicted workload characteristic of each equipment state abnormal point, and obtaining a second workload prediction result.
Preferably, correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result includes:
And determining the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result, and replacing and updating the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result based on the predicted workload characteristics of each equipment state abnormal point to obtain a target workload predicted result.
Preferably, the anomaly analysis model is a recurrent neural network model.
Preferably, the predictive model is an ARIMA model.
As a second aspect of an embodiment of the present invention, there is provided a scheduling simulation optimizing system based on progress prediction, including:
the data processing module is used for collecting historical data and constructing a time sequence data set, wherein the time sequence data set comprises an equipment state time sequence data set, a material supply time sequence data set and a workload time sequence data set;
the characteristic extraction module is used for determining a plurality of abnormal data points in the equipment state time sequence data set and the material supply time sequence data set, extracting a first time sequence characteristic of each abnormal data point and a second time sequence characteristic of an engineering project period to which the abnormal data point belongs, wherein the time sequence characteristic P= (e, m, y, t), e is the equipment state characteristic, m is the material supply characteristic, y is the workload characteristic, and t is the time characteristic;
The model training module is used for constructing a sample data set based on the first time sequence features and the second time sequence features, inputting the sample data set into the anomaly analysis model, and training to obtain the anomaly analysis model;
The characteristic splicing module is used for carrying out characteristic splicing on the first time sequence characteristic and the second time sequence characteristic of each abnormal data point to generate a first reference characteristic of each abnormal data point;
The data reconstruction module is used for processing the plurality of first reference features through the anomaly analysis model, generating a third time sequence feature of each anomaly data point, and carrying out data reconstruction on a workload time sequence data set in the time sequence data set based on the third time sequence feature of each anomaly data point to generate a target workload time sequence data set;
The prediction module is used for respectively extracting a workload time sequence and a device state time sequence from the target workload time sequence data set and the device state time sequence data set, respectively processing the workload time sequence and the device state time sequence through the prediction model, and generating a first workload prediction result and a device state prediction result in a time range to be predicted;
and the prediction correction module is used for generating a second reference characteristic based on the equipment state prediction result, inputting the second reference characteristic into the abnormality analysis model, generating a second workload prediction result, and correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result.
Preferably, for the feature stitching module, feature stitching is performed on the first timing feature and the second timing feature of each abnormal data point, to generate a first reference feature of each abnormal data point, including:
a time signature of the first timing signature, an equipment status signature of the second timing signature, and a material supply signature are selected to construct a first reference signature of the outlier data point.
Preferably, for the prediction correction module, correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result, including:
And determining the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result, and replacing and updating the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result based on the predicted workload characteristics of each equipment state abnormal point to obtain a target workload predicted result.
Compared with the prior art, the invention has the following advantages:
According to the invention, time sequence characteristics are acquired from historical data, a sample data set is constructed and trained to obtain an abnormal analysis model, interference information in the historical data is removed through the abnormal analysis model, more accurate time sequence data of the working amount and time sequence data of the equipment state are generated, the engineering state and the working amount are predicted through the prediction model, the equipment state is further analyzed through the abnormal analysis model, the influence of the equipment state change on the working amount is determined, the prediction result of the prediction model is corrected, the prediction accuracy is improved, the resource supply requirement of enterprise planning engineering is facilitated, and the engineering scheduling is adjusted based on market requirements.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a scheduling simulation optimization method based on progress prediction according to an embodiment of the present invention;
FIG. 2 is a block diagram of a scheduling simulation optimization system based on progress prediction according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, some embodiments of the present application will be described in further detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application. However, those of ordinary skill in the art will understand that in various embodiments of the present application, numerous technical details have been set forth in order to provide a better understanding of the present application. The claimed application may be practiced without these specific details and with various changes and modifications based on the following embodiments.
