A construction project laborer full life cycle management method and device and medium

Abstract

The application discloses a kind of construction project laborer whole life cycle management method, device and medium, comprising: based on convolutional neural network, the fusion processing of laborer's multi-source heterogeneous data is carried out to Transform network and graph neural network, constructs six-dimensional digital image including basic information, skill information, health information, safety information, performance information and credit information;Based on six-dimensional digital image and full connection neural network model, the post requirement of project and laborer feature are executed man-gang matching analysis, and matching degree sorting result and optimal personnel allocation scheme are generated;Based on matching degree sorting result and optimal personnel allocation scheme, six-dimensional digital image is executed closed loop update, and forms the management link of whole life cycle.

Description

Technical Field

[0001] This invention discloses a method, device, and medium for the full life cycle management of construction project workers, belonging to the field of life cycle digital technology. Background Technology

[0002] Existing methods for managing construction workers often employ rule-driven modular systems or multi-factor scoring-based recommendation matching. These methods achieve process control through modules such as demand analysis, contract management, attendance, and payroll. Some even match and recommend workers based on their professional level and project requirements. However, none of these solutions utilize multiple neural networks to fuse and model the multi-source heterogeneous data of workers, making it difficult to build a dynamically updated unified digital profile. This results in insufficient accuracy in matching people to positions, prominent data silos between different management stages, and a lack of a full lifecycle management chain from personnel profiling to matching configuration and then to closed-loop profile updates.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, device, and medium for the full life-cycle management of construction project laborers, which at least solves the technical problems in existing construction project laborer management methods, such as difficulty in integrating multi-source data, reliance on fixed rules for matching people to positions, and fragmentation of management links that cannot form a dynamic feedback loop.

[0005] According to one aspect of the present invention, in order to achieve the above-mentioned objective, a method for full life-cycle management of construction project laborers is provided, characterized in that it includes:

[0006] Based on convolutional neural networks, Transformer networks, and graph neural networks, multi-source heterogeneous data of laborers are fused and processed to construct a six-dimensional digital profile that includes basic information, skill information, health information, safety information, performance information, and credit information.

[0007] Based on a six-dimensional digital profile and a fully connected neural network model, a job matching analysis is performed on the project's job requirements and labor characteristics to generate matching degree ranking results and the optimal personnel allocation plan.

[0008] Based on the matching degree ranking results and the optimal personnel configuration plan, the six-dimensional digital profile is updated in a closed loop to form a management link for the entire life cycle.

[0009] Furthermore, based on convolutional neural networks, Transformer networks, and graph neural networks, multi-source heterogeneous data of laborers are fused and processed to construct a six-dimensional digital profile, including the following steps:

[0010] The OCR technology is used to extract the identification information of the workers and obtain structured identification data.

[0011] The facial features of laborers are collected based on facial recognition, and facial feature vectors are obtained.

[0012] Based on the collection of health data from laborers using intelligent medical examination equipment, basic health indicators are obtained.

[0013] Semantic feature vectors are obtained by semantic parsing of text data of laborers based on Transformer network;

[0014] Structured document data, facial feature vectors, basic health indicators, and semantic feature vectors are standardized and fused to generate a six-dimensional digital profile.

[0015] Furthermore, based on a six-dimensional digital profile and a fully connected neural network model, a person-job matching analysis is performed on the project's job requirements and labor characteristics to generate matching degree ranking results and optimal personnel allocation plans, including the following steps:

[0016] Extract the characteristics of job requirements for the project to obtain a job requirement vector;

[0017] Extracting worker features from a six-dimensional digital profile yields a worker feature vector;

[0018] The job requirement vector and personnel feature vector are input into a pre-trained fully connected neural network model to calculate the job-person matching degree and obtain a matching degree score.

[0019] Based on the ranking results of the matching scores, and combined with the team structure, the optimal personnel allocation plan is generated.

[0020] Furthermore, it also includes handling the attendance and safety behavior of workers, including:

[0021] Real-time processing of construction site video images is performed using convolutional neural networks to identify undetected attendance records and unsafe behavior characteristics.

[0022] The first-level warning signal is triggered based on the characteristics of unsafe behavior, and on-site alarm information is obtained;

[0023] Based on the Long Short-Term Memory Network, the location data and physiological data collected by smart wearable devices are analyzed to determine the working status and health status of workers and obtain status assessment results.

