A deep learning-based evaluation method for digital construction projects

Through the digital construction project evaluation method based on deep learning, the subjectivity and inaccuracy of traditional evaluation methods are solved, and the evaluation is carried out using convolutional neural network to ensure the accuracy and transparency of project evaluation.

CN118469331BActive Publication Date: 2024-11-22BEIJING ZHONGZIHAIWAI CONSULTATION CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410636902.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-22
Publication Date
2024-11-22
Estimated Expiration
2044-05-22

AI Technical Summary

Technical Problem

Traditional project evaluation methods focus on qualitative analysis, lack of quantitative data and comparative analysis, resulting in subjectivity and inaccuracy of evaluation results, affecting the progress and quality of the project.

Method used

The digital construction project evaluation method based on deep learning is adopted to obtain and preprocess historical data, extract key performance indicators, determine their importance, and use convolutional neural networks to train to generate evaluation reports to ensure quantitative data and comparative analysis.

Benefits of technology

The accuracy and objectivity of the evaluation results are achieved, ensuring the reliability of project progress and quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118469331B_ABST
    Figure CN118469331B_ABST
Patent Text Reader

Abstract

The present invention provides an evaluation method for digital construction projects based on deep learning, which belongs to the technical field of project evaluation, and includes: obtaining relevant data of historical digital construction projects, and preprocessing the relevant data; obtaining key performance indicators of digital construction projects according to the preprocessing results; determining the importance of the key performance indicators, and sequentially obtaining performance indicator information of target digital construction projects according to the importance; training a convolutional neural network based on the evaluation system of digital construction projects to obtain an evaluation model, and evaluating the change trend and difference degree of the various indicators, and generating a corresponding evaluation report. The method solves the problem of subjectivity and inaccuracy of evaluation results caused by the lack of quantitative data and comparative analysis in the past.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of project evaluation, and in particular to an evaluation method for digital construction projects based on deep learning. Background Art

[0002] As the world pays more and more attention to sustainable development, large-scale construction projects play an important role in promoting sustainable development.

[0003] Traditional evaluation methods often focus on qualitative analysis, lack quantitative data and comparative analysis, and are unable to accurately and intuitively obtain the differences and changes in various indicators, resulting in subjectivity and inaccuracy in the evaluation results, which in turn affects the overall progress and quality of the project.

[0004] Therefore, the present invention proposes an evaluation method for digital construction projects based on deep learning. Summary of the invention

[0005] The present invention provides an evaluation method for digital construction projects based on deep learning, which is used to solve the defects of the prior art that it focuses on qualitative analysis, lacks quantitative data and comparative analysis, cannot accurately and intuitively obtain the differences and changes of various indicators, and leads to subjectivity and inaccuracy in the evaluation results.

[0006] In one aspect, the present invention provides a method for evaluating a digital construction project based on deep learning, comprising:

[0007] Step 1: Acquire relevant data of historical digital construction projects and pre-process the relevant data;

[0008] Step 2: Obtain key performance indicators of the digital construction project based on the preprocessing results;

[0009] Step 3: Determine the importance of the key performance indicators, and obtain the performance indicator information of the target digital construction project in sequence according to the importance;

[0010] Step 4: Determine the change trend and difference degree of each indicator based on the performance indicator information of the target digital construction project based on the vertical and horizontal data comparison method;

[0011] Step 5: Based on the evaluation system of digital construction projects, the convolutional neural network is trained to obtain an evaluation model, and the changing trends and differences of the indicators are evaluated to generate a corresponding evaluation report.

[0012] According to a deep learning-based digital construction project evaluation method provided by the present invention, relevant data of historical digital construction projects are obtained and the relevant data are preprocessed, including:

[0013] Determine the data type and scope of the relevant data of the historical digital construction project that needs to be obtained, and determine the data source and format based on the data type and scope;

[0014] Determine appropriate tools and techniques to collect and extract data based on the data sources and formats described;

[0015] The collected and extracted data are formatted and preprocessed.

[0016] According to a deep learning-based digital construction project evaluation method provided by the present invention, key performance indicators of digital construction projects are obtained according to preprocessing results, including:

[0017] Acquire multiple requirements for historical digitization construction projects;

[0018] extracting all key information related to the digital construction project from the preprocessing results according to multiple requirements of the project;

[0019] Analyze the key information to obtain the project progress at different time points and stages;

[0020] Determine influencing factors according to the project progress, and obtain key performance indicators of the digital construction project based on the influencing factors.

