Project state evaluation method and device, electronic equipment and storage medium

Abstract

The invention discloses a project state evaluation method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence or other related fields, and the method comprises the steps: regularly collecting project related data of a target project from a specified data source, and obtaining multi-source project data; determining a project stage of each piece of multi-source project data according to the project standard flow table, and generating a project stage label; the multi-source project data and the project stage labels are input into an evaluation model, the evaluation model outputs an evaluation result, the evaluation model is optimized in advance based on stage weights, the evaluation model is used for analyzing the input data in combination with a fuzzy set theory, and project state probability distribution obtained through analysis serves as the evaluation result. Through the project progress and risk assessment method and device, the technical problem that the project progress and risk assessment accuracy is low due to the fact that limitation exists when project full-life-cycle data is processed and complex project condition changes cannot be adapted in the related technology is solved.

Description

Project status assessment methods and devices, electronic equipment, storage media Technical Field

[0001] This invention relates to the field of artificial intelligence technology or other related fields. Specifically, it relates to a method and apparatus for evaluating project status, an electronic device, and a storage medium. Background Technology

[0002] In project management, accurate project schedule forecasting is crucial for ensuring on-time project completion, cost control, and resource allocation. With economic development and the increasing scale and complexity of projects, project managers face significant challenges in effectively controlling schedule and risks during project execution. Traditional project schedule forecasting methods rely on manual judgment, which is insufficient in terms of forecast accuracy and dynamic adaptability.

[0003] Existing project progress and risk prediction often rely on human experience and historical data, which has problems such as strong subjectivity, low prediction accuracy, limited ability to process big data, lack of dynamic adaptability, and poor real-time performance. It is difficult to cope with complex and ever-changing project environments and uncertainties, and it has limitations in processing project lifecycle data. It cannot adapt to changes in complex project situations, resulting in low accuracy in project progress and risk assessment.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] The main objective of this application is to provide a method, apparatus, electronic device, and storage medium for assessing project status, so as to at least solve the technical problems in the related technologies that have limitations in processing project lifecycle data, cannot adapt to complex changes in project conditions, and result in low accuracy in assessing project progress and risks.

[0006] To achieve the above objectives, according to one aspect of this application, a method for evaluating project status is provided. The method includes: periodically collecting project-related data of a target project from N specified data sources to obtain multi-source project data, where N is a positive integer; calling a standard project workflow table and determining the project stage of each piece of multi-source project data according to the standard workflow table, and generating a project stage label for each piece of multi-source project data; inputting the multi-source project data and the project stage label into an evaluation model, and having the evaluation model output an evaluation result, wherein the evaluation model is pre-optimized based on stage weights, and the evaluation model is used to analyze the input data using fuzzy set theory, with the analyzed project status probability distribution used as the evaluation result.

[0007] Furthermore, the designated data source includes at least: project management software, resource planning system, financial software, and IoT sensors. The project-related data includes at least: basic attribute data, schedule data, resource data, and risk data. The basic attribute data includes: project type, project size, project budget, geographical location, and team composition. The schedule data includes: planned execution time period information, actual execution time period information, execution percentage, and execution path of project sub-tasks. The resource data includes: manpower input hours, equipment utilization rate, material consumption, and cost expenditure. The risk data includes: identified risk events, risk categories, scope of impact, and countermeasures taken.

[0008] Furthermore, after obtaining multi-source project data, the project status assessment method further includes: preprocessing the multi-source project data, wherein the preprocessing operations include at least: data cleaning, missing value imputation, outlier removal, and data standardization.

[0009] Further, the step of optimizing the evaluation model based on stage weights includes: defining a stage weight function according to the target project, wherein the stage weight function is used to determine the weight value corresponding to the project stage in which the target project is located; selecting a machine learning model and selecting a model loss function; introducing the stage weight function into the model loss function so that the stage weight of the project stage exhibits a positive growth with the loss gradient in the model loss function; and training the machine learning model using the model loss function after introducing the stage weight function until a preset convergence condition is reached to obtain the evaluation model.

[0010] Further, the step of inputting the multi-source project data and the project stage labels into the evaluation model, and then having the evaluation model output the evaluation result, includes: the evaluation model determining M project states corresponding to the target project based on the input data, and determining the project state interval corresponding to each project state, where M is an integer greater than 2; calculating the membership degree of the input data for each project state interval, obtaining a membership degree vector composed of M membership degrees, where the sum of the M membership degrees is 1, and the membership degree vector is used to characterize the probability distribution of the project state; and generating the evaluation result based on the membership degree vector.

[0011] Furthermore, the step of generating the evaluation result based on the membership vector includes: determining the project state interval corresponding to the maximum membership value as the evaluation result according to the maximum membership principle.

[0012] Furthermore, the step of generating the evaluation result based on the membership vector further includes: when the maximum membership value is greater than a preset confidence threshold, retrieving the decision plan for the project state interval corresponding to the maximum membership value, and generating the evaluation result based on the project state interval and the decision plan; or, when the maximum membership value is less than or equal to the preset confidence threshold, calculating the decision tendency based on all the membership degrees, and calling the corresponding tendency suggestion plan based on the decision tendency, and generating the evaluation result based on the decision tendency and the tendency suggestion plan.

[0013] To achieve the above objectives, according to another aspect of this application, a project status assessment device is also provided. This device includes: a data acquisition unit, configured to periodically collect project-related data of a target project from N specified data sources to obtain multi-source project data, where N is a positive integer; a generation unit, configured to call a standard project flowchart, determine the project stage of each piece of multi-source project data according to the standard project flowchart, and generate a project stage label for each piece of multi-source project data; and an assessment unit, configured to input the multi-source project data and the project stage label into an assessment model, and output an assessment result from the assessment model, wherein the assessment model is pre-optimized based on stage weights, and the assessment model is used to analyze the input data using fuzzy set theory, and the resulting project status probability distribution is used as the assessment result.

[0014] Furthermore, the designated data source includes at least: project management software, resource planning system, financial software, and IoT sensors. The project-related data includes at least: basic attribute data, schedule data, resource data, and risk data. The basic attribute data includes: project type, project size, project budget, geographical location, and team composition. The schedule data includes: planned execution time period information, actual execution time period information, execution percentage, and execution path of project sub-tasks. The resource data includes: manpower input hours, equipment utilization rate, material consumption, and cost expenditure. The risk data includes: identified risk events, risk categories, scope of impact, and countermeasures taken.

[0015] Furthermore, the project status assessment device further includes a preprocessing unit, used to preprocess the multi-source project data after obtaining the multi-source project data, wherein the preprocessing operation includes at least: data cleaning, missing value imputation, outlier removal, and data standardization.

[0016] Furthermore, the project status evaluation device further includes: an optimization unit, used to optimize the evaluation model based on stage weights, the optimization unit including: a definition module, used to define a stage weight function according to the target project, wherein the stage weight function is used to determine the weight value corresponding to the project stage in which the target project is located; a selection module, used to select a machine learning model and select a model loss function; an introduction module, used to introduce the stage weight function into the model loss function, so that the stage weight of the project stage shows a positive growth with the loss gradient in the model loss function; and a training module, used to train the machine learning model using the model loss function after introducing the stage weight function until a preset convergence condition is reached to obtain the evaluation model.

