Dynamic progress prediction method and device based on key line, equipment and medium

Abstract

The invention discloses a dynamic progress prediction method and device based on a key line, equipment and a medium. The method comprises the following steps: constructing an acquisition progress network based on pre-acquired process data acquisition data, and constructing a handover progress network based on pre-acquired handover technical data; according to the handing-over progress network and the acquisition progress network, respectively obtaining a handing-over progress diagram and an acquisition progress diagram; constructing a mapping relation model of the handing-over progress chart and the acquisition progress chart based on the handing-over progress chart and the acquisition progress chart, and obtaining correction parameters of the handing-over progress chart and the acquisition progress chart; based on the correction parameters, correcting the collection progress diagram to obtain a process progress diagram; inputting the process progress diagram into the trained construction progress prediction model to obtain a predicted construction progress diagram; and inputting the predicted construction progress graph, the process progress graph and a pre-constructed construction plan graph into a pre-trained adjustment plan network to obtain a feature graph containing the predicted key path. The method improves the efficiency and accuracy of engineering project management.

Description

Technical Field

[0001] The present application relates to a method, device, equipment and medium for dynamically predicting progress based on a critical path. Background Art

[0002] Engineering projects are comprehensive social activities that integrate economy, technology, management and organization. During the implementation process, they are easily affected by various risk factors such as funds, material supply and emergencies during construction. Therefore, risk management is an indispensable part of modern engineering project management, and risk warning is an important means of project management, which can measure the degree of deviation of a certain state from the warning line and remind people to take corresponding preventive measures. In actual projects, due to the uncertainty of the duration of various tasks and the complexity of the project itself, the deviation of engineering construction forecasts is large and the accuracy is low, which in turn makes the preparation of the construction schedule a major "bottleneck" in engineering progress control. Summary of the invention

[0003] In order to better realize the construction progress prediction in project management, the embodiments of the present application provide a method, device, equipment and medium for dynamically predicting progress based on key paths.

[0004] In a first aspect, an embodiment of the present application provides a method for dynamically predicting progress based on a critical path, the method comprising:

[0005] Construct a collection progress network based on the pre-acquired process data collection information, and construct a handover progress network based on the pre-acquired handover technical information;

[0006] According to the delivery progress network and the collection progress network, a delivery progress chart and a collection progress chart are obtained respectively;

[0007] Based on the handover progress chart and the collection progress chart, a mapping relationship model between the handover progress chart and the collection progress chart is constructed to obtain correction parameters of the handover progress chart and the collection progress chart;

[0008] Based on the correction parameters, the acquisition progress chart is corrected to obtain a process progress chart;

[0009] Inputting the process progress diagram into a trained construction progress prediction model to obtain a predicted construction progress diagram;

[0010] The predicted construction progress diagram, the process progress diagram and the pre-built construction plan diagram are input into a pre-trained adjustment plan network to obtain a feature diagram containing a predicted critical path.

[0011] In an optional implementation of the embodiment of the present application, the adjustment plan network includes a first adjustment structure, a second adjustment structure and a third adjustment structure; the predicted construction progress map, the process progress map and the pre-built construction plan map are input into the pre-trained adjustment plan network to obtain a feature map containing a predicted critical path, including:

[0012] Inputting the construction plan diagram into a third adjustment structure, and extracting features of the construction plan diagram through a third GCN network;

[0013] Predicting the features of the construction plan diagram through a third LSTM network to obtain a third preliminary prediction result;

[0014] Predicting the preliminary prediction result through a third RCN network to obtain a third prediction result;

[0015] Inputting the predicted construction progress graph into a first adjustment structure, and extracting features of the predicted construction progress graph through a first GCN network;

[0016] The features of the construction plan diagram are integrated with the features of the predicted construction progress diagram, and a first preliminary prediction result is obtained through a first LSTM network;

[0017] The third preliminary prediction result is integrated with the first preliminary prediction result to obtain a first prediction result through a first RCN network;

[0018] Inputting the process progress graph into a second adjustment structure, and extracting features of the process progress graph through a second GCN network;

[0019] The features of the construction plan diagram, the features of the predicted construction progress diagram, and the features of the process progress diagram are integrated to obtain a second preliminary prediction result through a second LSTM network;

[0020] The third preliminary prediction result, the first preliminary prediction result and the second preliminary prediction result are fused through a second RCN network to obtain a feature graph containing a prediction key path.

[0021] In an optional implementation of the embodiment of the present application, obtaining a handover progress chart and a collection progress chart respectively according to the handover progress network and the collection progress network includes:

[0022] An empty image is constructed according to the handover progress network, and the value of each node in the handover progress network is used as the pixel value of the pixel point at the corresponding position in the empty image to obtain a handover progress map;

[0023] An empty image is constructed according to the acquisition progress network, and the value of each node in the acquisition progress network is used as the pixel value of the pixel point at the corresponding position in the empty image to obtain an acquisition progress map.