[0021] According to a deep learning-based digital construction project evaluation method provided by the present invention, the importance of the key performance indicators is determined, and performance indicator information of the target digital construction project is obtained in sequence according to the importance, including:

[0022] Get the goals of the target digital construction project;

[0023] Classifying the key performance indicators based on the objectives of the digital construction project, and evaluating the impact on the project and the relevance to the project objectives according to the classification results;

[0024] Determine the importance of the key performance indicator according to the impact degree and relevance degree;

[0025] Prioritize the multiple key performance indicators according to the importance, and obtain the performance indicator information of the target digital construction project in sequence according to the ranking results.

[0026] According to a deep learning-based digital construction project evaluation method provided by the present invention, the change trend and difference degree of each indicator are determined based on the vertical and horizontal data comparison method according to the performance indicator information of the target digital construction project, including:

[0027] Determine horizontal indicator dimensions and vertical indicator dimensions according to the performance indicators of the target digital construction project;

[0028] Based on the horizontal indicator dimension and the vertical indicator dimension, the horizontal dimension description parameters and the vertical dimension description parameters are determined by using the vertical and horizontal data comparison method;

[0029] The changing trend and degree of difference of each indicator are determined based on the horizontal dimension description parameters and the vertical dimension description parameters.

[0030] According to the present invention, a method for evaluating a digital construction project based on deep learning is provided, characterized in that a convolutional neural network is trained based on the evaluation system of the digital construction project to obtain an evaluation model, and the change trend and difference degree of each indicator are evaluated to generate a corresponding evaluation report, including:

[0031] Obtain the evaluation criteria for various indicators of digital construction projects and build an evaluation system for digital construction projects based on the evaluation criteria;

[0032] Based on the evaluation system of digital construction projects, the model is trained through convolutional neural networks to obtain the evaluation model of digital construction projects;

[0033] Evaluate the change trend and difference degree of each indicator through the evaluation model to obtain the evaluation conclusion;

[0034] Generate a corresponding evaluation report based on the evaluation conclusions.

[0035] According to a deep learning-based digital construction project evaluation method provided by the present invention, obtaining a target of a target digital construction project includes:

[0036] Determine the project application scope and application scenarios of the target digital construction project;

[0037] Determine the target audience based on the project application scope and application scenarios of the target digital construction project;

[0038] Determine the goals of the target digital construction project based on the audience goals and the business functions of the target digital construction project.

[0039] A method for evaluating a digital construction project based on deep learning provided by the present invention further includes:

[0040] Decomposing the collected data related to the digital construction project to obtain the digital construction project category data and its corresponding service frequency data, and determining the usage intensity of the digital construction project category data according to the service frequency data;

[0041] Compare the usage intensity parameter with a preset usage intensity parameter, filter out the digital construction project category data whose usage intensity parameter is greater than the first preset usage intensity parameter, and arrange them from large to small based on the comparison difference to establish a first queue;

[0042] Compare the usage intensity parameter with a preset usage intensity parameter, filter out the digital construction project category data whose usage intensity parameter is less than the first preset usage intensity parameter, and arrange them from small to large based on the comparison difference to establish a second queue;

[0043] Establishing an association relationship between the first queuing queue and the second queuing queue, and performing corresponding matching based on the association relationship to obtain a project plan data group, a project fund use data group, and a project execution parameter data group;

[0044] Determine target values ​​of the project plan data group, the project fund usage data group, and the project execution parameter data group, and draw change curves of the project plan data group, the project fund usage data group, and the project execution parameter data group based on the target values;

[0045] Construct an evaluation model and determine the key element weights of the project plan data group, project fund use data group, and project execution parameter data group;

[0046] Based on the key element weight values, the project planning, cost and execution status of the digital construction project are analyzed to obtain key element evaluation values ​​of the performance indicators of the digital construction project.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] By preprocessing the data of historical construction projects, we can obtain the key performance indicators and importance of the projects, thereby obtaining the performance indicator information of the target digital construction projects. By comparing the indicator information vertically and horizontally to determine the changing trends and degree of difference of each indicator and conduct an evaluation, we can ensure quantitative data and comparative analysis, and ensure the subjectivity and accuracy of the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 It is a flowchart of a method for evaluating a digital construction project based on deep learning provided by an embodiment of the present invention;