[0017] Further, the evaluation unit includes: a determination module, used to determine M project states corresponding to the target project based on the input data by the evaluation model, and to determine the project state interval corresponding to each project state, wherein M is an integer greater than 2; a calculation module, used to calculate the membership degree of the input data for each project state interval, to obtain a membership degree vector composed of M membership degrees, wherein the sum of the M membership degrees is 1, and the membership degree vector is used to characterize the probability distribution of the project state; and a generation module, used to generate the evaluation result based on the membership degree vector.

[0018] Furthermore, the generation module includes a determination submodule, used to determine the project state interval corresponding to the maximum membership degree as the evaluation result based on the maximum membership degree principle.

[0019] Furthermore, the generation module further includes: a first generation submodule, configured to, when the maximum membership value is greater than a preset confidence threshold, retrieve the decision plan for the project state interval corresponding to the maximum membership value, and generate the evaluation result based on the project state interval and the decision plan; and a second generation submodule, configured to, when the maximum membership value is less than or equal to the preset confidence threshold, calculate the decision tendency based on all the membership values, and call the corresponding tendency suggestion plan based on the decision tendency, and generate the evaluation result based on the decision tendency and the tendency suggestion plan.

[0020] To achieve the above objectives, according to another aspect of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the project status evaluation method described in any one of the above claims.

[0021] To achieve the above objectives, according to another aspect of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the project status evaluation method described in any of the above.

[0022] To achieve the above objectives, according to another aspect of this application, a computer program product is also provided, including computer instructions, wherein when the computer instructions are executed by a processor, they implement the steps of the project status evaluation method described in any one of the above claims.

[0023] This invention proposes a method for evaluating project status. First, project-related data of the target project is collected periodically from N specified data sources to obtain multi-source project data, where N is a positive integer. Then, a standard project process table is invoked, and the project stage of each piece of multi-source project data is determined according to the standard process table, and a project stage label is generated for each piece of multi-source project data. Finally, the multi-source project data and project stage labels are input into an evaluation model, which outputs the evaluation result. The evaluation model is pre-optimized based on stage weights and is used to analyze the input data using fuzzy set theory. The project status probability distribution obtained from the analysis is used as the evaluation result.

[0024] This invention employs an integrated approach of multi-source heterogeneous data acquisition and intelligent staged annotation. By combining big data analysis with a machine learning model driven by fuzzy logic, it achieves the goal of accurately predicting project progress and risks, thereby realizing the technical effect of intelligent dynamic decision support. This solves the technical problem in related technologies that have limitations in processing data throughout the entire project lifecycle, cannot adapt to complex changes in project circumstances, and result in low accuracy in assessing project progress and risks.

[0025] Specifically, this invention periodically collects multi-dimensional information about target projects from multiple specified data sources to ensure the comprehensiveness and real-time nature of the data. Subsequently, each data sample is precisely labeled in stages according to standardized project management processes to enhance the structure of the data. This invention also proposes an evaluation model based on stage weight optimization. This model ensures the sensitivity and adaptability of the model to different project stages during training by dynamically introducing stage weights into the loss function. It can learn and predict more focusedly at the key stages of the project. The evaluation model can also output the probability distribution of the project status by combining fuzzy set theory, rather than the traditional single deterministic value, which greatly enriches the decision-making information and improves the comprehensive understanding of the project status. Finally, by inputting multi-source project data and project stage labels into the optimized evaluation model, the prediction results of project progress and risks can be fed back in real time and accurately. Based on the uncertainty of the prediction, more flexible and targeted decision adjustments can be made, which significantly improves the intelligence level and decision-making efficiency of project management. Attached Figure Description

[0026] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0027] Figure 1 shows a hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for assessing project status;

[0028] Figure 2 is a flowchart of an optional project status evaluation method according to an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram of an optional project status evaluation device according to an embodiment of the present invention;

[0030] Figure 4 is a structural block diagram of an electronic device for evaluating the status of an execution project according to an embodiment of the present invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0033] To facilitate understanding of the present invention by those skilled in the art, some terms or nouns involved in the various embodiments of the present invention are explained below:

[0034] Project schedule refers to the planned start and end dates of various activities within a project and the logical relationships between them. It is used to track the actual progress of the project and compare it with the planned schedule to determine whether the project is proceeding according to the predetermined timeline.

[0035] Risk prediction refers to the process of analyzing historical data and the current project status to predict the risk events, their probability of occurrence, and the scope of their impact that a project may encounter during its execution.

[0036] Machine learning algorithms are a class of algorithms that enable computers to learn from data. In this invention, an evaluation model is constructed.

[0037] The Weighted Loss Function (WRF) is a variant of the loss function used in machine learning model training to quantify prediction errors and guide model parameter adjustments. Unlike the standard loss function, the WRF assigns different weights to each prediction error. In this invention, the weights depend on the stage to which the project data belongs, thereby optimizing the model's learning focus and improving the accuracy of predictions for key stages.

[0038] Fuzzy Set Theory is a mathematical theory used to handle and describe the fuzziness and uncertainty in real life. Unlike the traditional concept of sets, the degree to which an element belongs to a fuzzy set is continuous, represented by membership degree, which ranges between 0 and 1. This invention utilizes fuzzy set theory to predict the probability distribution of project states, rather than traditional binary (yes / no) or discrete classification predictions, making the prediction results closer to the complexity of actual situations.

[0039] Membership degree, in fuzzy set theory, is a numerical value representing the degree to which an element belongs to a certain fuzzy set. In this invention, membership degree is used to measure the probability that specific project data belongs to different state intervals (such as early completion, on schedule, and delayed completion), and is an important component of the project state probability distribution.

[0040] Real-time prediction refers to the use of pre-trained models to quickly analyze current or recent data to provide immediate predictions of project progress and risks.

[0041] Decision support refers to the information and tools provided to project managers to help them make decisions based on forecasts in order to optimize project management and control.

[0042] It should be noted that the project status assessment method and apparatus in this application can be used in the field of artificial intelligence technology for intelligent risk and schedule prediction of project life cycle management, and can also be used in any field other than artificial intelligence technology for intelligent risk and schedule prediction of project life cycle management. This application does not limit the application field of the project status assessment method and apparatus.

[0043] It should be noted that all relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, and displayed data) involved in this application are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, processing, transmission, provision, disclosure, use, and handling of such data comply with the laws, regulations, and standards of the relevant regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse access. For example, this system has interfaces with relevant users or organizations. Before obtaining relevant information, a request to obtain the information needs to be sent to the aforementioned user or organization through the interface, and the relevant information is obtained only after receiving consent from the aforementioned user or organization.

[0044] The information collection (e.g., user voice, video, and text collection) and analysis operations involved in this application have provided users with corresponding operation entry points during execution, allowing users to choose to agree to or reject the automated decision results; if the user chooses to reject, the process will proceed to the expert decision-making process.

[0045] The following embodiments of the present invention can be applied to various systems / applications / devices that require project schedule prediction and risk assessment, enabling an intelligent auxiliary system for project management based on stage weight optimization and dynamic decision-making. The present invention uses a machine learning algorithm for stage weight optimization to analyze and predict multi-source project data, and then combines fuzzy set theory to output the probability distribution of project status, which can better reflect project uncertainty and provide dynamic decision support.

[0046] This invention also ensures that the system can respond quickly to any changes in the project status by monitoring changes in project data in real time and dynamically adjusting decision recommendations, providing timely warnings of potential risks and giving appropriate countermeasures, making the project management process more efficient and robust.

[0047] The present invention will now be described in detail with reference to various embodiments.