Example 1: referring to fig. 1, a flowchart of an exemplary scheduling simulation optimization method based on progress prediction according to an embodiment of the present invention is shown, and the scheduling simulation optimization method based on progress prediction includes the following steps:
S101, collecting historical data, constructing a time sequence data set, determining a plurality of abnormal data points in the equipment state time sequence data set and the material supply time sequence data set, and extracting a first time sequence characteristic of each abnormal data point and a second time sequence characteristic of an engineering project period to which the abnormal data point belongs.
It should be noted that, the history data includes data in the engineering project process under different periods, such as equipment status data, material supply data, workload data, etc., and the time information of the combined data is used to construct a time sequence data set, where the time sequence data set includes an equipment status time sequence data set, a material supply time sequence data set and a workload time sequence data set, and is used to record data of different nodes, such as daily equipment status data, material supply data, workload data, etc.
In this embodiment, the abnormal data points are specifically nodes with equipment failure or nodes with insufficient material supply, and the time sequence features of the nodes with abnormality are extracted from the data set, including a first time sequence feature of each abnormal data point and a second time sequence feature of an engineering project period to which the abnormal data point belongs, where the time sequence feature p= (e, m, y, t), e is an equipment state feature, m is a material supply feature, y is a workload feature, and t is a time feature.
In this embodiment, in order to facilitate subsequent data processing, the device status feature may specifically be determined based on the duration of the project affected by the device fault in one day and the standard working duration in one day, and, for example, if the working is stopped in 0.5h due to the device fault in one day and only in 7.5h for normal operation in 8h for the standard working duration, the device status feature may be 93.75%. For the material supply characteristics, the determination may be based on the effect on the working process due to material starvation, e.g. the proportion of the time period of actual operation per day relative to the standard working time period. For the time feature, considering seasonal environmental impact during the engineering project, the engineering project period may be preset, and the quarter feature information may be extracted, for example, for the time feature of number 5 of 4 months, the engineering project period is 5 days, and the time feature is 2-1-1, which sequentially represents the first period of the first month of the second quarter (summer). Those skilled in the art may also quantitatively characterize the plurality of feature information in other ways, and the present embodiment is not limited specifically, and it should be noted that the engineering project period does not represent the actual working period of the engineering, but is used as a cyclic unit for analyzing the periodicity of the time series data.
For the second timing characteristic, the characteristic information is specifically used to characterize the remaining nodes without anomalies in the engineering project period to which the abnormal data point belongs, it is conceivable that the equipment status characteristic and the material supply characteristic may be both 100%, and the workload characteristic may be a workload average of a plurality of nodes without anomalies.
S102, a sample data set is built based on a plurality of first time sequence features and second time sequence features, the sample data set is input into an anomaly analysis model, and the anomaly analysis model is obtained through training.
In the present embodiment, the device state characteristics, the material supply characteristics, and the time characteristics of the plurality of time series characteristics in the sample data set are taken as inputs of the abnormality analysis model, and the workload characteristics of the plurality of time series characteristics are taken as training targets of the abnormality analysis model, and the abnormality analysis model for predicting the workload by analyzing the device state, the material supply data, and the time data is trained. The anomaly analysis model is a neural network model, has a function of analyzing time sequence data, and is constructed by taking a cyclic neural network as a model framework in the embodiment. Other models with time series data analysis function can be selected by those skilled in the art, and the abnormality analysis model is not particularly limited herein.
S103, performing feature stitching on the first time sequence feature and the second time sequence feature of each abnormal data point to generate a first reference feature of each abnormal data point, and processing the plurality of first reference features through an abnormal analysis model to generate a third time sequence feature of each abnormal data point.
In this embodiment, the stitching of the first reference feature includes selecting a time feature of the first time sequence feature, an equipment state feature of the second time sequence feature, and a material supply feature, constructing a first reference feature of an abnormal data point based on the features, processing the first reference feature through an abnormal analysis model, predicting workload data under the first reference feature, and adding the workload data to the first reference feature to generate a third time sequence feature corresponding to the abnormal data point.