[0024] The second-level early warning signal is triggered based on the status assessment results, and an early warning notification is generated and pushed to the management terminal.

[0025] Furthermore, it also includes scheduling and dispatching of laborers' shifts and tasks, including:

[0026] Based on the status assessment results and the contactless attendance records, the available time periods and real-time workload of the workers are extracted to obtain the personnel availability parameters.

[0027] Based on deep reinforcement learning networks, with the optimization goals of shortest construction period, lowest cost and highest personnel utilization, personnel availability parameters, project schedule, personnel skill level and historical work efficiency are input into the deep reinforcement learning network to generate daily shift schedule and task allocation scheme.

[0028] In the event of an emergency, the emergency parameters are re-input into the deep reinforcement learning network to obtain an adjusted scheduling plan.

[0029] Furthermore, it also includes the calculation of the performance and compensation of laborers, including:

[0030] The actual work efficiency is obtained by collecting workload data of laborers using IoT devices.

[0031] The construction quality is automatically evaluated based on a convolutional neural network to obtain a quality score.

[0032] Input the quality score, safety records, attendance, and training participation into the performance evaluation model to generate a comprehensive performance evaluation result.

[0033] Based on comprehensive performance evaluation results and salary standards, the system automatically calculates the wages payable to service workers.

[0034] Furthermore, it also includes training and recommending workers, including:

[0035] Constructing a knowledge graph for construction industry training;

[0036] Based on the comprehensive performance evaluation results and the six-dimensional digital profile, the skill deficiencies of laborers are identified, and the characteristics of skill gaps are obtained.

[0037] Based on the skill deficiency features, a knowledge graph of construction industry training is retrieved to obtain a set of candidate training knowledge points.

[0038] The candidate training knowledge point set is sorted and filtered based on the recommendation algorithm to generate a personalized training course list;

[0039] The system automatically generates assessment questions for each course in the personalized training course list using natural language processing technology.

[0040] Furthermore, it also includes evaluating the creditworthiness of laborers, including:

[0041] By extracting performance records, safety records, performance, training participation, and attendance from the six-dimensional digital profile, a multi-dimensional credit factor is obtained.

[0042] By inputting multidimensional credit factors into a credit evaluation model constructed by a graph neural network, the credit score of the workers is calculated to obtain the initial credit score.

[0043] Credit ratings are determined based on initial credit scores, resulting in credit rating labels.

[0044] Credit scores and credit rating labels are linked to the credit information dimensions of a six-dimensional digital profile.

[0045] According to one embodiment of the present invention, a device for managing the entire life cycle of construction project workers is also provided, comprising:

[0046] The module is used to fuse multi-source heterogeneous data of laborers based on convolutional neural networks, Transformer networks and graph neural networks to build a six-dimensional digital profile that includes basic information, skill information, health information, safety information, performance information and credit information.

[0047] The generation module is used to perform job matching analysis on project job requirements and labor characteristics based on a six-dimensional digital profile and a fully connected neural network model, and generate matching degree ranking results and optimal personnel allocation plan.

[0048] The module is used to perform closed-loop updates on the six-dimensional digital profile based on the matching degree ranking results and the optimal personnel configuration plan, forming a management link for the entire life cycle.

[0049] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is executed, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of the present invention.

[0050] In this embodiment of the invention, multi-source heterogeneous data of laborers are fused and processed using convolutional neural networks, Transformer networks, and graph neural networks to construct a six-dimensional digital profile including basic skills, health, safety, performance, and credit. Based on this profile and a fully connected neural network model, a person-job matching analysis is performed, and the matching results are fed back to the digital profile for closed-loop updates. This achieves data-driven accurate matching, dynamic iteration of the profile, and collaborative linkage of all stages throughout the entire life cycle, significantly improving the automation level and decision-making accuracy of laborer management. This solves the technical problems in existing construction project laborer management methods, such as difficulty in multi-source data fusion, reliance on fixed rules for person-job matching, and fragmented management stages that cannot form a dynamic feedback loop. Attached Figure Description

[0051] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0052] Figure 1 This is a flowchart of a method for full life-cycle management of construction project workers according to one embodiment of the present invention;

[0053] Figure 2 This is a structural block diagram of a construction project laborer life cycle management device according to one embodiment of the present invention. Detailed Implementation

[0054] 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.