[0048] Example 1

[0049] According to an embodiment of the present invention, an embodiment of a method for evaluating project status is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0050] The project status assessment method embodiment provided in Embodiment 1 of the present invention can be executed in a mobile terminal, computer terminal, or similar computing device. Figure 1 shows a hardware structure block diagram of a computer terminal (or mobile device) for implementing the project status assessment method. As shown in Figure 1, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that the structure shown in Figure 1 is only illustrative and does not limit the structure of the above-described electronic device. For example, the computer terminal 10 may also include more or fewer components than shown in Figure 1, or have a different configuration than shown in Figure 1.

[0051] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0052] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the project status evaluation method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned project status evaluation method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0053] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0054] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0055] In the aforementioned operating environment, this invention provides a project status assessment method as shown in Figure 2. The main implementation body of this method is an intelligent project management auxiliary decision-making system, which combines big data analysis and artificial intelligence technology for progress prediction and risk assessment scenarios during project execution. In particular, it addresses the problems of low accuracy and poor adaptability of traditional prediction methods in complex project environments. Through stage weight optimization model training and dynamic decision support based on maximum membership, the method includes data collection and stage labeling, model training and optimization based on stage weights, fuzzy prediction and dynamic decision support, system integration and real-time services, etc., to significantly improve the accuracy of project progress prediction, enhance the reliability of risk assessment, and provide intelligent and dynamic decision support. This effectively copes with complex and ever-changing project environments and improves the efficiency and quality of project management.

[0056] The embodiments of the present invention will now be described in detail with reference to each specific step.

[0057] Figure 2 is a flowchart of an optional project status assessment method according to an embodiment of the present invention. As shown in Figure 2, the method includes the following steps:

[0058] Step S201: Periodically collect project-related data of the target project from N specified data sources to obtain multi-source project data, where N is a positive integer.

[0059] In this embodiment of the invention, regular data collection ensures that the data in the project management information system remains up-to-date, which helps in accurately predicting project progress and assessing risks. It also facilitates the formation of time-series data, providing a data foundation for dynamically analyzing project status trends. The frequency of regular data collection can be set based on project management needs and the actual data update speed. For example, for short-term, rapidly iterating projects, data can be collected daily or weekly; for long-term or large projects, monthly or quarterly. Key factors considered when setting the regular collection period include project characteristics, available resources, and the cost and efficiency of data collection.

[0060] In one optional embodiment, the designated data source includes at least: the designated data source should include project management software, resource planning and control system, financial management system and possible on-site monitoring system, etc. In addition to the above, it may also include industry database, supplier information system, market intelligence platform, social media data, weather forecast service, etc., to capture information that may affect the progress of the project, such as industry trends, supply chain changes, market reactions, and external environmental factors.

[0061] In the context of this invention, the target project refers to a specific project instance that needs to be monitored, analyzed, and predicted. It can be a project in any industry that involves scheduling, resource allocation, and risk management in project management. For example, the target project can be a software development project, a construction project, a financial product development project, a logistics and distribution plan, a marketing campaign, a research project, etc.

[0062] Furthermore, the basic attribute data in project-related data refers to project type, scale, budget, and participant information. Schedule data refers to planned and actual start / end times, completion percentages, and critical path nodes. Resource data refers to the input and consumption of human, material, and financial resources, such as work hour records, equipment usage statistics, and cost expenditure details. Risk data refers to risk identification, classification, probability of occurrence, impact estimation, and response strategies. It should be noted that project-related data will differ between different projects because different types of projects have specific data requirements. For example, software development projects focus more on technical indicators such as code commit records, defect density, and test coverage; while construction projects emphasize on-site management information such as material supply, construction progress, and safety inspection records.

[0063] Multi-source project data refers to comprehensive datasets about the same project collected from multiple different perspectives and sources. This data integration provides a more comprehensive and in-depth view of the project, helping to uncover correlations and patterns that a single data source might not reveal. Utilizing multi-source project data allows for the construction of more accurate and reliable predictive models, as the multi-dimensional information it contains helps the model capture complex project characteristics and external influences. Simultaneously, multi-source data also helps improve the robustness of prediction results, reducing prediction errors caused by anomalies or missing data sources.

[0064] Optionally, in the project status assessment method provided in this embodiment of the invention, the specified data sources include at least: project management software, resource planning system, financial software, and IoT sensors. The project-related data includes at least: basic attribute data, schedule data, resource data, and risk data. The basic attribute data includes: project type, project size, project budget, geographical location, and team composition. The schedule data includes: planned execution time period information, actual execution time period information, execution percentage, and execution path of project sub-tasks. The resource data includes: manpower input hours, equipment utilization rate, material consumption, and cost expenditure. The risk data includes: identified risk events, risk categories, scope of impact, and countermeasures taken.

[0065] In one optional embodiment, pre-specifying the data source refers to selectively collecting information from a series of pre-defined data systems during project management to ensure that the acquired data is directly relevant and covers key areas of project management. The aforementioned project management software primarily provides detailed records of project tasks, schedules, and resource allocation, serving as a core tool for project progress monitoring and resource planning. The resource planning system tracks and predicts the allocation and use of resources in the project, including human resources, equipment, and materials, and is an important source for resource optimization and cost control. Financial software records project financial information, such as budget allocation, cost expenditures, and revenue status, for managing project budgets and controlling costs. IoT sensors are responsible for collecting real-time physical environment data related to project execution, such as temperature, humidity, and equipment operating status at the construction site, for monitoring changes in project environmental factors.

[0066] Another point to note is that the basic attribute data in the project-related data reflects the essential characteristics of the project, including project type (e.g., IT development, construction), scale (e.g., budget size, number of participants), budget, geographical location, and team composition. This information is fundamental to understanding the project background and developing management strategies. Schedule data covers the timeline and actual execution of project tasks, including the planned execution time period, actual execution time period, percentage of execution, and critical path analysis for sub-tasks. This information is a key indicator for assessing whether the project is progressing as planned. Resource data records in detail the usage of various resources in the project, including man-hours, equipment utilization, material consumption, and project cost expenditures. This information helps in the rational allocation of resources and cost control. Risk data not only includes identified risk events and their categories but also covers the scope of potential risks and the countermeasures already taken, helping the implementing entity to fully understand the risk situation and prepare contingency plans in advance.

[0067] In addition to the aforementioned data sources, data from customer relationship management systems, supply chain management systems, and human resource management systems can be integrated to obtain more comprehensive project operation information and business environment information. Furthermore, by monitoring project-related social media discussions and news reports, unstructured information such as external evaluations and market trends can be collected to assist in risk assessment. Market research and industry reports can also be used to understand industry trends, competitor activities, and regulatory changes. Integrating opinions from project management experts and databases of past projects, especially historical data from similar projects, can provide additional training samples for model training, improving the generalization ability of predictive models. Blockchain technology can be used to collect and verify information from multiple decentralized data sources, ensuring data authenticity and immutability, and improving data credibility and integrity in projects involving multi-party collaborations.

[0068] Optionally, in the project status assessment method provided in the embodiments of the present invention, after obtaining multi-source project data, the project status assessment method further includes: preprocessing the multi-source project data, wherein the preprocessing operation includes at least: data cleaning, missing value imputation, outlier removal, and data standardization.