[0051] Figure 2 It is another flow chart of a method for evaluating a digital construction project based on deep learning provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Embodiment 1:

[0054] An evaluation method for a digital construction project based on deep learning is provided in an embodiment of the present invention. Figure 1 As shown, the method mainly includes the following steps:

[0055] Step 1: Acquire relevant data of historical digital construction projects and pre-process the relevant data;

[0056] Step 2: Obtain key performance indicators of the digital construction project based on the preprocessing results;

[0057] Step 3: Determine the importance of the key performance indicators, and obtain the performance indicator information of the target digital construction project in sequence according to the importance;

[0058] Step 4: Determine the change trend and difference degree of each indicator based on the performance indicator information of the target digital construction project based on the vertical and horizontal data comparison method;

[0059] Step 5: Based on the evaluation system of digital construction projects, the convolutional neural network is trained to obtain an evaluation model, and the changing trends and differences of the indicators are evaluated to generate a corresponding evaluation report.

[0060] In this embodiment, the relevant data includes: project background and objectives, user requirements, development progress and resource allocation, project cycle, testing and acceptance criteria.

[0061] In this embodiment, data preprocessing refers to cleaning, converting, integrating and normalizing the original data.

[0062] In this embodiment, the key performance indicators of the construction project generally include: project progress, project cost, project quality, customer satisfaction and risk management.

[0063] In this embodiment, the vertical and horizontal data comparison method is a method for comparing data from different sources to detect inconsistencies or errors in the data. This method divides the data into two parts, horizontal and vertical, for comparison. The horizontal comparison involves comparing data from different sources at the same point in time, such as comparing the construction progress recorded by the system and the construction progress recorded by the workers on the same day. The vertical comparison involves comparing data from different sources at the same point in time, such as comparing the operation of the same working machine in different time periods.

[0064] In this embodiment, the convolutional neural network is a special type of neural network that mainly processes input data with spatial local correlation. It uses a convolutional layer to extract local features of the data and a pooling layer to reduce redundant information and the number of parameters.

[0065] The beneficial effects of the above technical solution are: by preprocessing the data of historical construction projects, obtaining the key performance indicators and importance of the projects, thereby obtaining the performance indicator information of the target digital construction projects, and comparing the indicator information vertically and horizontally to determine the changing trends and degree of difference of each indicator and conduct evaluation, which can ensure quantitative data and comparative analysis, and ensure the objectivity and accuracy of the evaluation results.

[0066] Embodiment 2:

[0067] An evaluation method for a digital construction project based on deep learning is provided in an embodiment of the present invention. Figure 2 As shown, relevant data of historical digital construction projects are obtained and preprocessed, including:

[0068] S01: Determine the data type and scope of the relevant data of the historical digital construction project that needs to be obtained, and determine the data source and format according to the data type and scope;

[0069] S02: Determine appropriate tools and techniques to collect and extract data based on the data sources and formats;

[0070] S03: Convert the format of the collected and extracted data, and pre-process the converted data.

[0071] In this embodiment, the data types include: customer data, product data, sales data, human resources data, and financial data.

[0072] In this embodiment, the data range includes: internal data, external data, historical data, and real-time data.

[0073] In this embodiment, the data source may be: files or databases in multiple different formats.

[0074] In this embodiment, the data format may be: CSV, JSON, Excel table.

[0075] In this embodiment, the tools for extracting data include: crawler tools, data visualization tools, and big data analysis tools.

[0076] In this embodiment, format conversion refers to a process of converting one data format into another data format, such as converting an Excel file into PDF, or converting a PPT file into HTML.

[0077] The beneficial effects of the above technical solution are: by obtaining the data source and format of historical digital construction projects, determining appropriate tools and technologies to collect and extract data, and performing format conversion, the availability of data can be improved and the accuracy and security of data can be ensured.

[0078] Embodiment 3:

[0079] An evaluation method for a digital construction project based on deep learning provided by an embodiment of the present invention obtains key performance indicators of the digital construction project according to preprocessing results, including:

[0080] Acquire multiple requirements for historical digitization construction projects;

[0081] extracting all key information related to the digital construction project from the preprocessing results according to multiple requirements of the project;

[0082] Analyze the key information to obtain the project progress at different time points and stages;

[0083] Determine influencing factors according to the project progress, and obtain key performance indicators of the digital construction project based on the influencing factors.