[0024] In an optional implementation of the embodiment of the present application, the mapping relationship model between the handover progress chart and the collection progress chart is constructed based on the handover progress chart and the collection progress chart to obtain correction parameters of the handover progress chart and the collection progress chart, including:

[0025] Sampling is performed in the handover progress chart and the collection progress chart respectively to obtain a preset number of progress matching data pairs;

[0026] Based on the preset number of progress matching data pairs, a mapping matrix is ​​obtained by using the mapping relationship model constructed by the following formula 1:

[0027] C = HJ formula 1;

[0028] Wherein, H is the mapping matrix; C is the acquisition progress chart; J is the handover progress chart;

[0029] Based on the handover progress chart, the collection progress chart and the mapping matrix, correction parameters of the handover progress chart and the collection progress chart are obtained.

[0030] In an optional implementation of the embodiment of the present application, the acquisition progress chart is corrected based on the correction parameter to obtain a process progress chart, including:

[0031] Based on the correction parameter, the acquisition progress graph is corrected by the following formula 2 to obtain the process progress graph:

[0032] G=CH -1 B Formula 2;

[0033] Wherein, G is the process progress chart; B is the correction parameter; H -1 is the inverse matrix of the mapping matrix; C is the acquisition progress chart.

[0034] In an optional implementation of the embodiment of the present application, the trained construction progress prediction model is obtained by the following method:

[0035] Based on the LSTM network, a construction progress prediction model is built;

[0036] Based on a plurality of historical construction progress diagrams obtained in advance, the construction progress prediction model is trained until the loss function of the LSTM network converges, thereby obtaining a trained construction progress prediction model.

[0037] In an optional implementation of the embodiment of the present application, it is characterized by further comprising:

[0038] Comparing the characteristics of the feature map with the characteristics of the construction plan map, determining whether the difference between the construction period of each feature point in the feature map and the construction period of the corresponding feature point in the characteristics of the construction plan map is greater than a preset threshold;

[0039] If so, a progress warning is issued for the items corresponding to the feature points.

[0040] In an optional implementation of the embodiment of the present application, it is characterized by further comprising:

[0041] Before inputting the process progress chart into the trained construction progress prediction model, the process progress data in the process progress chart is cleaned to obtain a cleaned process progress chart.

[0042] In a second aspect, an embodiment of the present application provides a dynamic progress prediction device based on a critical path, the device comprising:

[0043] The first construction module is used to construct a collection progress network based on the pre-acquired process data collection information, and to construct a handover progress network based on the pre-acquired handover technical information;

[0044] The second construction module is used to obtain a handover progress chart and a collection progress chart respectively according to the handover progress network and the collection progress network;

[0045] A third construction module is used to construct a mapping relationship model between the handover progress diagram and the collection progress diagram based on the handover progress diagram and the collection progress diagram, and obtain correction parameters of the handover progress diagram and the collection progress diagram;

[0046] A correction module, used for correcting the acquisition progress chart based on the correction parameters to obtain a process progress chart;

[0047] A first prediction module is used to input the process progress diagram into a trained construction progress prediction model to obtain a predicted construction progress diagram;

[0048] The second prediction module is used to input the predicted construction progress diagram, the process progress diagram and the pre-built construction plan diagram into a pre-trained adjustment plan network to obtain a feature diagram containing a predicted critical path.

[0049] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for dynamically predicting progress based on critical paths.

[0050] In a fourth aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the dynamic progress prediction method based on the critical path as described above is implemented.

[0051] In a fifth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer device, enables the computer device to execute the above-mentioned method for dynamically predicting progress based on critical paths.

[0052] In a sixth aspect, an embodiment of the present application provides a chip, the chip including a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the dynamic progress prediction method based on the critical path as described above.

[0053] The beneficial effects of the above technical solution provided by the embodiment of the present application include at least:

[0054] The dynamic prediction progress method based on the critical path provided by the embodiment of the present application obtains the collection progress chart and the handover progress chart respectively through the pre-acquired process data collection data and the handover technical data, and then obtains the process progress chart, and then inputs the process progress chart into the trained construction progress prediction model to obtain the predicted construction progress chart, and finally, inputs the predicted construction progress chart, the process progress chart and the pre-built construction plan chart into the pre-trained adjustment plan network to obtain the feature chart containing the predicted critical path. The method predicts the critical path information through the pre-trained adjustment plan network, can realize automatic monitoring and planning of the construction progress, and derives a reliable construction route, which provides a strong support for the construction of the project, helps the project to achieve efficient and reliable management, and improves the efficiency and accuracy of the project management; based on the process progress data acquired in real time, the dynamic prediction of the critical path and the construction status analysis can be carried out, which can more accurately predict the project progress, and the construction can be adjusted in real time through the results of the dynamic prediction of the critical path, which can better control the project progress and avoid project delays.

[0055] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or be understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the written description, claims, and drawings.

[0056] The technical solution of the present application is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings are used to provide a further understanding of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings:

[0058] Figure 1 A schematic diagram of the steps of a method for dynamically predicting progress based on key paths provided in an embodiment of the present application;

[0059] Figure 2 An example image containing multiple pixel coordinates provided in an embodiment of the present application;

[0060] Figure 3 A schematic diagram of the structure of the adjustment plan network provided in the embodiment of the present application;

[0061] Figure 4 A schematic diagram of the overall framework of a dynamic progress prediction method based on key paths provided in an embodiment of the present application;

[0062] Figure 5 A schematic diagram of the structure of a dynamic progress prediction device based on key paths provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0064] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0065] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0066] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0067] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0068] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0069] It should be understood that the size of the serial numbers of the steps in the following embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0070] In order to illustrate the technical solution of the present application, a specific embodiment is provided below for illustration.