[0069] In the project status assessment method provided in this invention, preprocessing the acquired multi-source project data aims to improve data quality and consistency, thereby ensuring the accuracy and effectiveness of subsequent analysis. Data cleaning aims to remove noise and irrelevant information from the data, such as deleting duplicate records, correcting formatting errors, and removing meaningless fields. In specific implementation scenarios, data cleaning may include removing HTML tags, repairing damaged files, and standardizing date formats. Specifically, regular expressions or specialized data cleaning libraries can be used to identify and process data items that do not conform to the expected format, and to delete completely irrelevant columns or rows.

[0070] Additionally, project data may contain missing values ​​in certain fields. During preprocessing, missing value imputation can be used to fill these gaps, preventing interference or errors during model training. Common methods include mean / median imputation, forward / backward imputation, and imputation based on statistical relationships within the dataset. In this specific implementation, relevant features can be used to predict missing values, or missing values ​​can be replaced with the mean, mode, or median of the column. For time series data, time windows can be used to imput missing data points forward or backward.

[0071] Outlier removal is another operation. It's important to note that outliers are data points that significantly deviate from the rest of the dataset, potentially distorting the analysis results. Outlier removal can be performed using simple statistical methods or more complex machine learning algorithms. In this specific embodiment, a normal range for the data can be defined first, and then statistical thresholds or distance-based algorithms can be used to identify and remove data points that exceed this range.

[0072] Furthermore, data standardization is the process of transforming data from different scales to the same scale to ensure comparability between different features. Common standardization methods include min-max scaling, Z-score standardization, and decimal scaling normalization. In this specific implementation, data standardization can be achieved by subtracting the mean and dividing by the standard deviation (Z-score standardization) or scaling to a fixed range (such as 0 to 1, min-max scaling), ensuring that all features are compared and processed on the same scale.

[0073] In addition to the steps mentioned above, preprocessing may also include the following supplementary steps:

[0074] Feature engineering involves creating new features or transforming existing features based on project characteristics to extract more valuable information. For example, lagged features can be extracted from time series data, or cross features can be created to represent the interactions between different data sources.

[0075] Data fusion: Merging similar data from different data sources to eliminate data redundancy and improve data consistency and integration.

[0076] Metadata analysis and governance: Understanding the source, quality, meaning, and usage of data, and implementing data governance strategies, such as data dictionary maintenance and data quality auditing, to ensure the effective use and compliance of data.

[0077] Privacy Protection and De-identification: When processing project data involving personal or sensitive information, data de-identification and de-identification are performed to comply with laws and regulations and protect the privacy of participants.

[0078] Through the preprocessing operations described above, this embodiment ensures that the data used in the project status assessment process is as accurate, consistent, and meaningful as possible, providing a solid foundation for subsequent predictive models and decision support. Furthermore, the application of advanced preprocessing techniques such as feature engineering and data fusion can further uncover deeper information behind the data, enhancing the depth and breadth of the analysis.

[0079] Step S202: Call the project standard process table, determine the project stage of each multi-source project data according to the project standard process table, and generate project stage labels for each multi-source project data.

[0080] Specifically, a standard project flowchart is a reference framework used in project management to define the various stages and sequence of a project from initiation to completion. It assigns specific stage names and descriptions to each step based on project management knowledge systems or specific industry standards. In terms of content, a standard project flowchart includes definitions, key activities, expected deliverables, and transition conditions for each stage. For example, a standard project flowchart might list the following stages: Initiation, Planning, Execution, Monitoring and Controlling, and Closure.

[0081] In one specific implementation, for a software development project, project phases may include requirements analysis, design, coding, testing, and deployment. For a construction project, project phases may include preliminary design, detailed design, construction, quality inspection, and project delivery. While all projects go through similar initiation, planning, execution, monitoring, and closure phases, the specific activities and sub-phases within these major phases will differ across industries or project types, reflecting the project's unique processes and requirements.

[0082] After invoking the project's standard workflow table, the system of this invention will match the characteristics of the project data (such as task name, key milestones, timestamps, etc.) with the stage definitions in the standard workflow table to determine the project stage to which each data record belongs. ,in, M represents the total number of phases. For example, if a data record shows that a task is named "Requirements Gathering," the implementation system will mark the task and related data as belonging to the project initiation phase.

[0083] Step S203: Input the multi-source project data and project stage labels into the evaluation model, and output the evaluation results from the evaluation model. The evaluation model is pre-optimized based on stage weights. The evaluation model is used to analyze the input data in combination with fuzzy set theory, and the project state probability distribution obtained from the analysis is used as the evaluation result.

[0084] It's important to note that the evaluation model can be a machine learning-based model, such as support vector machines, random forests, gradient boosting trees, deep neural networks, etc., with the specific architecture depending on the model type and the complexity of the prediction task. For example, a deep neural network can contain multiple hidden layers to learn complex patterns in the data, with each layer consisting of multiple neurons performing weighted summations and nonlinear transformations. During the model training phase, a stage weight function is used... Adjusting the loss function ensures that the model pays more attention to the stages that have a greater impact on the final outcome of the project. This optimization method can improve the model's sensitivity to key stages, thereby improving prediction accuracy.

[0085] In this embodiment of the invention, fuzzy set theory is used to address the uncertainty of the model output, ensuring that the model outputs a probability distribution of project states, rather than just a single predicted value. This probability distribution reflects the model's assessment of multiple possible states and the likelihood of each state. In this embodiment, fuzzy set theory allows the model to handle the fuzziness and uncertainty of information, defining the degree to which a data point belongs to a certain fuzzy set through a membership function. This set is not limited to black-and-white classifications. In project state assessment, fuzzy set theory is used to describe the uncertainty of project progress, cost, risk, and other states. For example, a project may not be completely low-risk or high-risk, but rather in a fuzzy state between medium and high risk.

[0086] The evaluation model outputs a probability distribution of project states, providing the likelihood of each possible project state (e.g., early completion, on-time completion, delayed completion). Based on multi-source project data (including but not limited to schedule data, resource data, and risk data), combined with project phase annotations and model parameters obtained through training, the evaluation model comprehensively analyzes the current project state. By comparing planned and actual progress, resource consumption and budget, identified risks and corresponding countermeasures, it assesses the overall health of the project and outputs a distribution describing possible states and the degree of uncertainty based on fuzzy set theory.

[0087] Combining fuzzy set theory with evaluation models can handle the uncertainty in project status assessment and avoid overly arbitrary judgments. For example, for a schedule data point, a traditional model would give a clear prediction of "delay" or "on time," while a model based on fuzzy set theory would output a probability distribution that covers "slight delay," "on time," and "significantly ahead of schedule." This result is more in line with the complexity and uncertainty of the real world.

[0088] Optionally, in the project status evaluation method provided in this embodiment of the invention, the step of optimizing the evaluation model based on stage weights includes: defining a stage weight function according to the target project, wherein the stage weight function... Used to determine the project stage of the target project. The corresponding weight values; select the machine learning model and the model loss function; introduce the stage weight function into the model loss function so that the stage weights of the project stages in the model loss function show a positive growth with the loss gradient; use the model loss function after introducing the stage weight function to train the machine learning model until the preset convergence condition is reached to obtain the evaluation model.

[0089] The implementer of this invention can define a stage weighting function based on project characteristics, stage importance, and degree of impact. This includes, but is not limited to, determining the actual contribution of different stages to the final project outcome through analysis of past project data. The stage weighting function can be determined by various factors, such as the duration of the project stage, the number of key tasks within the stage, and the correlation between the stage and the project outcome. The weight value is expressed numerically, reflecting the relative importance of the stage within the project. Linear growth, exponential growth, or a user-defined function can be used in the calculation. When it is determined that later stages of the project (such as execution, monitoring, and control) have a greater direct impact on the project outcome, higher weights are ensured for these later stages.