[0055] 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, apparatus, product, or device 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 devices.

[0056] According to an embodiment of the present invention, a method for full life-cycle management of construction project laborers is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer device 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.

[0057] This method embodiment can be executed in an electronic device or similar computing device that includes a memory and a processor. Taking operation on a vehicle terminal as an example, the vehicle terminal may include one or more processors (processors may include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), digital signal processing (DSP) chips, microcontroller units (MCUs), field-programmable gate arrays (FPGAs), neural network processors (NPUs), tensor processors (TPUs), artificial intelligence (AI) type processors, etc.) and a memory for storing data. Optionally, the vehicle terminal may also include transmission devices, input / output devices, and display devices for communication functions. Those skilled in the art will understand that the above structural description is merely illustrative and does not limit the structure of the vehicle terminal. For example, the vehicle terminal may include more or fewer components than described above, or have a different configuration than described above.

[0058] The memory can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the construction project labor lifecycle management method in this embodiment of the invention. The processor executes various functional applications and data processing by running the computer program stored in the memory, thereby realizing the aforementioned construction project labor lifecycle management 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 mobile 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.

[0059] The transmission device is used to receive or send data via a network. Specific examples of the network mentioned above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device 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 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0060] Display devices can be, for example, touchscreen liquid crystal displays (LCDs) and touch displays (also referred to as "touchscreens" or "touch displays"). The LCD allows users to interact with the user interface of the mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), which allows users to interact with the GUI through finger contact and / or gestures on a touch-sensitive surface. Optional human-computer interaction functions include: creating web pages, drawing, word processing, creating electronic documents, playing games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital video, playing digital music, and / or web browsing, etc. Executable instructions for performing the above human-computer interaction functions are configured / stored in one or more processor-executable computer program products or readable storage media.

[0061] Figure 1 This is a flowchart of a method for full life-cycle management of construction project workers according to one embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0062] Step S110 involves fusing multi-source heterogeneous data of laborers based on convolutional neural networks, Transformer networks, and graph neural networks to construct a six-dimensional digital profile including basic information, skill information, health information, safety information, performance information, and credit information. The specific steps are as follows:

[0063] In step S110, constructing a six-dimensional digital profile of construction project workers first requires collecting multi-source heterogeneous data on the workers. Specifically, OCR text recognition technology is used to automatically extract information from workers' ID cards, academic certificates, skill certificates, and special operation permits to obtain structured document data; facial recognition devices are used to collect workers' facial images, extract facial feature vectors, and establish a facial feature database; intelligent medical examination devices are used to collect workers' basic health data such as height, weight, blood pressure, and heart rate to obtain basic health indicators; and the construction site management system and historical project database are used to obtain textual data such as workers' work experience, past project performance, safety violation records, training participation, and performance ratings.

[0064] After completing the multi-source data collection, various neural networks were used to fuse the heterogeneous data. For structured data such as identification information and health data, normalization was directly performed to form numerical features. For textual data such as work experience, safety records, and performance feedback, semantic parsing was performed based on a Transformer network to extract key entity and relational features, generating semantic feature vectors. Specifically, the textual descriptions of the workers were input into a pre-trained BERT model, which outputs a 768-dimensional semantic embedding vector. This vector represents the deep semantic information of the workers regarding their skills, experience, and compliance.

[0065] For video image data collected at the construction site, a convolutional neural network (CNN) is used for feature extraction. Monitoring video frames of workers entering and leaving the site are input into an improved YOLOv8 network to detect facial regions and extract identity features. Simultaneously, it identifies whether workers are wearing safety equipment such as helmets and reflective vests, generating safety behavior feature vectors. This CNN consists of a sequentially connected input layer, multiple convolutional layers, pooling layers, and a fully connected layer, outputting worker identification results and safety status labels.

[0066] To model the collaborative relationships, team structures, and credit associations among laborers, a graph neural network is employed. Each laborer is treated as a graph node, with their historical collaboration records, task transfer relationships, and credit guarantee relationships as edges, constructing a laborer relationship graph. The graph neural network aggregates the feature information of neighboring nodes through a message-passing mechanism, outputting a credit feature vector and relationship embedding representation for each laborer.