[0071] The inventor found that in the prior art, the traditional project management method mainly controls the project progress through a static schedule, but due to the existence of various uncertainties in the project execution process, it is difficult to ensure the accuracy and timeliness of the schedule, which does not meet the inventor's expectations. Based on this, the inventor further developed and proposed this application to provide a method, device, equipment and medium for dynamic prediction of progress based on critical paths.

[0072] Embodiment 1

[0073] The present application embodiment provides a method for dynamically predicting progress based on a key path, referring to Figure 1 As shown, the method includes:

[0074] S101: constructing a collection progress network based on the pre-acquired process data collection information, and constructing a handover progress network based on the pre-acquired handover technical information.

[0075] S102: Obtaining a handover progress chart and a collection progress chart respectively according to the handover progress network and the collection progress network.

[0076] S103: Based on the handover progress chart and the collection progress chart, a mapping relationship model between the handover progress chart and the collection progress chart is constructed to obtain correction parameters of the handover progress chart and the collection progress chart.

[0077] S104: Based on the correction parameters, the acquisition progress chart is corrected to obtain a process progress chart.

[0078] S105: Input the process progress chart into the trained construction progress prediction model to obtain a predicted construction progress chart.

[0079] S106: Input the predicted construction progress diagram, the process progress diagram and the pre-built construction plan diagram into a pre-trained adjustment plan network to obtain a feature diagram containing a predicted critical path.

[0080] Before the implementation of the project, it is necessary to manually carry out the progress planning work, including unit project division, WBS planning and planning at all levels. The specific configuration can be done in the following ways:

[0081] Before the implementation of the project, the construction process of the entire project can be decomposed, the unit project can be divided, and the responsible units can be associated. The minimum granularity can be identified to the sub-project, and a division of labor database can be built (that is, the unit project division and WBS planning of the project can be carried out). The division of labor database contains multiple task decomposition data, and each task decomposition data includes a link consisting of "project → task → work → daily activities".

[0082] According to the specified construction period, the various levels of items in "project → task → work → daily activities" are planned, that is, plans are compiled at all levels, and time is allocated to each level of items, that is, a pre-built construction plan is obtained. According to the compiled plan, the process standardization decomposition is carried out for the processes in daily activities, that is, the complex production process is decomposed into a series of standardized and operable processes, so that each process has clear input, output, resource requirements and time requirements. This work package standard process decomposition method helps to improve production efficiency, reduce errors, simplify quality control, and make the production process easier to manage and optimize.

[0083] According to the importance of the process, each process is assigned a weight. The larger the weight, the greater the impact of the process on the implementation of the entire project. After the weight is configured, a construction network will be obtained. The construction network includes "project → task → work → daily activities", which includes the time and weight of each process and the construction of the process. Among them, the construction method of the construction network can be constructed according to the "project → task → work → daily activities" tree. For each node in the tree that is related, the related nodes are also connected to form a construction network. There may be some closed loops in the construction network, and the corresponding matters between the nodes connected by the closed loops affect and relate to each other.

[0084] After the construction network is set up in the above way, the project enters the implementation stage. During the construction of the project, the construction personnel who carry out daily activities will regularly (once a day, once a week, or once a month) report the work progress and workload, and record them in the technical data for handover, but there may be errors in the reporting. Therefore, the supervision unit should also regularly check the actual work situation and record it in the process data collection data, but the collection method may also have some errors. Therefore, it is necessary to combine the data of the handover technical data and the process data collection data to obtain accurate process progress data.

[0085] In a specific embodiment, the technical data submitted by the Guangyuan City Petroleum Construction Project states:

[0086] On August 20, 2023, the Guangyuan Petroleum Construction Project (project), pipeline tasks (tasks, tasks include pipelines, booster stations and gas stations), laying work items (work, work includes pipeline line design, line drawing, trenching, laying and landfilling, etc.), transported 1,000 meters of pipelines (daily activities).

[0087] The process data collection materials of Guangyuan City Petroleum Construction Project record:

[0088] On August 20, 2023, the Guangyuan Petroleum Construction Project (project), pipeline tasks (tasks, tasks include pipelines, booster stations and gas stations), laying work items (work, work includes pipeline line design, line drawing, trenching, laying and landfilling, etc.), transported 900 meters of pipelines (daily activities).

[0089] It can be seen from the contents recorded in the above-mentioned technical data for handover and process data collection data that there are errors in the process data collection data and the technical data for handover, which need to be corrected.

[0090] The error can be corrected manually. However, manual correction is inconvenient, and may be untrue and unfair. On the other hand, there may be conflicts in confrontation. In addition, sometimes the error may be relatively small and have no impact on the entire project, but many systems currently alarm because of small errors, resulting in a high false alarm rate in the system. Therefore, the inventor believes that a relatively fair, simple and effective correction method is very helpful for the entire project management. The purpose of correcting the error can be achieved by combining the process data collection information with the handover technical information, and correcting the process data collection information through the handover technical information to obtain the process progress data.