[0090] An optional stage weight function can be defined as: ,in, It is a stage number. It is the total number of stages. It is a normalization coefficient that ensures the sum of weights is a reasonable value (e.g., the sum is 1). It is an adjustable parameter greater than 0, used to control the steepness of weight growth, by adjusting... It allows you to control how much attention the model pays to later stages.

[0091] The implementing entity can choose from various models, such as linear regression models, whose architecture includes an input layer, one or more linear combination layers, and an output layer; support vector machines, which can find the optimal hyperplane for data classification based on the maximum margin principle; deep neural networks, which contain multi-layered neural network architectures, such as convolutional neural networks or recurrent neural networks; and ensemble learning models, such as gradient boosting decision trees and random forests. The selection criteria may include the model's predictive power, ability to handle complex data, training speed, and model size.

[0092] Furthermore, for classification tasks, optional loss functions include cross-entropy loss, which measures the difference between the model's predicted probability distribution and the true distribution; for regression tasks, the loss function can be mean squared error (MSE), used to calculate the average squared difference between predicted and true values, or root mean square error (RMSE), which is more sensitive to error penalty. The selection criteria include the type of prediction task, the characteristics of the data distribution, and the sensitivity to error.

[0093] For a selected model loss function, the implementer can multiply each sample (or the sum of data for each time period) by the weight value of the stage to which that data point belongs when calculating the loss. This indicates that during model training, data from important stages will have a greater impact on gradient descent. This operation allows the stage weights of project stages to exhibit a positive correlation with the loss gradient in the model loss function; that is, as the project stage progresses, the stage weights increase, and during model training, the prediction errors of later stages have a greater impact on the updating of model parameters.

[0094] Taking regression tasks as an example, using mean squared error (MSE) as the loss function, the weighted loss function is: It is the total number of samples. It is the actual value (e.g., the actual construction period). These are model predictions (e.g., predicted project duration). These are the weights corresponding to the project stage to which the target sample belongs. In this way, when the model performs backpropagation and parameter updates, the sample errors from high-weight stages (such as the execution stage) will generate larger gradients, forcing the model to prioritize learning the feature patterns of these key stages, thereby improving the model's ability to predict the final project results.

[0095] This invention employs a loss function incorporating stage weights, adjusting model parameters through backpropagation algorithms and optimizers (such as stochastic gradient descent (SGD), Adam, etc.) to minimize the weighted loss. Training continues until the model reaches preset convergence conditions, such as the loss function no longer decreasing significantly, reaching the maximum number of iterations, or performance on the validation set no longer improving. Convergence can be determined by the loss function stabilizing within a certain threshold and no longer changing significantly; or by the prediction performance on the validation set reaching predetermined performance metrics (such as accuracy, F1 score, etc.). Preset convergence conditions typically combine the stability of the loss function with an evaluation of the model's generalization ability, ensuring that the model performs well not only on training data but also makes accurate predictions on unknown data.

[0096] Optionally, in the project status assessment method provided in this embodiment of the invention, the step of inputting multi-source project data and project stage labels into the assessment model, and then having the assessment model output the assessment result, includes: the assessment model determining M types of project status corresponding to the target project based on the input data, and determining the project status interval corresponding to each project status, where M is an integer greater than 2; calculating the membership degree of the input data for each project status interval to obtain a membership degree vector composed of M membership degrees, where the sum of the M membership degrees is 1, and the membership degree vector is used to characterize the probability distribution of the project status; and generating the assessment result based on the membership degree vector.

[0097] It's important to note that, based on the input multi-source project data and project stage annotations, the trained evaluation model can automatically identify and predict the M possible states of the target project through its internal parameters and rules. These states are determined based on the objectives set during model training. For example, project states include, but are not limited to: ahead of schedule, on schedule, minor delay, severe delay, good cost control, cost exceeding budget, low risk, medium risk, and high risk. These states are inherently related; for example, project delays can affect cost states and even increase the probability of risk states. The project state interval refers to discretizing a continuous numerical range (such as the number of days of project delay, the percentage of cost exceeding the budget, and the impact of risk events) into multiple intervals, each interval corresponding to a project state. For example, project delay states can be divided into three intervals: less than 10 days, 10 to 30 days, and more than 30 days. Furthermore, the input data, i.e., the multi-source project data and project stage annotations, is the foundation for determining the project state intervals. By analyzing the distribution of historical data, interval boundaries can be reasonably set to ensure that the intervals cover the main distribution in the dataset while distinguishing the characteristics of different states.

[0098] Calculating the membership degree of input data for each project state interval requires a membership function based on fuzzy set theory. This function describes the degree to which an input data point belongs to a certain interval (or state). Each calculated membership degree represents the correlation or degree of belonging between the input data and a specific project state. For example, a high membership degree for the "severely delayed" state indicates that the project is experiencing significant time delays. The membership function can be a trigonometric function, trapezoidal function, or Gaussian function, etc. For each project state interval, the implementing entity can use the trained membership function to map the input data to a value between [0,1]. This value represents the probability or degree to which the data point belongs to that state interval. The sum of the membership degrees for all M state intervals needs to be normalized to 1 to maintain the properties of the probability distribution. The M-dimensional membership vector composed of all membership degrees serves as the evaluation result, providing not only the predicted probability of each state but also reflecting the relative likelihood between states. This distributed output helps project managers comprehensively understand the current state of the project, as well as the risks and opportunities associated with different states.

[0099] In an alternative implementation, in classification tasks (risk levels: low, medium, high), instead of using Softmax to output the probability of a single class, a vector containing the membership degrees of all risk levels is output. ,in, Represents membership degree, and For regression tasks (such as project schedule prediction), continuous project schedule values ​​can be discretized into multiple intervals (more than 10% ahead of schedule, on schedule, less than 10% behind schedule, 10%-20% behind schedule, and more than 20% behind schedule). The model output can also be used to determine the membership vector of the above intervals.

[0100] The above steps not only enable accurate prediction of project progress and risks, but also provide project managers with flexible and timely decision support based on fuzzy prediction results with maximum membership degree and dynamic decision-making logic, significantly improving the efficiency and effectiveness of project management.

[0101] Optionally, in the project status evaluation method provided in the embodiments of the present invention, the step of generating evaluation results based on membership vectors further includes: determining the project status interval corresponding to the maximum membership value as the evaluation result according to the principle of maximum membership.

[0102] It should be noted that the maximum membership degree principle is a decision criterion widely used in fuzzy logic and fuzzy set theory. This principle states that, given the fuzzy set membership degree distribution of an object (such as project status), the fuzzy set with the largest membership degree value should be selected as the category or status description of the object, and this should be used for decision-making or classification. That is, the category or interval with the highest membership degree is selected as the final prediction result. In this embodiment of the invention, the maximum membership degree principle is applied to evaluate the membership degree vector output by the model to determine which state interval the item is most likely to be in.

[0103] In the context of project status assessment, the maximum membership principle helps the inventors of this invention make decisions from a series of possible project states, determining the most likely state range for the current project. This process simplifies the decision-making process and ensures that decisions are based on the most likely project state, improving the relevance and effectiveness of the decisions. However, it is worth noting that the maximum membership principle only provides a point estimate (i.e., the most likely project state) and cannot fully reflect the uncertainty of the prediction results. Therefore, in practical applications, the inventors of this invention can also combine other principles of fuzzy set theory (such as the centroid principle or the weighted average principle) to supplement and refine the decision-making process to obtain a more comprehensive overview of the project status.