S104, carrying out data reconstruction on a workload time sequence data set in the time sequence data set based on the third time sequence characteristic of each abnormal data point, generating a target workload time sequence data set, and respectively extracting a workload time sequence and an equipment state time sequence from the target workload time sequence data set and the equipment state time sequence data set.
In this embodiment, the meaning of performing data reconstruction on the workload time sequence data set is to reduce the influence of interference factors in the historical data, such as insufficient material supply, equipment production faults and the like, on the progress of the engineering project, so as to obtain data with greater correlation between the progress of the engineering project and time, namely, the target workload time sequence data set. Specifically, according to the time characteristics in the third time sequence characteristics of each abnormal data point, the corresponding workload data of each abnormal data point in the workload time sequence data set is determined, and the workload characteristics of each abnormal data point are used for replacing the workload data corresponding to each abnormal data point in the workload time sequence data set, so that the target workload time sequence data set is obtained.
And respectively analyzing the change relation of the workload and the equipment state along with time, and extracting to obtain a workload time sequence and an equipment state time sequence.
S105, respectively processing the workload time sequence and the equipment state time sequence through a prediction model to generate a first workload prediction result and an equipment state prediction result in a time range to be predicted.
In this embodiment, the prediction model is specifically an ARIMA model, and the ARIMA model may implement prediction of future data points by analyzing a time sequence, and for a time range to be predicted, for example, an engineering progress within a half month in the future, analyze a workload time sequence by the ARIMA model, predict workload data within a half month in the future, generate a first workload prediction result, analyze a device state time sequence, predict device state data within a half month in the future, and generate a device state prediction result. For the time frame to be predicted, the determination may be based on known material supply, for example, the known material supply may ensure the requirement of at least 10 days under the condition of normal operation of the engineering, and then the workload data of 10 days in the future may be predicted by the prediction model.
S106, generating a second reference characteristic based on the equipment state prediction result, inputting the second reference characteristic into the anomaly analysis model, generating a second workload prediction result, correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result, and adjusting the work schedule according to the target workload prediction result.
In this embodiment, for the first workload prediction result, the material supply data may be analyzed in combination with the actual situation, but the change of the equipment state has a certain unknowness, so the prediction accuracy is not high, and further analysis of the equipment state is required. Under the condition that the prediction time and the material supply are known, the equipment state is further analyzed by combining the equipment state prediction result and the abnormality analysis model, the influence of the equipment state change on the workload is determined, the first workload prediction result is corrected, a more accurate target workload prediction result is generated, the material supply requirement of an enterprise is conveniently planned, and meanwhile, the work schedule is adjusted.
Preferably, in the step S106, a second reference feature is generated based on the device state prediction result and is input into the anomaly analysis model, a second workload prediction result is generated, and the first workload prediction result is corrected based on the second workload prediction result, so as to obtain a target workload prediction result, which specifically includes:
and determining a plurality of equipment state abnormal points in the time range to be predicted according to the equipment state prediction result to obtain equipment state characteristics of each equipment state abnormal point.
Specifically, the abnormal point of the equipment state is a node with the equipment state characteristic lower than 100% in the equipment state prediction result, the time characteristic of the abnormal point of the equipment state is determined according to the time to which the abnormal point of the equipment state belongs, the material supply characteristic of the abnormal point of the equipment state is determined according to the material supply data, wherein after the known material supply data is analyzed to determine the time range to be predicted, the material supply characteristic is specifically 100%, and the second reference characteristic of each abnormal point of the equipment state is generated according to the related characteristic data.
And processing the second reference characteristic of each equipment state abnormal point through an abnormality analysis model to generate a second workload prediction result comprising the predicted workload characteristic of each equipment state abnormal point, wherein the second workload prediction result is specifically workload data caused by equipment abnormal change.
After the second workload prediction result is obtained, determining the corresponding prediction workload of each equipment state abnormal point in the first workload prediction result, namely, assuming theoretical workload under the condition that equipment failure does not occur in the equipment state abnormal point, replacing and updating the prediction workload corresponding to each equipment state abnormal point in the first workload prediction result based on the prediction workload characteristics of each equipment state abnormal point, and replacing the prediction workload corresponding to the equipment state abnormal point by a plurality of prediction workload characteristics respectively to obtain a more accurate target workload prediction result.