[0067] Finally, the identity and safety behavior features output by the convolutional neural network, the semantic feature vector output by the Transformer network, the credit feature vector output by the graph neural network, and the structured document data, facial feature vectors, and basic health indicators extracted by OCR are standardized and fused to generate a six-dimensional digital profile containing basic information, skill information, health information, safety information, performance information, and credit information. The basic information dimension includes name, age, ID number, and job type; the skill information dimension includes skill certificates, years of experience, and areas of expertise; the health information dimension includes physical examination indicators, medical history, and physical fitness assessment; the safety information dimension includes violation records, safety training results, and accident rate; the performance information dimension includes task completion rate, quality score, and attendance rate; and the credit information dimension includes contract performance records, credit score, and rewards and punishments. This six-dimensional digital profile is stored in a database and iterates in real time as the worker's subsequent work data, training data, and assessment results are updated, ensuring the timeliness and accuracy of the profile.

[0068] Step S120: Based on the six-dimensional digital profile and fully connected neural network model, perform a person-job matching analysis on the project's job requirements and labor characteristics to generate matching degree ranking results and the optimal personnel allocation plan. The specific steps are as follows:

[0069] In step S120, after constructing the six-dimensional digital profile of the labor force, this embodiment further performs a job-person matching analysis. First, job feature vectors are extracted from the task requirements of the construction project. Specifically, for each job to be recruited or assigned in the current project, the system automatically parses key elements such as job title, required skill type, years of work experience required, qualification certificate requirements, work location, salary budget, and project urgency, encoding these elements into a fixed-dimensional job requirement vector. For example, for the scaffolder job, the skill dimension corresponds to scaffolding erection and dismantling, the experience dimension requires more than three years, the qualification dimension requires a special operation certificate, the work location dimension locks the project site coordinates, the salary dimension sets the daily base wage, and the urgency dimension is set as high priority according to the project schedule. The above features are normalized to form a standardized job requirement vector.

[0070] Simultaneously, matching feature vectors for each worker are extracted from the constructed six-dimensional digital profile. The basic information dimension of the six-dimensional digital profile provides the worker's age, job category, and ID card information; the skills information dimension provides a list of skills certificates, years of experience, and proficient work processes; the health information dimension provides physical examination status and physical fitness assessment level; the safety information dimension provides historical violation records and safety training results; the performance information dimension provides past task completion rates, quality scores, and attendance rates; and the credit information dimension provides contract performance records and credit scores. The system numerically encodes the key features of these six dimensions to form a personnel feature vector corresponding to each worker. This vector resides in the same feature space as the job requirement vector, facilitating subsequent matching degree calculations.

[0071] After extracting the job requirement vector and personnel feature vector, they are input into a pre-trained fully connected neural network model. This fully connected neural network model includes an input layer, multiple hidden layers, and an output layer. The input layer receives the concatenated job requirement vector and personnel feature vector. The hidden layers employ several fully connected layers and a ReLU activation function for non-linear transformation. The output layer uses a sigmoid function to output a matching score between zero and one; a higher score indicates a higher degree of suitability between the worker and the job. The training process of the fully connected neural network model uses historical job-person matching data as training samples and actual post-hiring performance as supervision labels, optimizing the network weights through a backpropagation algorithm.

[0072] After calculating the matching degree for each candidate worker in the current project using a fully connected neural network model, the system sorts them from highest to lowest matching degree score, generating a matching degree ranking result. This ranking result visually presents the suitability priority of each worker and the target position for management reference. Based on this, the system further optimizes the allocation by considering the overall team structure requirements of the project. Specifically, the system obtains the planned personnel quota for each position in the project plan, the collaborative dependencies between positions, and the skill matching requirements of the teams. Based on the matching degree ranking result, it uses a greedy algorithm or the Hungarian algorithm to solve for the optimal personnel allocation scheme, ensuring that the workers assigned to each position have the highest overall matching degree, while also satisfying conditions such as skill complementarity within the team, reasonable experience levels, and total cost constraints. Finally, the system outputs the optimal personnel allocation scheme, which clarifies the specific list of workers corresponding to each position and their matching degree scores, and pushes this scheme to the project management personnel's terminal.