[0104] The evaluation result is selected from the project state interval corresponding to the maximum value in the membership vector, indicating that this interval is the state that the model believes the project is most likely to be in. For example, if the membership vector output by the model shows that the "slight delay" state has the highest membership, then this state becomes the evaluation result, indicating that the project manager is at risk of a slight delay in the current project and needs to pay attention to relevant factors and take appropriate management measures.

[0105] The above steps establish a project schedule and risk prediction system based on stage weight optimization and dynamic decision-making. This system overcomes the shortcomings of existing technologies, such as low model training efficiency, failure of prediction results to reflect uncertainty, and rigid decision support. By integrating multi-source heterogeneous project data, optimizing the machine learning model training process, and introducing fuzzy set theory for dynamic prediction and decision-making, the system significantly improves the intelligence level of project management. It helps project managers make more accurate and flexible decisions based on real-time and comprehensive project status assessment information, thereby optimizing project schedule control, cost management, and risk response, and improving the overall success rate and economic benefits of the project.

[0106] Optionally, in the project status assessment method provided in this embodiment of the invention, the step of generating assessment results based on membership vectors further includes: when the maximum membership value is greater than a preset confidence threshold, retrieving the decision plan for the project status interval corresponding to the maximum membership value, and generating assessment results based on the project status interval and the decision plan; or, when the maximum membership value is less than or equal to a preset confidence threshold, calculating the decision tendency based on all membership degrees, and calling the corresponding tendency suggestion plan based on the decision tendency, and generating assessment results based on the decision tendency and the tendency suggestion plan.

[0107] It should be noted that a pre-set reliability threshold is used. In project status assessment, its role is to determine the reliability or confidence level of the model's prediction results. When the maximum membership value exceeds this threshold (…),… This indicates that the model has a high confidence level in predicting the state of a certain project, and the prediction results are relatively clear and reliable; conversely, when the maximum membership degree is lower than or equal to the threshold ( A confidence threshold of 0.8 indicates significant uncertainty in the model's predictions, requiring more cautious handling. The basis for setting the confidence threshold typically comes from experience, prior knowledge, or statistical analysis, reflecting the level of prediction uncertainty that decision-makers are willing to accept. For example, if project management demands extremely high accuracy in predictions, a higher confidence threshold, such as 0.8, is needed; conversely, if higher demands are placed on real-time predictions and response speed, the confidence threshold can be lowered to facilitate rapid decision-making.

[0108] To further explain, a maximum membership degree exceeding a pre-set confidence threshold indicates that the model's predictions show a high level of confidence in a particular project state. In this case, the implementation system can retrieve pre-prepared decision-making plans, providing specific response strategies or action plans for that state range. For example, if the maximum membership degree corresponds to the "high risk of resource shortage" state, the high risk of resource shortage plan can be invoked to guide project managers to increase resource allocation, optimize resource configuration, or make advance procurements to mitigate the negative impact of resource shortages.

[0109] Decision-making plans are developed before project commencement or at least in the early stages of project execution. These plans are formulated by the project management team based on project characteristics, industry experience, historical data, and risk management strategies. The plans include specific response measures, responsibility allocation, and action timelines, enabling swift action once a corresponding situation is predicted. The process of invoking these plans is automated and condition-driven. The evaluation model determines when a specific plan needs to be invoked based on input data and project status intervals. The system then automatically retrieves and provides that plan, and can also automatically execute some or all of its measures to quickly respond to changes in project status.

[0110] To further clarify, a maximum membership degree less than or equal to a pre-set confidence threshold indicates high uncertainty in the model's predictions, requiring additional consideration before making a decision. In this case, the implementation system can calculate a decision propensity score based on all membership degrees, serving as an indicator that synthesizes all possible states and helps decision-makers understand the overall trend of the project's status. The decision propensity score can be calculated using a membership vector weighted average or other weighted aggregation techniques. Recommended contingency plans are also prepared in advance during the project planning phase, but they focus more on guiding direction and strategy selection rather than specific action details. These plans help decision-makers make decisions based on overall trends and risk preferences when facing uncertainty. For example, if the decision propensity score indicates a risk of project delays and resource shortages, the plan might suggest increasing contingency reserves, advance procurement, or optimizing resource allocation, but it won't specify which resources to adjust or when.

[0111] The above steps establish an intelligent, flexible, and reliable project status assessment and decision support system. By integrating multi-source data, optimizing the prediction model, and introducing fuzzy logic and dynamic decision-making mechanisms, the system significantly improves the efficiency and effectiveness of project management. This invention not only provides clear prediction results but also offers corresponding decision-making plans and propensity suggestions based on the confidence level of the prediction results and the project's uncertainties. This helps project managers formulate more scientific and targeted project management strategies based on real-time prediction information and historical experience.

[0112] When generating evaluation results, the implementation system of this invention first checks whether the maximum value of the membership degree exceeds the confidence threshold. If it does, a clear decision plan is provided based on the state interval corresponding to the maximum membership degree; if it does not exceed the threshold, a decision tendency is provided, and a suggested plan based on the tendency is invoked to guide project managers on how to balance risks and opportunities in an uncertain environment. The binary decision-making process based on the confidence threshold effectively combines the accuracy of model predictions with the actual needs of the project, providing strong support for project management decisions.

[0113] Through the above steps S201 to S203, project-related data of the target project can be collected periodically from N specified data sources to obtain multi-source project data, where N is a positive integer. Then, the standard project process table is called, and the project stage of each piece of multi-source project data is determined according to the standard project process table. Project stage labels are generated for each piece of multi-source project data. Finally, the multi-source project data and project stage labels are input into the evaluation model, and the evaluation model outputs the evaluation results. The evaluation model is pre-optimized based on stage weights. The evaluation model is used to analyze the input data in combination with fuzzy set theory, and the project state probability distribution obtained from the analysis is used as the evaluation result.

[0114] In this embodiment of the invention, an integrated approach of multi-source heterogeneous data acquisition and intelligent staged annotation is adopted. By combining big data analysis and fuzzy logic-driven machine learning models, the goal of accurately predicting project progress and risks is achieved, thereby realizing the technical effect of intelligent dynamic decision support. This solves the technical problem in related technologies that have limitations in processing project lifecycle data, cannot adapt to complex project changes, and result in low accuracy in assessing project progress and risks.

[0115] Specifically, this invention collects multi-dimensional information about target projects periodically from multiple specified data sources to ensure the comprehensiveness and real-time nature of the data. Subsequently, each data sample is precisely labeled in stages according to the standardized process of project management, enhancing the structure of the data. This invention also proposes an evaluation model based on stage weight optimization. This model ensures the sensitivity and adaptability of the model to different project stages during training by dynamically introducing stage weights into the loss function. It can learn and predict more focusedly at the key stages of the project. The evaluation model can also output the probability distribution of the project status by combining fuzzy set theory, rather than the traditional single deterministic value, which greatly enriches the decision-making information and improves the comprehensive understanding of the project status. Finally, by inputting multi-source project data and project stage labels into the optimized evaluation model, the prediction results of project progress and risk can be fed back in real time and accurately. Based on the uncertainty of the prediction, more flexible and targeted decision adjustments can be made, significantly improving the intelligence level and decision-making efficiency of project management.