[0084] In this embodiment, multiple requirements for a digital construction project include: determining project goals and scope, defining project scope, developing a project plan and schedule, and determining required resources and budget.

[0085] In this embodiment, the key information is, for example, the formulation of a project plan and schedule, and the key information includes the start and end time of the project and the work cycle.

[0086] In this embodiment, the influencing factors may be: time constraints, resource availability, budget constraints, and risk management.

[0087] In this embodiment, the key performance indicators of the digital construction project generally include: project progress, project cost, project quality, customer satisfaction and risk management.

[0088] The beneficial effects of the above technical solution are: by extracting and analyzing all key information related to the digital construction project according to multiple requirements of the project, determining the project progress, and determining the factors affecting the progress based on the project progress, thereby obtaining the key performance indicators of the digital construction project, better managing the project and improving the transparency of the project.

[0089] Embodiment 4:

[0090] An evaluation method for a digital construction project based on deep learning provided by an embodiment of the present invention determines the importance of the key performance indicators, and sequentially obtains performance indicator information of a target digital construction project according to the importance, including:

[0091] Get the goals of the target digital construction project;

[0092] Classifying the key performance indicators based on the objectives of the digital construction project, and evaluating the impact on the project and the relevance to the project objectives according to the classification results;

[0093] Determine the importance of the key performance indicator according to the impact degree and relevance degree;

[0094] Prioritize the multiple key performance indicators according to the importance, and obtain the performance indicator information of the target digital construction project in sequence according to the ranking results.

[0095] In this embodiment, the main goals of the digital construction project include: improving the operational efficiency of the enterprise, reducing costs, improving the quality and innovation capabilities of products or services, and enhancing customer satisfaction and loyalty.

[0096] In this embodiment, the key performance indicators can be classified according to different dimensions, such as:

[0097] Strategic level: For example, revenue growth, market share expansion, improved customer satisfaction, etc.

[0098] Operational level: for example, increased production efficiency, improved inventory turnover, and reduced employee turnover.

[0099] Customer service level: for example, customer satisfaction scores, reduced complaint rates, faster after-sales service response times, etc.

[0100] Innovation level: for example, an increase in the number of new products launched, a higher return on R&D investment, and an increase in the number of patent applications.

[0101] The beneficial effect of the above technical solution is: by determining the impact of key indicators on the project and the degree of relevance to the project goals, determining the importance of the key performance indicators and prioritizing them, it is possible to ensure the realization of project goals, thereby obtaining performance indicator information of the target digital construction project and improving the accuracy of the evaluation results.

[0102] Embodiment 5:

[0103] An evaluation method for a digital construction project based on deep learning provided by an embodiment of the present invention determines the change trend and difference degree of each indicator based on the vertical and horizontal data comparison method according to the performance indicator information of the target digital construction project, including:

[0104] Determine horizontal indicator dimensions and vertical indicator dimensions according to the performance indicators of the target digital construction project;

[0105] Based on the horizontal indicator dimension and the vertical indicator dimension, the horizontal dimension description parameters and the vertical dimension description parameters are determined by using the vertical and horizontal data comparison method;

[0106] The changing trend and degree of difference of each indicator are determined based on the horizontal dimension description parameters and the vertical dimension description parameters.

[0107] In this embodiment, the horizontal indicator dimension refers to the dimension for comparing and analyzing the same indicator at different time points. For example, in the field of construction projects, the horizontal indicator dimension may include time (quarter, year) and geographical location (city, province).

[0108] In this embodiment, the vertical indicator dimension refers to a dimension for comparing and analyzing the same indicator at the same time point. For example, in the field of construction projects, the vertical indicator dimension may include project cycle, capital investment, and market share.

[0109] In this embodiment, the vertical and horizontal data comparison method is a comparative analysis method used to compare the differences and similarities between two or more variables, usually by comparing the values ​​or measurement indicators of each variable and grouping them according to different classification standards.

[0110] In this embodiment, the horizontal dimension description parameters refer to descriptive indicators used in horizontal data analysis to measure and describe the differences between different categories. These indicators are usually quantitative and can be used to sort, group or segment different categories. Common horizontal dimension description parameters include: project progress and risk assessment.