[0073] Step S140: Based on the matching degree ranking results and the optimal personnel configuration plan, perform a closed-loop update on the six-dimensional digital profile to form a full lifecycle management link. The specific steps are as follows:

[0074] In step S140, after generating the matching ranking results and the optimal personnel allocation plan, this embodiment feeds back the results to the six-dimensional digital profile of the workers, realizing dynamic closed-loop updates to the profile. Specifically, the system records the job information of each worker who has been successfully matched, the matching score, and the job allocation results in the final personnel allocation plan into the worker's digital profile. For workers who are hired or assigned to new positions, the performance information dimension of their six-dimensional digital profile is updated with the current project name, job responsibilities, and expected duration; the skills information dimension marks the key skills required for the position according to the job requirements, providing a basis for subsequent skills assessment; and the credit information dimension records the start time of successful performance of this match, serving as one of the basic data for credit score calculation.

[0075] Meanwhile, actual feedback data during the implementation of the optimal personnel allocation plan is continuously collected and used to update the profile. For example, attendance records, safety behavior recognition results, performance evaluation scores, and payroll calculation data generated by workers in their actual work are all automatically written back to the corresponding six-dimensional digital profile dimensions by the system. Attendance records update the attendance rate indicator in the health information dimension, safety behavior recognition results update the violation records and warning times in the safety information dimension, performance evaluation scores update the quality score and task completion rate in the performance information dimension, and payroll calculation data is archived as income records in the basic information dimension. Each data update triggers a recalculation of the profile, ensuring that the six-dimensional digital profile always reflects the latest status of the workers.

[0076] As the project progresses, when workers complete their current tasks or move to the next position, the system, based on the updated six-dimensional digital profile, executes the first two steps of claim 1 again: re-integrating multi-source heterogeneous data to construct a new profile, and performing a new person-job matching analysis based on this latest profile to generate a new round of matching degree ranking results and optimal personnel allocation plan. This process is repeated cyclically, ensuring that the digital profile of workers remains dynamically updated and forms a closed loop with the actual management process throughout their entire career path, from onboarding, job matching, task execution, performance evaluation, salary settlement to credit accumulation and even re-employment after leaving the company. This forms a full lifecycle management chain from data collection to profile construction, from profile to matching configuration, from configuration result feedback to profile updates, achieving full traceability, real-time optimization, and closed-loop iterative management of construction project workers.

[0077] Based on steps S110 to S140 above, in this embodiment of the invention, multi-source heterogeneous data of laborers are fused and processed using convolutional neural networks, Transformer networks, and graph neural networks to construct a six-dimensional digital profile including basic skills, health, safety, performance, and credit. Based on this profile and a fully connected neural network model, a person-job matching analysis is performed, and the matching results are fed back to the digital profile for closed-loop updates. This achieves data-driven accurate matching, dynamic iteration of the profile, and collaborative linkage of all stages throughout the entire life cycle, significantly improving the automation level and decision-making accuracy of laborer management. This solves the technical problems in existing construction project laborer management methods, such as difficulty in multi-source data fusion, reliance on fixed rules for person-job matching, and fragmented management stages that cannot form a dynamic feedback loop.

[0078] The method of this invention, based on convolutional neural networks, Transformer networks, and graph neural networks, fuses multi-source heterogeneous data of laborers to construct a six-dimensional digital profile. The method includes the following steps: extracting laborers' identification information using OCR technology to obtain structured identification data; collecting laborers' facial features using face recognition to obtain facial feature vectors; collecting laborers' health data using intelligent medical examination equipment to obtain basic health indicators; performing semantic parsing on laborers' text data using Transformer networks to obtain semantic feature vectors; and standardizing and fusing the structured identification data, facial feature vectors, basic health indicators, and semantic feature vectors to generate a six-dimensional digital profile.

[0079] This embodiment integrates heterogeneous data from multiple sources, including OCR, facial recognition, intelligent physical examination, and Transformer semantic parsing, to automatically generate a six-dimensional digital profile encompassing basic skills, health, safety, performance, and credit. This achieves comprehensive aggregation and structured representation of laborer information, laying a unified and dynamic data foundation for subsequent accurate matching and closed-loop management.