The scheduling simulation optimization method based on progress prediction is suitable for optimizing scheduling of enterprises based on small-scale non-standardized mode engineering projects, a prediction model for predicting the engineering work progress is constructed through comprehensively analyzing information such as material supply data, equipment state data and the like in the generation process, interference information in historical data is removed through an abnormal analysis model, prediction precision of the prediction model is improved, changing factors such as resources and environment in the historical data are considered, interference caused by the resource environment is reduced to the greatest extent, comprehensive analysis is performed on periodic characteristics in the data, accurate prediction of workload is achieved, the enterprise is convenient to analyze the engineering progress, a material supply plan which meets the enterprise requirements is formulated, work scheduling planning and the like are more scientific and reasonable, and flexibility of the enterprises is improved.
Example 2: referring to fig. 2, a block diagram of an exemplary scheduling simulation optimizing system based on progress prediction according to an embodiment of the present invention, specifically includes:
The data processing module is used for collecting historical data and constructing a time sequence data set;
Wherein the time series data set comprises an equipment state time series data set, a material supply time series data set and a workload time series data set;
The characteristic extraction module is used for determining a plurality of abnormal data points in the equipment state time sequence data set and the material supply time sequence data set, and extracting a first time sequence characteristic of each abnormal data point and a second time sequence characteristic of an engineering project period to which the abnormal data point belongs;
Wherein the time sequence feature p= (e, m, y, t), e is the equipment state feature, m is the material supply feature, y is the workload feature, and t is the time feature;
The model training module is used for constructing a sample data set based on the first time sequence features and the second time sequence features, inputting the sample data set into the anomaly analysis model, and training to obtain the anomaly analysis model;
the anomaly analysis model is a cyclic neural network model.
The characteristic splicing module is used for carrying out characteristic splicing on the first time sequence characteristic and the second time sequence characteristic of each abnormal data point to generate a first reference characteristic of each abnormal data point;
Specifically, a time characteristic of the first timing characteristic, an equipment status characteristic of the second timing characteristic, and a material supply characteristic are selected to construct a first reference characteristic of the outlier data point.
The data reconstruction module is used for processing the plurality of first reference features through the anomaly analysis model, generating a third time sequence feature of each anomaly data point, and carrying out data reconstruction on a workload time sequence data set in the time sequence data set based on the third time sequence feature of each anomaly data point to generate a target workload time sequence data set;
specifically, correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result, including:
And determining the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result, and replacing and updating the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result based on the predicted workload characteristics of each equipment state abnormal point to obtain a target workload predicted result.
The prediction module is used for respectively extracting a workload time sequence and a device state time sequence from the target workload time sequence data set and the device state time sequence data set, respectively processing the workload time sequence and the device state time sequence through the prediction model, and generating a first workload prediction result and a device state prediction result in a time range to be predicted;
and the prediction correction module is used for generating a second reference characteristic based on the equipment state prediction result, inputting the second reference characteristic into the abnormality analysis model, generating a second workload prediction result, and correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result.
It will be understood that modifications and variations will be apparent to those skilled in the art from the foregoing description, and it is intended that all such modifications and variations be included within the scope of the following claims. Parts of the specification not described in detail belong to the prior art known to those skilled in the art.