[0111] In this embodiment, the longitudinal dimension description parameter refers to an indicator of a specific object in the same time period. Common longitudinal dimension description parameters include: machine operation parameters.

[0112] The beneficial effects of the above technical solution are: by obtaining the horizontal and vertical indicator dimensions of the target digital construction project, using the vertical and horizontal data comparison method to determine the horizontal dimension description parameters and the vertical dimension description parameters, and determining the changes and differences of the indicators, it can help evaluate and intervene in the effects of indicators at different time points or in different groups. By comparing data from different periods and different groups, it can be determined whether a certain variable has a significant impact, so as to optimize and improve the indicators in a targeted manner.

[0113] Embodiment 6:

[0114] An evaluation method for a digital construction project based on deep learning provided by an embodiment of the present invention trains a convolutional neural network based on an evaluation system of a digital construction project to obtain an evaluation model, evaluates the change trend and difference degree of each indicator, and generates a corresponding evaluation report, including:

[0115] Obtain the evaluation criteria for various indicators of digital construction projects and build an evaluation system for digital construction projects based on the evaluation criteria;

[0116] Based on the evaluation system of digital construction projects, the model is trained through convolutional neural networks to obtain the evaluation model of digital construction projects;

[0117] Evaluate the change trend and difference degree of each indicator through the evaluation model to obtain the evaluation conclusion;

[0118] Generate a corresponding evaluation report based on the evaluation conclusions.

[0119] In this embodiment, the evaluation standard refers to the evaluation of an indicator, such as the standard 29 days of the construction period. If this time is exceeded, the evaluation of this indicator will be low or unqualified.

[0120] In this embodiment, the evaluation system refers to a set of standards and methods for evaluating and accepting various indicators.

[0121] In this embodiment, the convolutional neural network simulates the visual processing process of the human brain through modules such as convolutional layers and pooling layers, thereby realizing feature extraction and classification of input data.

[0122] In this embodiment, in deep learning, the performance of the evaluation model is usually measured using a loss function, which is used to determine the difference between the model prediction value and the true value, and used as an indicator for optimizing the model.

[0123] The beneficial effects of the above technical solution are: according to the evaluation system of the digital construction project, the model is trained through a convolutional neural network, the evaluation model is determined, and various indicators are evaluated according to the evaluation model, which can ensure the accuracy of the evaluation and improve the efficiency of the evaluation results.

[0124] Embodiment 7:

[0125] An evaluation method for a digital construction project based on deep learning provided by an embodiment of the present invention obtains a target of a target digital construction project, including:

[0126] Determine the project application scope and application scenarios of the target digital construction project;

[0127] Determine the target audience based on the project application scope and application scenarios of the target digital construction project;

[0128] Determine the goals of the target digital construction project based on the audience goals and the business functions of the target digital construction project.

[0129] In this embodiment, the project application scope of the digital construction project includes: production management, supply chain management, human resource management, and financial management.

[0130] In this embodiment, the application scenarios of the digital construction project include: industrial manufacturing, warehousing and logistics, retail, and medical health.

[0131] In this embodiment, the business functions of the digital construction project include: data collection and storage, data analysis, reporting and dashboards.

[0132] The beneficial effects of the above technical solution are: determining the target audience according to the project application scope and application scenario, thereby determining the target construction project, which can improve the implementation efficiency of the project and save time and cost.

[0133] Embodiment 8:

[0134] An evaluation method for a digital construction project based on deep learning provided by an embodiment of the present invention also includes:

[0135] Decomposing the collected data related to the digital construction project to obtain the digital construction project category data and its corresponding service frequency data, and determining the usage intensity of the digital construction project category data according to the service frequency data;

[0136] Compare the usage intensity parameter with a preset usage intensity parameter, filter out the digital construction project category data whose usage intensity parameter is greater than the first preset usage intensity parameter, and arrange them from large to small based on the comparison difference to establish a first queue;

[0137] Compare the usage intensity parameter with a preset usage intensity parameter, filter out the digital construction project category data whose usage intensity parameter is less than the first preset usage intensity parameter, and arrange them from small to large based on the comparison difference to establish a second queue;

[0138] Establishing an association relationship between the first queuing queue and the second queuing queue, and performing corresponding matching based on the association relationship to obtain a project plan data group, a project fund use data group, and a project execution parameter data group;

[0139] Determine target values ​​of the project plan data group, the project fund usage data group, and the project execution parameter data group, and draw change curves of the project plan data group, the project fund usage data group, and the project execution parameter data group based on the target values;

[0140] Construct an evaluation model and determine the key element weights of the project plan data group, project fund use data group, and project execution parameter data group;

[0141] Based on the key element weight values, the project planning, cost and execution status of the digital construction project are analyzed to obtain key element evaluation values ​​of the performance indicators of the digital construction project.