[0091] In the above step S101, the collection progress network is constructed based on the pre-acquired process data collection information, and the handover progress network is constructed based on the pre-acquired handover technical information. Specifically, according to the process data collection information and the handover technical information, in the manner of "project→task→work→daily activities", the progress networks of the process data collection information and the handover technical information are respectively constructed, namely, the collection progress network and the handover progress network. The size of the collection progress network and the handover progress network can be the same as the above-mentioned construction network, or smaller than the construction network.

[0092] In the above step S102, obtaining a handover progress chart and a collection progress chart respectively according to the handover progress network and the collection progress network specifically includes:

[0093] An empty image is constructed according to the handover progress network, and the value of each node in the handover progress network is used as the pixel value of the pixel point at the corresponding position in the empty image to obtain a handover progress map;

[0094] An empty image is constructed according to the acquisition progress network, and the value of each node in the acquisition progress network is used as the pixel value of the pixel point at the corresponding position in the empty image to obtain an acquisition progress map.

[0095] In the embodiment of the present application, the handover progress map and the collection progress map can be similar to the format of the image, and the nodes of the handover progress network and the collection progress network can be matched with the pixel points of the image to obtain the handover progress map and the collection progress map. Therefore, the handover progress map and the collection progress map contain nodes, and in addition to the item information (corresponding to the position coordinates in the image) marked in the nodes, they also contain the completion progress data of the item information during the implementation process of the engineer (corresponding to the pixel value in the image).

[0096] For an image, the representation of a pixel in the image is pixel (x, y), the pixel value of the pixel (x, y) is p(x, y), and the image is represented in the form of a matrix, such as Figure 2 Shown is an example image.

[0097] In an embodiment of the present application, in order to construct a handover progress chart and a collection progress chart, it is necessary to label the projects, tasks, work and daily activities in the collection progress network and the handover progress network in numerical form, such as project 1, project 2, project 3 and project 4, etc., task 1, task 2, task 3 and task 4, etc., work 1, work 2, work 3 and work 4, etc., daily activity item 1, daily activity item 2, daily activity item 3 and daily activity item 4, etc.

[0098] The handover progress chart and collection progress chart also exist in the form of a matrix, but the representation dimension of their nodes is determined according to the unit project division and WBS planning, that is, the representation dimension of their nodes is based on the project division into several levels, grades and projects (that is, the level of division of projects or projects), that is, the depth of the tree.

[0099] In the embodiment of the present application, in the construction drawing or progress diagram, for nodes belonging to the same tree, the value of the parent node is equal to the sum of the values ​​of its child nodes. For the construction drawing, the initial value of its smallest child node is configured when the work plan is specified; for the handover progress diagram and the collection progress diagram, the initial value of its smallest child node comes from the data registered in the process data collection data and the handover technical data, which are collected in real time by the supervisor and the constructor.

[0100] After constructing the tree (net) in the above way, when constructing the construction drawing or progress map, construct an empty image that can include the tree (net). For the pixel points in the image, if they exist in the tree (net), assign the value of the node in the tree (net) to the pixel point as the pixel value of the pixel point; if the pixel point in the image cannot find the corresponding node in the tree (net), set the pixel value of the pixel point to 0. After all the pixels in the image are assigned values, the corresponding map of the tree (net) is obtained. In this way, the handover progress map and the collection progress map are constructed respectively.

[0101] In the above step S103, based on the handover progress chart and the collection progress chart, a mapping relationship model between the handover progress chart and the collection progress chart is constructed to obtain correction parameters of the handover progress chart and the collection progress chart, specifically including:

[0102] Sampling is performed in the handover progress chart and the collection progress chart respectively to obtain a preset number of progress matching data pairs;

[0103] Based on the preset number of progress matching data pairs, a mapping matrix is ​​obtained by using the mapping relationship model constructed by the following formula 1:

[0104] C = HJ formula 1;

[0105] Wherein, H is the mapping matrix; C is the acquisition progress chart; J is the handover progress chart;

[0106] Based on the handover progress chart, the collection progress chart and the mapping matrix, correction parameters of the handover progress chart and the collection progress chart are obtained in a preset manner.

[0107] In the embodiment of the present application, samples are collected from the handover progress graph and the collection progress graph respectively to obtain a preset number of progress matching data pairs, each pair of progress matching data pairs includes handover data from the handover progress graph and collection data from the collection progress graph. The preset number is set according to the actual situation, for example, at least three pairs of progress matching data pairs can be collected.

[0108] In each pair of progress matching data, when the Euclidean distance between the handover data and the collection data is less than the set parameter, the two are considered to be matched. The set parameter can be 0.1, 0.2 or 0.3, which can be set according to the actual scenario of the project, or the variance between the handover progress chart and the collection progress chart can be used as the set parameter.

[0109] In the embodiment of the present application, the preset number of progress matching data pairs are substituted into the above formula 1 to solve the H mapping matrix. The mapping matrix represents the correspondence between the points in the handover progress map and the collection progress map, and the correspondence between the values ​​of the points in the handover progress map and the collection progress map needs to be obtained through correction parameters.