Claims (6)
1. The scheduling simulation optimization method based on progress prediction is characterized by comprising the following steps of:
collecting historical data and constructing a time sequence data set, wherein the time sequence data set comprises an equipment state time sequence data set, a material supply time sequence data set and a workload time sequence data set;
determining a plurality of abnormal data points in the equipment state time sequence data set and the material supply time sequence data set, extracting a first time sequence characteristic of each abnormal data point and a second time sequence characteristic of an engineering project period to which the abnormal data point belongs, wherein the time sequence characteristic P= (e, m, y, t), e is the equipment state characteristic, m is the material supply characteristic, y is the workload characteristic, and t is the time characteristic;
Constructing a sample data set based on a plurality of first time sequence features and second time sequence features, inputting the sample data set into an anomaly analysis model, and training to obtain the anomaly analysis model, wherein the anomaly analysis model is a cyclic neural network model;
performing feature stitching on the first time sequence feature and the second time sequence feature of each abnormal data point to generate a first reference feature of each abnormal data point; the method comprises the following steps:
selecting a time feature of the first time sequence feature, a device state feature of the second time sequence feature and a material supply feature to construct a first reference feature of an abnormal data point;
Processing the plurality of first reference features through an anomaly analysis model to generate a third time sequence feature of each anomaly data point, and carrying out data reconstruction on a workload time sequence data set in the time sequence data set based on the third time sequence feature of each anomaly data point to generate a target workload time sequence data set; the method comprises the following steps:
determining the workload data corresponding to each abnormal data point in the workload time sequence data set according to the time characteristics in the third time sequence characteristics of each abnormal data point, and replacing the workload data corresponding to each abnormal data point in the workload time sequence data set by the workload characteristics of each abnormal data point to obtain a target workload time sequence data set;
Respectively extracting a workload time sequence and an equipment state time sequence from a target workload time sequence data set and an equipment state time sequence data set;
Respectively processing the workload time sequence and the equipment state time sequence through a prediction model to generate a first workload prediction result and an equipment state prediction result in a time range to be predicted, wherein the prediction model is an ARIMA model;
and generating a second reference characteristic based on the equipment state prediction result, inputting the second reference characteristic into the anomaly analysis model, generating a second workload prediction result, correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result, and adjusting the work schedule according to the target workload prediction result.
2. The scheduling simulation optimization method based on progress prediction according to claim 1, wherein generating a second reference feature based on the device state prediction result and inputting the second reference feature into the anomaly analysis model, generating a second workload prediction result comprises:
determining a plurality of equipment state abnormal points in a time range to be predicted according to equipment state prediction results, obtaining equipment state characteristics of each equipment state abnormal point, determining time characteristics of the equipment state abnormal points according to the time of the equipment state abnormal points, determining material supply characteristics of the equipment state abnormal points according to material supply data, and generating second reference characteristics of each equipment state abnormal point;
And processing the second reference characteristic of each equipment state abnormal point through the abnormality analysis model to generate a predicted workload characteristic of each equipment state abnormal point, and obtaining a second workload prediction result.
3. The scheduling simulation optimization method based on progress prediction according to claim 2, wherein correcting the first workload prediction result based on the second workload prediction result to obtain the target workload prediction result comprises:
And determining the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result, and replacing and updating the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result based on the predicted workload characteristics of each equipment state abnormal point to obtain a target workload predicted result.
4. A scheduling simulation optimizing system based on progress prediction, characterized in that a scheduling simulation optimizing method based on progress prediction as claimed in any one of claims 1 to 3 is used, comprising:
the data processing module is used for collecting historical data and constructing a time sequence data set, wherein the time sequence data set comprises an equipment state time sequence data set, a material supply time sequence data set and a workload time sequence data set;
the characteristic extraction module is used for determining a plurality of abnormal data points in the equipment state time sequence data set and the material supply time sequence data set, extracting a first time sequence characteristic of each abnormal data point and a second time sequence characteristic of an engineering project period to which the abnormal data point belongs, wherein the time sequence characteristic P= (e, m, y, t), e is the equipment state characteristic, m is the material supply characteristic, y is the workload characteristic, and t is the time characteristic;
The model training module is used for constructing a sample data set based on the first time sequence features and the second time sequence features, inputting the sample data set into the anomaly analysis model, and training to obtain the anomaly analysis model;
The characteristic splicing module is used for carrying out characteristic splicing on the first time sequence characteristic and the second time sequence characteristic of each abnormal data point to generate a first reference characteristic of each abnormal data point;
The data reconstruction module is used for processing the plurality of first reference features through the anomaly analysis model, generating a third time sequence feature of each anomaly data point, and carrying out data reconstruction on a workload time sequence data set in the time sequence data set based on the third time sequence feature of each anomaly data point to generate a target workload time sequence data set;
The prediction module is used for respectively extracting a workload time sequence and a device state time sequence from the target workload time sequence data set and the device state time sequence data set, respectively processing the workload time sequence and the device state time sequence through the prediction model, and generating a first workload prediction result and a device state prediction result in a time range to be predicted;
and the prediction correction module is used for generating a second reference characteristic based on the equipment state prediction result, inputting the second reference characteristic into the abnormality analysis model, generating a second workload prediction result, and correcting the first workload prediction result based on the second workload prediction result to obtain a target workload prediction result.