[0080] Furthermore, based on a six-dimensional digital profile and a fully connected neural network model, a person-job matching analysis is performed on the project's job requirements and labor characteristics to generate a matching degree ranking result and an optimal personnel allocation plan. This includes the following steps: extracting project job requirement features to obtain a job requirement vector; extracting labor characteristics from the six-dimensional digital profile to obtain a personnel feature vector; inputting the job requirement vector and personnel feature vector into a pre-trained fully connected neural network model to calculate the person-job matching degree and obtain a matching degree score; and generating an optimal personnel allocation plan based on the matching degree score ranking result and the team structure.

[0081] This embodiment uses a fully connected neural network model to perform nonlinear mapping and matching degree calculation between job demand vectors and labor personnel feature vectors, and generates the optimal configuration scheme by combining the team structure, thereby realizing intelligent and precise matching of people and jobs, and significantly improving the scientificity and efficiency of personnel allocation.

[0082] Furthermore, it also includes processing the attendance and safety behavior of workers, including: performing real-time processing of video images from the construction site based on convolutional neural networks to identify undetectable attendance records and unsafe behavior characteristics; triggering a first-level early warning signal based on unsafe behavior characteristics to obtain on-site alarm information; analyzing location data and physiological data collected by smart wearable devices based on long short-term memory networks to determine the workers' working status and health status, and obtaining status assessment results; triggering a second-level early warning signal based on the status assessment results to generate an early warning notification pushed to the management terminal.

[0083] This embodiment uses a convolutional neural network to identify undetectable attendance and unsafe behaviors in videos in real time and triggers a first-level on-site warning. At the same time, it combines a long short-term memory network to analyze the location and physiological data of wearable devices to assess work status and health status and trigger a second-level push warning. This realizes dual-modal real-time monitoring and hierarchical warning of the safety and health of personnel at the construction site, which significantly improves the comprehensiveness and timeliness of risk prevention and control.

[0084] Furthermore, it also includes scheduling and dispatching of laborers' shifts and tasks, including: extracting available time slots and real-time workloads of laborers based on status assessment results and seamless attendance records to obtain personnel availability parameters; using a deep reinforcement learning network, with the optimization objectives of shortest construction period, lowest cost, and highest personnel utilization rate, inputting personnel availability parameters, project schedule, personnel skill level, and historical work efficiency into the deep reinforcement learning network to generate daily shift plans and task allocation plans; in the event of emergencies, the emergency parameters are re-inputted into the deep reinforcement learning network to obtain an adjusted shift plan.

[0085] This embodiment extracts personnel availability parameters based on state assessment and seamless attendance tracking, and inputs them along with project progress, skill level, and historical efficiency into a deep reinforcement learning network. It dynamically generates daily scheduling plans with multiple objectives, including the shortest construction period, lowest cost, and highest personnel utilization. In case of emergencies, the scheduling can be adjusted in real time by re-inputting parameters, thereby significantly improving the intelligence level, adaptability, and overall operational efficiency of labor resource scheduling.

[0086] Furthermore, it also includes calculating the performance and compensation of laborers, including: collecting laborers' workload data based on IoT devices to obtain actual work efficiency; automatically evaluating construction quality based on convolutional neural networks to obtain quality scores; inputting the quality scores, safety records, attendance, and training participation into the performance evaluation model to generate comprehensive performance evaluation results; and automatically calculating the wages payable to laborers based on the comprehensive performance evaluation results and compensation standards.

[0087] This embodiment automatically collects workload data through IoT devices and uses convolutional neural networks to score construction quality. It then combines safety records, attendance, and training participation to generate a comprehensive performance evaluation result. Based on the salary standard, it automatically calculates the salary payable, realizing multi-dimensional, objective, and quantitative integrated performance-based salary calculation, which significantly improves the fairness and efficiency of salary management.

[0088] Furthermore, it also includes training recommendations for laborers, including: constructing a knowledge graph for construction industry training; identifying the skill deficiencies of laborers based on comprehensive performance evaluation results and a six-dimensional digital profile, obtaining skill deficiency characteristics; retrieving the knowledge graph for construction industry training based on skill deficiency characteristics, obtaining a set of candidate training knowledge points; sorting and filtering the set of candidate training knowledge points based on recommendation algorithms, generating a personalized training course list; and automatically generating assessment questions for each course in the personalized training course list based on natural language processing technology.

[0089] This embodiment constructs a knowledge graph for training in the construction industry and combines comprehensive performance evaluation and six-dimensional digital profiling to accurately identify skill gaps. Then, based on graph retrieval and recommendation algorithms, it automatically generates personalized training course lists and corresponding assessment questions, achieving intelligent diagnosis of training needs and precise matching of training resources, significantly improving the relevance and effectiveness of training.