[0116] The invention will now be described in conjunction with another alternative embodiment.

[0117] Example 2

[0118] This invention also provides a project status assessment device. It should be noted that the project status assessment device of this invention includes multiple implementation units, which can be used to execute the project status assessment method provided in the first embodiment above. Each implementation unit corresponds to each implementation step in the first embodiment above.

[0119] Figure 3 is a schematic diagram of an optional project status evaluation device according to an embodiment of the present invention. As shown in Figure 3, the device may include: a data acquisition unit 31, a generation unit 32, and an evaluation unit 33.

[0120] The acquisition unit 31 is used to periodically collect project-related data of the target project from N specified data sources to obtain multi-source project data, where N is a positive integer.

[0121] The generation unit 32 is used to call the project standard process table, determine the project stage of each multi-source project data according to the project standard process table, and generate the project stage label for each multi-source project data.

[0122] Evaluation unit 33 is used to input multi-source project data and project stage labels into the evaluation model, and the evaluation model outputs the evaluation results. The evaluation model is pre-optimized based on stage weights and is used to analyze the input data in combination with fuzzy set theory, and the project state probability distribution obtained from the analysis is used as the evaluation result.

[0123] The aforementioned project status assessment device can first use the acquisition unit 31 to periodically collect project-related data of the target project from N specified data sources to obtain multi-source project data, where N is a positive integer. Then, the generation unit 32 is used to call the project standard process table and determine the project stage of each piece of multi-source project data according to the project standard process table, and generate a project stage label for each piece of multi-source project data. Finally, the assessment unit 33 is used to input the multi-source project data and project stage labels into the assessment model, and the assessment model outputs the assessment result. The assessment model is pre-optimized based on stage weights and is used to analyze the input data in combination with fuzzy set theory, and the project status probability distribution obtained from the analysis is used as the assessment result.

[0124] In this embodiment of the invention, an integrated approach of multi-source heterogeneous data acquisition and intelligent staged annotation is adopted. By combining big data analysis and fuzzy logic-driven machine learning models, the goal of accurately predicting project progress and risks is achieved, thereby realizing the technical effect of intelligent dynamic decision support. This solves the technical problem in related technologies that have limitations in processing project lifecycle data, cannot adapt to complex project changes, and result in low accuracy in assessing project progress and risks.

[0125] Specifically, this invention collects multi-dimensional information about target projects periodically from multiple specified data sources to ensure the comprehensiveness and real-time nature of the data. Subsequently, each data sample is precisely labeled in stages according to the standardized process of project management, enhancing the structure of the data. This invention also proposes an evaluation model based on stage weight optimization. This model ensures the sensitivity and adaptability of the model to different project stages during training by dynamically introducing stage weights into the loss function. It can learn and predict more focusedly at the key stages of the project. The evaluation model can also output the probability distribution of the project status by combining fuzzy set theory, rather than the traditional single deterministic value, which greatly enriches the decision-making information and improves the comprehensive understanding of the project status. Finally, by inputting multi-source project data and project stage labels into the optimized evaluation model, the prediction results of project progress and risk can be fed back in real time and accurately. Based on the uncertainty of the prediction, more flexible and targeted decision adjustments can be made, significantly improving the intelligence level and decision-making efficiency of project management.

[0126] Furthermore, the specified data sources include at least: project management software, resource planning systems, financial software, and IoT sensors. Project-related data includes at least: basic attribute data, schedule data, resource data, and risk data. Among them, basic attribute data includes: project type, project size, project budget, geographical location, and team composition; schedule data includes: planned execution time period information, actual execution time period information, execution percentage, and execution path of project sub-tasks; resource data includes: manpower input hours, equipment utilization rate, material consumption, and cost expenditure; risk data includes: identified risk events, risk categories, scope of impact, and countermeasures taken.

[0127] Furthermore, the project status assessment device also includes a preprocessing unit, used to preprocess the multi-source project data after obtaining the multi-source project data, wherein the preprocessing operations include at least: data cleaning, missing value imputation, outlier removal, and data standardization.

[0128] Furthermore, the project status evaluation device also includes: an optimization unit for optimizing the evaluation model based on stage weights. The optimization unit includes: a definition module for defining a stage weight function according to the target project, wherein the stage weight function is used to determine the weight value corresponding to the project stage in which the target project is located; a selection module for selecting a machine learning model and selecting a model loss function; an introduction module for introducing the stage weight function into the model loss function so that the stage weight of the project stage shows a positive growth with the loss gradient in the model loss function; and a training module for training the machine learning model using the model loss function after introducing the stage weight function until a preset convergence condition is reached to obtain the evaluation model.

[0129] Furthermore, the evaluation unit includes: a determination module, used by the evaluation model to determine M project states corresponding to the target project based on the input data, and to determine the project state interval corresponding to each project state, where M is an integer greater than 2; a calculation module, used to calculate the membership degree of the input data for each project state interval, to obtain a membership degree vector composed of M membership degrees, where the sum of the M membership degrees is 1, and the membership degree vector is used to characterize the probability distribution of the project state; and a generation module, used to generate evaluation results based on the membership degree vector.

[0130] Furthermore, the generation module includes a determination submodule, which is used to determine the project status interval corresponding to the maximum membership degree as the evaluation result based on the principle of maximum membership degree.

[0131] Furthermore, the generation module also includes: a first generation submodule, used to retrieve the decision plan for the project status interval corresponding to the maximum membership degree when the maximum membership degree is greater than a preset confidence threshold, and generate an evaluation result based on the project status interval and the decision plan; and a second generation submodule, used to calculate the decision tendency degree based on all membership degrees when the maximum membership degree is less than or equal to a preset confidence threshold, and to call the corresponding tendency suggestion plan based on the decision tendency degree, and generate an evaluation result based on the decision tendency degree and the tendency suggestion plan.

[0132] It should be noted that the acquisition unit 31, generation unit 32, and evaluation unit 33 mentioned above correspond to steps S201 to S203 in Embodiment 1. The instances and application scenarios implemented by the above units and corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules or units can also be part of a device and can run in the computer terminal 10 provided in Embodiment 1.

[0133] The invention will now be described in conjunction with another alternative embodiment.

[0134] Example 3

[0135] This invention can also provide an electronic device. FIG4 is a structural block diagram of an electronic device for evaluating the status of an execution project according to an embodiment of the present invention. As shown in FIG4, the electronic device may include: one or more (only one is shown in FIG4) processors 702, memory 704, storage controller, and peripheral interfaces, wherein the peripheral interfaces are connected to a radio frequency module, an audio module, and a display.

[0136] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the project status evaluation method and apparatus in this application embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the aforementioned project status evaluation method. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0137] The processor can access information and applications stored in memory via a transmission device to perform the following steps: periodically collect project-related data of the target project from N specified data sources to obtain multi-source project data, where N is a positive integer; call the project standard process table and determine the project stage of each piece of multi-source project data according to the project standard process table, and generate a project stage label for each piece of multi-source project data; input the multi-source project data and project stage labels into the evaluation model, and the evaluation model outputs the evaluation result, wherein the evaluation model is pre-optimized based on stage weights, and the evaluation model is used to analyze the input data in combination with fuzzy set theory, and the project state probability distribution obtained from the analysis is used as the evaluation result.

[0138] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: preprocessing multi-source project data, wherein the preprocessing operations include at least: data cleaning, missing value imputation, outlier removal, and data standardization.