5. The schedule modeling optimization system based on progress prediction of claim 3, wherein for the feature stitching module, feature stitching the first time series feature and the second time series feature for each outlier data point to generate a first reference feature for each outlier data point comprises:
a time signature of the first timing signature, an equipment status signature of the second timing signature, and a material supply signature are selected to construct a first reference signature of the outlier data point.
6. The schedule modeling optimization system based on progress prediction of claim 3, wherein for the prediction correction module, correcting the first workload prediction result based on the second workload prediction result to obtain the target workload prediction result comprises:
And determining the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result, and replacing and updating the predicted workload corresponding to each equipment state abnormal point in the first workload predicted result based on the predicted workload characteristics of each equipment state abnormal point to obtain a target workload predicted result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410332683.4A CN117952564B (en) | 2024-03-22 | 2024-03-22 | Scheduling simulation optimization method and system based on progress prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410332683.4A CN117952564B (en) | 2024-03-22 | 2024-03-22 | Scheduling simulation optimization method and system based on progress prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117952564A CN117952564A (en) | 2024-04-30 |
| CN117952564B true CN117952564B (en) | 2024-06-07 |
Family
ID=90803089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410332683.4A Active CN117952564B (en) | 2024-03-22 | 2024-03-22 | Scheduling simulation optimization method and system based on progress prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117952564B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2020071845A (en) * | 2018-11-02 | 2020-05-07 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | Abnormality detection device, abnormality detection method, and abnormality detection program |
| WO2021212752A1 (en) * | 2020-04-23 | 2021-10-28 | 平安科技（深圳）有限公司 | Device index data-based anomaly detection method and apparatus, device, and storage medium |
| CN113988268A (en) * | 2021-11-03 | 2022-01-28 | 西安交通大学 | Heterogeneous multi-source time sequence anomaly detection method based on unsupervised full-attribute graph |
| CN115936428A (en) * | 2022-11-17 | 2023-04-07 | 江苏东港能源投资有限公司 | External damage prevention fixed value optimization system for incremental power distribution network |
| WO2024000852A1 (en) * | 2022-06-29 | 2024-01-04 | 深圳前海微众银行股份有限公司 | Data processing method and apparatus, device, and storage medium |
| CN117436763A (en) * | 2023-11-30 | 2024-01-23 | 广州墨斗信息科技有限公司 | Method and system for realizing complete process fine management of building labor based on digitization |
| CN117473514A (en) * | 2023-12-28 | 2024-01-30 | 华东交通大学 | Intelligent operation and maintenance method and system of industrial control system |
| CN117494292A (en) * | 2024-01-02 | 2024-02-02 | 中亿丰数字科技集团有限公司 | Engineering progress management method and system based on BIM and AI large model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115208680B (en) * | 2022-07-21 | 2023-04-07 | 中国科学院大学 | Dynamic network risk prediction method based on graph neural network |
-
2024
- 2024-03-22 CN CN202410332683.4A patent/CN117952564B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2020071845A (en) * | 2018-11-02 | 2020-05-07 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | Abnormality detection device, abnormality detection method, and abnormality detection program |
| WO2021212752A1 (en) * | 2020-04-23 | 2021-10-28 | 平安科技（深圳）有限公司 | Device index data-based anomaly detection method and apparatus, device, and storage medium |