[0110] Based on the handover progress chart, the collection progress chart and the mapping matrix, the correction parameters of the handover progress chart and the collection progress chart can be obtained through a preset method. The preset method can be one of the following methods:

[0111] Method 1: Obtain the variance of the values ​​of the points in the handover progress diagram and the collection progress diagram, and use the variance as the correction parameter;

[0112] Method 2: Obtain the average Euclidean distance of all points in the handover progress graph and the collection progress graph, and use the average Euclidean distance as the correction parameter.

[0113] Method three: perform weighted summation of the Euclidean distances of all points in the handover progress graph and the collection progress graph and the Euclidean distances of all point values ​​to obtain the correction parameter.

[0114] The correction parameter obtained in the above manner may be a number or a correction parameter matrix, and the correction parameter matrix is ​​used to perform global correction on the entire acquisition progress network.

[0115] For the correction parameter matrix, the values ​​of the points in the correction parameter matrix can be obtained by obtaining the difference between the values ​​of the corresponding point pairs in the handover progress diagram and the collection progress diagram, and using the difference to perform weighted summation with the correction parameters obtained in any of the above-mentioned methods one, two and three to obtain the global parameters. The entire correction parameter matrix contains multiple numerical global correction parameters.

[0116] In the above step S104, the acquisition progress chart is corrected based on the correction parameter to obtain the process progress chart, which specifically includes:

[0117] Based on the correction parameter, the acquisition progress graph is corrected by the following formula 2 to obtain the process progress graph:

[0118] G=CH -1 B Formula 2;

[0119] Wherein, G is the process progress chart; B is the correction parameter; H -1 is the inverse matrix of the mapping matrix; C is the acquisition progress chart.

[0120] In the embodiment of the present application, the acquisition progress graph can be corrected to obtain the process progress graph by the above formula 2. After obtaining the process progress graph, the process can be reversed according to the method of obtaining the graph according to the network in S102 to realize the conversion of the process progress graph into the process progress network.

[0121] In an embodiment of the present application, before the process progress chart is input into a trained construction progress prediction model, the process progress data in the process progress chart is cleaned to obtain a cleaned process progress chart.

[0122] In the embodiment of the present application, since there are a lot of redundancies and noises in the process progress data of the process progress graph, it is necessary to clean the process progress data, wherein the cleaning method can use a Gaussian model to perform a denoising operation.

[0123] After cleaning, the process progress data after cleaning is obtained, that is, relatively accurate process progress data is obtained, which can be converted into a process progress chart after cleaning, and then the process progress chart after cleaning is input into the trained construction progress prediction model to obtain a predicted construction progress chart. Based on the process progress data after cleaning, the construction progress at a certain time node in the future can be predicted, which can improve the accuracy of the prediction and the effectiveness of project management.

[0124] In the embodiment of the present application, the trained construction progress prediction model is obtained by the following method:

[0125] Based on the LSTM network, a construction progress prediction model is built;

[0126] Based on a plurality of historical construction progress diagrams obtained in advance, the construction progress prediction model is trained until the loss function of the LSTM network converges, thereby obtaining a trained construction progress prediction model.

[0127] In the embodiment of the present application, the LSTM network is used to construct a construction progress prediction model. In addition, the time series prediction method can also be used to predict the construction progress. Here, the construction progress prediction model constructed by the LSTM network is described:

[0128] In the above step S105, the process progress diagram is input into the trained construction progress prediction model to obtain the predicted construction progress diagram. Specifically, the cleaned process progress diagram is input into the pre-trained LSTM network, and the LSTM network outputs the predicted construction progress diagram. The process progress network can also be input into the pre-trained LSTM network, and the LSTM network outputs the predicted construction progress network.

[0129] For the training of the LSTM network, multiple historical construction progress graphs (networks) can be obtained and used as training samples to train the LSTM network until the loss function of the LSTM network converges. The LSTM network training is completed and a trained LSTM network is obtained, that is, a trained construction progress prediction model is obtained.

[0130] In the above step S106, the adjustment plan network includes a first adjustment structure, a second adjustment structure and a third adjustment structure; the predicted construction progress map, the process progress map and the pre-built construction plan map are input into the pre-trained adjustment plan network to obtain a feature map including a predicted critical path, including:

[0131] Inputting the construction plan diagram into a third adjustment structure, and extracting features of the construction plan diagram through a third GCN network;

[0132] Predicting the features of the construction plan diagram through a third LSTM network to obtain a third preliminary prediction result;

[0133] Predicting the preliminary prediction result through a third RCN network to obtain a third prediction result;

[0134] Inputting the predicted construction progress graph into a first adjustment structure, and extracting features of the predicted construction progress graph through a first GCN network;

[0135] The features of the construction plan diagram are integrated with the features of the predicted construction progress diagram, and a first preliminary prediction result is obtained through a first LSTM network;

[0136] The third preliminary prediction result is integrated with the first preliminary prediction result to obtain a first prediction result through a first RCN network;

[0137] Inputting the process progress graph into a second adjustment structure, and extracting features of the process progress graph through a second GCN network;

[0138] The features of the construction plan diagram, the features of the predicted construction progress diagram, and the features of the process progress diagram are integrated to obtain a second preliminary prediction result through a second LSTM network;

[0139] The third preliminary prediction result, the first preliminary prediction result and the second preliminary prediction result are fused through a second RCN network to obtain a feature graph containing a prediction key path.