[0139] The processor can also access information and applications stored in the memory via a transmission device to perform the following steps: define a stage weight function based on the target project, wherein the stage weight function is used to determine the weight value corresponding to the project stage in which the target project is located; select a machine learning model and select a model loss function; introduce the stage weight function into the model loss function so that the stage weight of the project stage in the model loss function shows a positive increase with the loss gradient; train the machine learning model using the model loss function after introducing the stage weight function until the preset convergence condition is reached to obtain the evaluation model.

[0140] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: the evaluation model determines the M project states corresponding to the target project based on the input data, and determines the project state interval corresponding to each project state, where M is an integer greater than 2; calculates the membership degree of the input data for each project state interval, and obtains a membership degree vector composed of M membership degrees, where the sum of the M membership degrees is 1, and the membership degree vector is used to characterize the probability distribution of the project state; and generates the evaluation result based on the membership degree vector.

[0141] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: Based on the principle of maximum membership, determine the project state interval corresponding to the maximum membership value as the evaluation result.

[0142] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: when the maximum membership degree is greater than the preset confidence threshold, retrieve the decision plan for the project status interval corresponding to the maximum membership degree, and generate an evaluation result based on the project status interval and the decision plan; or, when the maximum membership degree is less than or equal to the preset confidence threshold, calculate the decision tendency degree based on all membership degrees, call the corresponding tendency suggestion plan based on the decision tendency degree, and generate an evaluation result based on the decision tendency degree and the tendency suggestion plan.

[0143] This invention provides a project status assessment scheme. By integrating multi-source heterogeneous data acquisition with intelligent staged annotation, and combining big data analysis with a fuzzy logic-driven machine learning model, it achieves accurate prediction of project progress and risks. This realizes the technical effect of intelligent dynamic decision support, thereby solving the technical problem of limitations in related technologies when processing project lifecycle data, their inability to adapt to complex project changes, and the resulting low accuracy in project progress and risk assessment.

[0144] Those skilled in the art will understand that the structure shown in Figure 4 is merely illustrative, and the electronic device may also be a smartphone, tablet computer, PDA, mobile internet device (MID), PAD, or other terminal device. Figure 4 does not limit the structure of the aforementioned electronic device. For example, the electronic device may include more or fewer components (such as network interfaces, display devices, etc.) than shown in Figure 4, or may have a different configuration than that shown in Figure 4.

[0145] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0146] The invention will now be described in conjunction with another alternative embodiment.

[0147] Example 4

[0148] This invention also provides a computer-readable storage medium. Optionally, in this invention, the computer-readable storage medium can be used to store the program code executed by the project status evaluation method provided in Embodiment 1.

[0149] Optionally, in this embodiment of the invention, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0150] This invention also provides a computer program product, which, when executed on a data processing device, is suitable for performing the steps of a project status assessment method: periodically collecting project-related data of the target project from N specified data sources to obtain multi-source project data, where N is a positive integer; calling a standard project flowchart and determining the project stage of each piece of multi-source project data according to the standard project flowchart, and generating a project stage label for each piece of multi-source project data; inputting the multi-source project data and project stage labels into an assessment model, and having the assessment model output an assessment result, wherein the assessment model is pre-optimized based on stage weights, and the assessment model is used to analyze the input data using fuzzy set theory, and the project status probability distribution obtained from the analysis is used as the assessment result.

[0151] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0152] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0156] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0157] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for assessing project status, characterized in that, include: Periodically collect project-related data of the target project from N specified data sources to obtain multi-source project data, where N is a positive integer; call the project standard process table, and determine the project stage of each piece of multi-source project data according to the project standard process table, and generate a project stage label for each piece of multi-source project data; input the multi-source project data and the project stage label into the evaluation model, and the evaluation model outputs the evaluation result, wherein the evaluation model is pre-optimized based on stage weights, and the evaluation model is used to analyze the input data in combination with fuzzy set theory, and the project state probability distribution obtained from the analysis is used as the evaluation result.

2. The evaluation method according to claim 1, characterized in that, The designated data sources include at least: project management software, resource planning systems, financial software, and IoT sensors. The project-related data includes at least: basic attribute data, schedule data, resource data, and risk data. The basic attribute data includes: project type, project size, project budget, geographical location, and team composition. The schedule data includes: planned execution time periods, actual execution time periods, execution percentages, and execution paths for project sub-tasks. The resource data includes: manpower input hours, equipment utilization rate, material consumption, and cost expenditures. The risk data includes: identified risk events, risk categories, scope of impact, and countermeasures taken.

3. The evaluation method according to claim 1, characterized in that, After obtaining multi-source project data, the project status assessment method further includes: preprocessing the multi-source project data, wherein the preprocessing operations include at least: data cleaning, missing value imputation, outlier removal, and data standardization.

4. The evaluation method according to claim 1, characterized in that, The steps for optimizing the evaluation model based on stage weights include: defining a stage weight function according to the target project, wherein the stage weight function is used to determine the weight value corresponding to the project stage in which the target project is located; selecting a machine learning model and a model loss function; introducing the stage weight function into the model loss function so that the stage weight of the project stage exhibits a positive growth with the loss gradient in the model loss function; and training the machine learning model using the model loss function after introducing the stage weight function until a preset convergence condition is reached to obtain the evaluation model.

5. The evaluation method according to claim 1, characterized in that, The steps of inputting the multi-source project data and the project stage labels into the evaluation model, and then having the evaluation model output the evaluation result, include: the evaluation model determining M project states corresponding to the target project based on the input data, and determining the project state interval corresponding to each project state, where M is an integer greater than 2; calculating the membership degree of the input data for each project state interval, obtaining a membership degree vector composed of M membership degrees, where the sum of the M membership degrees is 1, and the membership degree vector is used to characterize the probability distribution of the project state; and generating the evaluation result based on the membership degree vector.

6. The evaluation method according to claim 5, characterized in that, The step of generating the evaluation result based on the membership vector includes: determining the project state interval corresponding to the maximum membership degree as the evaluation result according to the principle of maximum membership degree.

7. The evaluation method according to claim 6, characterized in that, The step of generating the evaluation result based on the membership vector further includes: when the maximum membership value is greater than a preset confidence threshold, retrieving the decision plan for the project state interval corresponding to the maximum membership value, and generating the evaluation result based on the project state interval and the decision plan; or, when the maximum membership value is less than or equal to the preset confidence threshold, calculating the decision tendency degree based on all the membership degrees, and calling the corresponding tendency suggestion plan based on the decision tendency degree, and generating the evaluation result based on the decision tendency degree and the tendency suggestion plan.

8. A device for evaluating the status of a project, characterized in that, include: The data collection unit is used to periodically collect project-related data of the target project from N specified data sources to obtain multi-source project data, where N is a positive integer; the generation unit is used to call the project standard process table, determine the project stage of each piece of multi-source project data according to the project standard process table, and generate a project stage label for each piece of multi-source project data; the evaluation unit is used to input the multi-source project data and the project stage label into the evaluation model, and output the evaluation result from the evaluation model, wherein the evaluation model is pre-optimized based on stage weights, and the evaluation model is used to analyze the input data in combination with fuzzy set theory, and the project state probability distribution obtained from the analysis is used as the evaluation result.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the project status assessment method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the project status evaluation method according to any one of claims 1 to 7.

11. A computer program product, characterized in that, Includes computer instructions, wherein when executed by a processor, the computer instructions implement the steps of the method for evaluating the project status as described in any one of claims 1 to 7.

Images