| CN113988268A (en) * | 2021-11-03 | 2022-01-28 | 西安交通大学 | Heterogeneous multi-source time sequence anomaly detection method based on unsupervised full-attribute graph |
| WO2024000852A1 (en) * | 2022-06-29 | 2024-01-04 | 深圳前海微众银行股份有限公司 | Data processing method and apparatus, device, and storage medium |
| CN115936428A (en) * | 2022-11-17 | 2023-04-07 | 江苏东港能源投资有限公司 | External damage prevention fixed value optimization system for incremental power distribution network |
| CN117436763A (en) * | 2023-11-30 | 2024-01-23 | 广州墨斗信息科技有限公司 | Method and system for realizing complete process fine management of building labor based on digitization |
| CN117473514A (en) * | 2023-12-28 | 2024-01-30 | 华东交通大学 | Intelligent operation and maintenance method and system of industrial control system |
| CN117494292A (en) * | 2024-01-02 | 2024-02-02 | 中亿丰数字科技集团有限公司 | Engineering progress management method and system based on BIM and AI large model |
Non-Patent Citations (2)
| Title |
|---|
| 基于模型的热轧轧制计划排程系统的设计与实现;黄可为;杜斌;刘青;牛长胜;;上海金属;20090715(04);全文 * |
| 面向LED封装车间管理的MES排程模块研究与设计;李晨晨;臧海娟;吴涛;;电脑知识与技术;20200305(07);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117952564A (en) | 2024-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12112271B2 (en) | Resource data modeling, forecasting and simulation | |
| Yang et al. | Robust scheduling based on extreme learning machine for bi-objective flexible job-shop problems with machine breakdowns | |
| Singh et al. | Entropy based software reliability analysis of multi-version open source software | |
| US20060106477A1 (en) | Production plan creation system, method, and program | |
| CN101408769B (en) | On-line energy forecasting system and method based on product ARIMA model | |
| Zhang et al. | Investigating the impact of operational variables on manufacturing cost by simulation optimization | |
| Ungermann et al. | Data analytics for manufacturing systems–a data-driven approach for process optimization | |
| CN111754024A (en) | Power industry time sequence prediction method and device based on regression analysis | |
| CN110991985A (en) | Project progress data processing method | |
| JP2017151980A (en) | System and method for optimization of recommended service intervals | |
| Kivanç et al. | A scalable multi-objective maintenance optimization model for systems with multiple heterogeneous components and a finite lifespan | |
| CN117952564B (en) | Scheduling simulation optimization method and system based on progress prediction | |
| Friederich et al. | A Framework for Validating Data-Driven Discrete-Event Simulation Models of Cyber-Physical Production Systems | |
| Francis et al. | Simulation of scheduling logic using dynamic functions | |
| Siriweera et al. | Constraint-driven dynamic workflow for automation of big data analytics based on graphplan | |
| Barton | Simulation experiment design | |
| Briano et al. | Design of experiment and montecarlo simulation as support for gas turbine power plant availabilty estimation | |
| CN113780655A (en) | Steel multi-variety demand prediction method based on intelligent supply chain | |
| Szaller et al. | Real-time prediction of manufacturing lead times in complex production environments | |
| CN117371662A (en) | Evaluation system and method for adjustment capability of virtual power plant | |
| CN118863148A (en) | Scheduling prediction method, device, equipment and medium for flexible manufacturing unit | |
| CN117728404A (en) | Method and system for predicting power generation capacity of medium-and-long-term photovoltaic field station | |
| Vouligny et al. | Asset Management in Electrical Utilities in the Context of Business and Operational Complexity | |
| Frey et al. | Tackling the multitude of uncertainties in energy systems analysis by model coupling and high-performance computing | |
| Wai et al. | Application of Discrete Event Simulation in Modelling a SME Manufacturing System |
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 |