[0140] In the embodiment of the present application, the critical path is comprehensively planned based on the predicted construction progress, the actual construction progress, and the originally set construction plan, and the actual construction progress is adjusted in combination with the predicted construction progress and the planned construction period. Among them, the predicted construction progress is represented by a predicted construction progress chart, the actual construction progress is represented by a cleaned process progress chart, and the originally set construction plan is represented by a pre-built construction plan chart (a construction plan chart obtained from the construction network).

[0141] In the embodiment of the present application, the adjustment plan network is composed of three groups of adjustment structures, namely the first adjustment structure, the second adjustment structure and the third adjustment structure, and each group of adjustment structures includes a GCN, an LSTM and an RCN network. When constructing the adjustment plan network, the output of the second adjustment structure is used as the output of the entire adjustment plan network, and the output of the first adjustment structure and the output of the third adjustment structure are used to constrain the output of the second adjustment structure. In this way, the process and plan can be adjusted dynamically to achieve reliable, convenient and efficient management of engineering projects.

[0142] In the embodiment of the present application, the structure of the planned network is adjusted as follows: Figure 3 As shown, the input of the first adjustment structure is the predicted construction progress diagram, the input of the second adjustment structure is the process progress diagram after cleaning, and the input of the third adjustment structure is the pre-built construction plan diagram.

[0143] By obtaining historical predicted construction progress diagrams, process progress diagrams and construction plan diagrams as training samples, the adjustment plan network is trained in advance until the loss function of the adjustment plan network converges, the training is completed, and a trained adjustment plan network is obtained.

[0144] In the embodiment of the present application, the predicted construction progress map, the process progress map and the pre-built construction plan map are respectively input into the first adjustment structure, the second adjustment structure and the third adjustment structure in the trained adjustment structure network for prediction. The specific prediction process is as follows:

[0145] In the third adjustment structure, the features of the construction plan drawing are extracted through the third GCN network (GCN3). After the features of the construction plan drawing are extracted, the features of the construction plan drawing are input into the third LSTM network (LSTM3) for process prediction, and the prediction result of LSTM3 (the third preliminary prediction result) is input into the third RCN network (RCN3). RCN3 performs further prediction based on the third preliminary prediction result to obtain the third prediction result.

[0146] In the first adjustment structure, the features of the predicted construction progress chart are extracted through the first GCN network (GCN1). After the features of the predicted construction progress chart are extracted, they are fused with the features of the construction plan chart output by GCN3, and then input into the first LSTM network (LSTM1) for process prediction. The prediction result of LSTM1 (the first preliminary prediction result) is fused with the third preliminary prediction result, and then input into the first RCN network (RCN1). RCN1 makes further predictions based on the first preliminary prediction results to obtain the first prediction results.

[0147] In the second adjustment structure, the features of the process progress chart are extracted through the second GCN network (GCN2). After the features of the process progress chart are extracted, the features of the process progress chart are input into the second LSTM network (LSTM2). LSTM2 combines the first preliminary prediction result, the third preliminary prediction result and the features of the process progress chart for process prediction to obtain the second preliminary prediction result. The third preliminary prediction result is input into the second RCN network (RCN2). RCN2 combines the first prediction result, the second prediction result and the second preliminary prediction result for further prediction to obtain the feature map. The feature map can be used as a critical path, that is, the feature information contained in the feature map is the most important feature of the entire construction process. It combines the above-mentioned weights, construction period and other factors in it. Then the information contained in the feature map is the most critical construction path information, that is, the feature map containing the predicted critical path is obtained.

[0148] In the above prediction process, the first preliminary prediction result, the third preliminary prediction result and the features of the process schedule chart can be fused by weighted sum to obtain a new feature; the first prediction result, the second prediction result and the second preliminary prediction result can also be fused by weighted sum.

[0149] The above prediction results are a comprehensive prediction of multiple factors. The rationality and reliability of the planning path and the prediction results have been improved, and can be used as a reference standard for construction management. Based on the earned value analysis method and fuzzy algorithm of the construction progress in the project, a dynamic analysis model is constructed to dynamically predict and adjust the progress of key lines, and intelligently detect, analyze and correct the construction status, providing a basis for construction progress management.

[0150] In the embodiment of the present application, after obtaining the feature graph including the predicted critical path, the following is further included:

[0151] Comparing the characteristics of the feature map with the characteristics of the construction plan map, determining whether the difference between the construction period of each feature point in the feature map and the construction period of the corresponding feature point in the characteristics of the construction plan map is greater than a preset threshold;

[0152] If yes, a progress warning is issued for the items corresponding to the characteristic points;

[0153] Otherwise, there is no need to issue a progress warning for the matters corresponding to the characteristic points.

[0154] In the embodiment of the present application, according to the obtained feature map containing the predicted critical path, the features of the construction plan map output by GCN3 are combined for comparison. If the duration difference between a feature point in the feature map and the corresponding feature point in the features of the construction plan map output by GCN3 is greater than a preset threshold, it means that the construction progress of the matter corresponding to the feature point has a high risk of affecting the entire project duration, and the progress of the construction progress of the matter can be automatically warned, otherwise no warning is required. By setting a progress warning, problems can be discovered early in the construction process of a long project, trends can be accurately and timely judged, and active control of the project progress can be achieved. At the same time, the determination of the warning limit can also change with changing conditions, and regular adjustments and modifications can be made. In addition, data can also be input into a visualization tool for visualization.