[0142] In this embodiment, data decomposition refers to the process of breaking down the relevant data of a large, complex digital construction project into multiple smaller, easier to understand and manage components. Typically, the original data set contains a large amount of information, which is further refined, classified, and categorized to make the data easier to process and analyze.

[0143] In this embodiment, digital construction project category data refers to data that classifies and encodes different types of digital construction projects. The digital construction project category data can be collected by encoding each project according to the category to which it belongs, such as using a one-dimensional or two-dimensional coding system, or using a more advanced classification system, such as natural language processing or machine learning technology to help automatically classify the data.

[0144] In this embodiment, the service frequency data refers to the frequency data of the time when the customer or user receives each type of project data.

[0145] In this embodiment, usage intensity refers to the frequency and duration of use of a certain category of data within a certain period of time.

[0146] In this embodiment, the usage intensity parameter is a measure of how frequently a user uses a certain category of data. These indicators are usually used to monitor user behavior, such as how often they use data, how long they spend, and how quickly they complete tasks.

[0147] In this embodiment, the first queue refers to a queue that uses the actual usage intensity parameter of the project data that is greater than the first preset usage intensity parameter and the preset usage intensity parameter as the difference, and arranges the queue from large to small according to the difference. For example, if the preset usage intensity parameter is 20, then project data with a usage intensity parameter greater than 20 is obtained. For example, the project schedule usage intensity parameter is 30, the project cost usage intensity parameter is 21, and the project target usage intensity parameter is 29. Then the order of the first queue is project progress, project target, and project cost.

[0148] In this embodiment, the project plan data group is a collection of multiple data used to track and manage the progress of the project. These data may include information such as the project name, team member names, work progress, budget, and schedule.

[0149] In this embodiment, the project fund usage data group is a collection of multiple data used to track and manage the flow of funds for the project. These data may include information such as expenditure type (such as material costs, labor costs, etc.), amount, time point, and source.

[0150] In this embodiment, the project execution parameter data group refers to a data set used to describe and manage different parameters in the project. These parameters may be constants in program codes, settings in configuration files, or values ​​input by users.

[0151] The beneficial effect of the above technical solution is: by decomposing the data, taking the digital construction project category data and its corresponding service frequency data as the basis for data analysis, establishing queues and association relationships according to the usage intensity of the project category data, and obtaining the project plan data group, project fund usage data group and project execution parameter data group, and by drawing the change curves of the project plan data group, the project fund usage data group and the project execution parameter data group for analysis, an effective evaluation of the key elements of the performance indicators of the digital construction project can be achieved, thereby strictly and effectively controlling the operating conditions of the digital construction project.