[0155] In the present application embodiment, Figure 4 As shown in the figure, it is a schematic diagram of the overall framework of the dynamic prediction progress method based on the critical path, which can be mainly divided into four parts: progress planning, data collection, progress status evaluation and early warning, and data visualization. Among them, the progress planning part is the preparatory work for the construction of the project, including the division of unit projects and WBS planning, further planning at all levels, and further decomposition and weight configuration of the standard process of the work package to obtain the construction network; the data collection part is mainly to collect the process data progress and the submission of the handover technical data, and correct the process progress data; in the progress status evaluation and early warning part, firstly, through the construction package progress calculation, professional progress calculation, sub-project progress calculation and sub-project progress calculation, the critical path calculation and unit project progress calculation are carried out. The specific calculation process refers to the above steps S105 and S106. After obtaining the characteristic diagram containing the predicted critical path, the progress deviation analysis and progress risk early warning can be further carried out; the data visualization part can visualize the progress report and the relevant content of the engineering sand table.

[0156] The dynamic prediction progress method based on the critical path provided by the embodiment of the present application obtains the collection progress chart and the handover progress chart respectively through the pre-acquired process data collection data and the handover technical data, and then obtains the process progress chart, and then inputs the process progress chart into the trained construction progress prediction model to obtain the predicted construction progress chart, and finally, inputs the predicted construction progress chart, the process progress chart and the pre-built construction plan chart into the pre-trained adjustment plan network to obtain the feature chart containing the predicted critical path. The method predicts the critical path information through the pre-trained adjustment plan network, can realize automatic monitoring and planning of the construction progress, and derives a reliable construction route, which provides a strong support for the construction of the project, helps the project to achieve efficient and reliable management, and improves the efficiency and accuracy of the project management; based on the process progress data acquired in real time, the dynamic prediction of the critical path and the construction status analysis can be carried out, which can more accurately predict the project progress, and the construction can be adjusted in real time through the results of the dynamic prediction of the critical path, which can better control the project progress and avoid project delays.

[0157] Embodiment 2

[0158] Based on the same inventive concept, the embodiment of the present application also provides a dynamic progress prediction device based on a key path, referring to Figure 5 As shown, the device comprises:

[0159] The first construction module 101 is used to construct a collection progress network based on the pre-acquired process data collection information, and to construct a handover progress network based on the pre-acquired handover technical information;

[0160] The second construction module 102 is used to obtain a handover progress chart and a collection progress chart respectively according to the handover progress network and the collection progress network;

[0161] The third construction module 103 is used to construct a mapping relationship model between the handover progress chart and the collection progress chart based on the handover progress chart and the collection progress chart, and obtain correction parameters of the handover progress chart and the collection progress chart;

[0162] A correction module 104, used to correct the acquisition progress chart based on the correction parameters to obtain a process progress chart;

[0163] The first prediction module 105 is used to input the process progress diagram into the trained construction progress prediction model to obtain a predicted construction progress diagram;

[0164] The second prediction module 106 is used to input the predicted construction progress diagram, the process progress diagram and the pre-built construction plan diagram into a pre-trained adjustment plan network to obtain a feature diagram containing a predicted critical path.

[0165] Embodiment 3

[0166] Based on the same inventive concept, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for dynamically predicting progress based on critical paths as described in the first embodiment above is implemented.

[0167] Embodiment 4

[0168] Based on the same inventive concept, an embodiment of the present application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the dynamic progress prediction method based on the critical path as described in the above-mentioned embodiment 1 is implemented.

[0169] Embodiment 5

[0170] Based on the same inventive concept, an embodiment of the present application also provides a computer program product comprising instructions. When the computer program product is executed on a computer device, the computer device executes the dynamic progress prediction method based on the critical path as described in the above embodiment 1.

[0171] Embodiment 6

[0172] Based on the same inventive concept, an embodiment of the present application also provides a chip, the chip includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the dynamic progress prediction method based on the critical path as described in the above embodiment one.

[0173] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.

[0174] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0175] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0176] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0177] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A dynamic progress prediction method based on critical lines, It is characterized in that include: Construct a collection progress network based on the pre-acquired process data collection information, and construct a handover progress network based on the pre-acquired handover technical information; According to the delivery progress network and the collection progress network, a delivery progress chart and a collection progress chart are obtained respectively; Based on the handover progress chart and the collection progress chart, a mapping relationship model between the handover progress chart and the collection progress chart is constructed to obtain correction parameters of the handover progress chart and the collection progress chart; Based on the correction parameters, the acquisition progress chart is corrected to obtain a process progress chart; Inputting the process progress diagram into a trained construction progress prediction model to obtain a predicted construction progress diagram; The predicted construction progress diagram, the process progress diagram and the pre-built construction plan diagram are input into a pre-trained adjustment plan network to obtain a feature diagram containing a predicted critical path.

2. The method according to claim 1, It is characterized in that The adjustment plan network includes a first adjustment structure, a second adjustment structure and a third adjustment structure; the predicted construction progress map, the process progress map and the pre-built construction plan map are input into the pre-trained adjustment plan network to obtain a feature map containing a predicted critical path, including: Inputting the construction plan diagram into a third adjustment structure, and extracting features of the construction plan diagram through a third GCN network; Predicting the features of the construction plan diagram through a third LSTM network to obtain a third preliminary prediction result; Predicting the preliminary prediction result through a third RCN network to obtain a third prediction result; Inputting the predicted construction progress graph into a first adjustment structure, and extracting features of the predicted construction progress graph through a first GCN network; The features of the construction plan diagram are integrated with the features of the predicted construction progress diagram, and a first preliminary prediction result is obtained through a first LSTM network; The third preliminary prediction result is integrated with the first preliminary prediction result to obtain a first prediction result through a first RCN network; Inputting the process progress graph into a second adjustment structure, and extracting features of the process progress graph through a second GCN network; The features of the construction plan diagram, the features of the predicted construction progress diagram, and the features of the process progress diagram are integrated to obtain a second preliminary prediction result through a second LSTM network; The third preliminary prediction result, the first preliminary prediction result and the second preliminary prediction result are fused through a second RCN network to obtain a feature graph containing a prediction key path.

3. The method according to claim 1, It is characterized in that The step of obtaining a handover progress chart and a collection progress chart respectively according to the handover progress network and the collection progress network includes: An empty image is constructed according to the handover progress network, and the value of each node in the handover progress network is used as the pixel value of the pixel point at the corresponding position in the empty image to obtain a handover progress map; An empty image is constructed according to the acquisition progress network, and the value of each node in the acquisition progress network is used as the pixel value of the pixel point at the corresponding position in the empty image to obtain an acquisition progress map.

4. The method according to claim 1, It is characterized in that The step of constructing a mapping relationship model between the handover progress chart and the collection progress chart based on the handover progress chart and the collection progress chart to obtain correction parameters of the handover progress chart and the collection progress chart includes: Sampling is performed in the handover progress chart and the collection progress chart respectively to obtain a preset number of progress matching data pairs; Based on the preset number of progress matching data pairs, a mapping matrix is ​​obtained by using the mapping relationship model constructed by the following formula 1: C = HJ formula 1; Wherein, H is the mapping matrix; C is the acquisition progress chart; J is the handover progress chart; Based on the handover progress chart, the collection progress chart and the mapping matrix, correction parameters of the handover progress chart and the collection progress chart are obtained.

5. The method according to claim 1, It is characterized in that The step of correcting the acquisition progress chart based on the correction parameter to obtain a process progress chart includes: Based on the correction parameter, the acquisition progress graph is corrected by the following formula 2 to obtain the process progress graph: G=CH -1 B Formula 2; Wherein, G is the process progress chart; B is the correction parameter; H -1 is the inverse matrix of the mapping matrix; C is the acquisition progress chart.

6. The method according to claim 1, It is characterized in that The trained construction progress prediction model is obtained by the following method: Based on the LSTM network, a construction progress prediction model is built; Based on a plurality of historical construction progress diagrams obtained in advance, the construction progress prediction model is trained until the loss function of the LSTM network converges, thereby obtaining a trained construction progress prediction model.

7. The method according to claim 2, It is characterized in that Also includes: Comparing the characteristics of the feature map with the characteristics of the construction plan map, determining whether the difference between the construction period of each feature point in the feature map and the construction period of the corresponding feature point in the characteristics of the construction plan map is greater than a preset threshold; If so, a progress warning is issued for the items corresponding to the feature points.

8. The method according to any one of claims 1 to 7, It is characterized in that Also includes: Before inputting the process progress chart into the trained construction progress prediction model, the process progress data in the process progress chart is cleaned to obtain a cleaned process progress chart.

9. A dynamic progress prediction device based on key paths, It is characterized in that include: The first construction module is used to construct a collection progress network based on the pre-acquired process data collection information, and to construct a handover progress network based on the pre-acquired handover technical information; The second construction module is used to obtain a handover progress chart and a collection progress chart respectively according to the handover progress network and the collection progress network; A third construction module is used to construct a mapping relationship model between the handover progress diagram and the collection progress diagram based on the handover progress diagram and the collection progress diagram, and obtain correction parameters of the handover progress diagram and the collection progress diagram; A correction module, used for correcting the acquisition progress chart based on the correction parameters to obtain a process progress chart; A first prediction module is used to input the process progress diagram into a trained construction progress prediction model to obtain a predicted construction progress diagram; The second prediction module is used to input the predicted construction progress diagram, the process progress diagram and the pre-built construction plan diagram into a pre-trained adjustment plan network to obtain a feature diagram containing a predicted critical path.

10. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the program is executed by a processor, the processor executes the method for dynamically predicting progress based on a critical path according to any one of claims 1 to 8.

11. A computer device, It is characterized in that The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for dynamically predicting progress based on a critical path as claimed in any one of claims 1 to 8 is implemented.

12. A computer program product comprising instructions, which, when executed on a computer device, enables the computer device to execute the method for dynamically predicting progress based on a critical path according to any one of claims 1 to 8.

13. A chip, comprising a processor and a communication interface, wherein the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the method for dynamically predicting progress based on a critical path as described in any one of claims 1 to 8.

Images