[0152] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating digital construction projects based on deep learning, characterized in that: include: Step 1: Acquire relevant data of historical digital construction projects and pre-process the relevant data; Step 2: Obtain key performance indicators of the digital construction project based on the preprocessing results; Step 3: Determine the importance of the key performance indicators, and obtain the performance indicator information of the target digital construction project in sequence according to the importance; Step 4: Determine the change trend and difference degree of each indicator based on the performance indicator information of the target digital construction project based on the vertical and horizontal data comparison method; Step 5: Based on the evaluation system of the digital construction project, the convolutional neural network is trained to obtain an evaluation model, and the change trend and difference degree of each indicator are evaluated to generate a corresponding evaluation report; Among them, according to the performance indicator information of the target digital construction project, the change trend and difference degree of each indicator are determined based on the vertical and horizontal data comparison method, including: Determine the horizontal indicator dimension and the vertical indicator dimension according to the performance indicators of the target digital construction project, wherein the horizontal indicator dimension refers to the dimension for comparing and analyzing the same indicator at different time points. In the field of construction projects, the horizontal indicator dimension includes time and geographic location. The vertical indicator dimension refers to the dimension for comparing and analyzing the same indicator at the same time point. In the field of construction projects, the vertical indicator dimension includes project cycle, capital investment, and market share. Based on the horizontal indicator dimension and the vertical indicator dimension, the horizontal dimension description parameters and the vertical dimension description parameters are determined using the vertical and horizontal data comparison method. The horizontal dimension description parameters refer to the descriptive indicators used in horizontal data analysis to measure and describe the differences between different categories; the vertical dimension description parameters refer to the indicators of a specific object in the same time period; Determine the change trend and degree of difference of each indicator based on the horizontal dimension description parameters and the vertical dimension description parameters; Among them, the evaluation model is obtained by training the convolutional neural network based on the evaluation system of the digital construction project, and the change trend and difference degree of each indicator are evaluated to generate a corresponding evaluation report, which also includes: Obtain the evaluation criteria for various indicators of digital construction projects and build an evaluation system for digital construction projects based on the evaluation criteria; Based on the evaluation system of digital construction projects, the convolutional neural network is trained to obtain the evaluation model of digital construction projects; Evaluate the change trend and difference degree of each indicator through the evaluation model to obtain the evaluation conclusion; Generate a corresponding assessment report based on the assessment conclusions; The deep learning-based digital construction project evaluation methods also include: Decomposing the collected data related to the digital construction project to obtain the digital construction project category data and its corresponding service frequency data, and determining the usage intensity of the digital construction project category data according to the service frequency data; Compare the usage intensity parameter with the preset usage intensity parameter, filter out the digital construction project category data whose usage intensity parameter is greater than the first preset usage intensity parameter, and arrange them from large to small based on the comparison difference to establish a first queue; Compare the usage intensity parameter with the preset usage intensity parameter, filter out the digital construction project category data whose usage intensity parameter is less than the first preset usage intensity parameter, and arrange them from small to large based on the comparison difference to establish a second queue; Establishing an association relationship between the first queuing queue and the second queuing queue, and performing corresponding matching based on the association relationship to obtain a project plan data group, a project fund use data group, and a project execution parameter data group; Determine target values ​​of the project plan data group, the project fund usage data group, and the project execution parameter data group, and draw change curves of the project plan data group, the project fund usage data group, and the project execution parameter data group based on the target values; Construct an evaluation model and determine the key element weights of the project plan data group, project fund use data group, and project execution parameter data group; Based on the key element weight values, the project planning, cost and execution status of the digital construction project are analyzed to obtain key element evaluation values ​​of the performance indicators of the digital construction project.

2. The deep learning-based digital construction project evaluation method according to claim 1 is characterized in that: Obtain relevant data of historical digital construction projects and pre-process the relevant data, including: Determine the data type and scope of the relevant data of the historical digital construction project that needs to be obtained, and determine the data source and format based on the data type and scope; Determine appropriate tools and techniques to collect and extract data based on the data sources and formats described; The collected and extracted data are formatted and preprocessed.

3. The evaluation method for digital construction projects based on deep learning according to claim 1 is characterized in that: The key performance indicators of the digital construction project are obtained based on the preprocessing results, including: Acquire multiple requirements for historical digitization construction projects; extracting all key information related to the digital construction project from the preprocessing results according to multiple requirements of the project; Analyze the key information to obtain the project progress at different time points and stages; Determine influencing factors according to the project progress, and obtain key performance indicators of the digital construction project based on the influencing factors.

4. The evaluation method for digital construction projects based on deep learning according to claim 1 is characterized in that: Determine the importance of the key performance indicators, and obtain performance indicator information of the target digital construction project in sequence according to the importance, including: Get the goals of the target digital construction project; Classifying the key performance indicators based on the objectives of the digital construction project, and evaluating the impact on the project and the relevance to the project objectives according to the classification results; Determine the importance of the key performance indicator according to the impact degree and relevance degree; The plurality of key performance indicators are prioritized according to the importance, and the performance indicator information of the target digital construction project is obtained in sequence according to the ranking results.

5. The deep learning-based digital construction project evaluation method according to claim 4 is characterized in that: Get the goals of your digital construction project, including: Determine the project application scope and application scenarios of the target digital construction project; Determine the target audience based on the project application scope and application scenarios of the target digital construction project; Determine the goals of the target digital construction project based on the audience goals and the business functions of the target digital construction project.