[0090] Furthermore, it also includes evaluating the credit of laborers, including: extracting performance records, safety records, performance, training participation, and attendance from the six-dimensional digital profile to obtain multi-dimensional credit factors; inputting the multi-dimensional credit factors into a credit evaluation model constructed by a graph neural network to calculate the credit score of the laborers to obtain an initial credit score; classifying credit levels based on the initial credit score to obtain credit level labels; and associating the credit score and credit level labels with the credit information dimensions of the six-dimensional digital profile.

[0091] This embodiment extracts performance, safety records, performance, training participation, and attendance from a six-dimensional digital profile as multi-dimensional credit factors. It then uses a credit evaluation model constructed using a graph neural network to calculate credit scores and classify levels. Finally, it links the scores and tags back to the credit information dimensions of the six-dimensional digital profile, realizing a multi-factor integrated dynamic credit evaluation and closed-loop profile update. This provides a scientific and quantitative basis for decision-making regarding the recruitment, task allocation, and salary adjustment of laborers.

[0092] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0093] This invention also provides a device for managing the entire lifecycle of construction project workers, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0094] Figure 2 According to one embodiment of the present invention, a construction project laborers' life-cycle management device includes:

[0095] Module 201 is used to fuse multi-source heterogeneous data of laborers based on convolutional neural networks, Transformer networks and graph neural networks to construct a six-dimensional digital profile that includes basic information, skill information, health information, safety information, performance information and credit information.

[0096] The generation module 202 is used to perform job matching analysis on project job requirements and labor characteristics based on a six-dimensional digital profile and a fully connected neural network model, and generate matching degree ranking results and optimal personnel configuration plan.

[0097] The module 203 is used to perform closed-loop updates on the six-dimensional digital profile based on the matching degree ranking results and the optimal personnel configuration plan, forming a management link for the entire life cycle.

[0098] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0099] According to one embodiment of the present invention, an electronic device is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the above-described method for the full life-cycle management of construction project laborers during runtime.

[0100] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0101] Step S1: Based on convolutional neural networks, Transformer networks and graph neural networks, multi-source heterogeneous data of laborers are fused and processed to construct a six-dimensional digital profile including basic information, skill information, health information, safety information, performance information and credit information;

[0102] Step S2: Based on the six-dimensional digital profile and the fully connected neural network model, perform a job matching analysis on the project's job requirements and labor characteristics to generate matching degree ranking results and the optimal personnel allocation plan;

[0103] Step S3: Based on the matching degree ranking results and the optimal personnel configuration plan, perform a closed-loop update on the six-dimensional digital profile to form a full life cycle management link.

[0104] According to one embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the storage medium is located to execute the above-described method for full life-cycle management of construction project laborers.

[0105] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0106] Step S1: Based on convolutional neural networks, Transformer networks and graph neural networks, multi-source heterogeneous data of laborers are fused and processed to construct a six-dimensional digital profile including basic information, skill information, health information, safety information, performance information and credit information;

[0107] Step S2: Based on the six-dimensional digital profile and the fully connected neural network model, perform a job matching analysis on the project's job requirements and labor characteristics to generate matching degree ranking results and the optimal personnel allocation plan;

[0108] Step S3: Based on the matching degree ranking results and the optimal personnel configuration plan, perform a closed-loop update on the six-dimensional digital profile to form a full life cycle management link.

[0109] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0110] According to one embodiment of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements the above-described method for full life-cycle management of construction project laborers.

[0111] Optionally, in this embodiment, the above-mentioned computer program product can be configured as a computer program that performs the following steps:

[0112] Step S1: Based on convolutional neural networks, Transformer networks and graph neural networks, multi-source heterogeneous data of laborers are fused and processed to construct a six-dimensional digital profile including basic information, skill information, health information, safety information, performance information and credit information;

[0113] Step S2: Based on the six-dimensional digital profile and the fully connected neural network model, perform a job matching analysis on the project's job requirements and labor characteristics to generate matching degree ranking results and the optimal personnel allocation plan;

[0114] Step S3: Based on the matching degree ranking results and the optimal personnel configuration plan, perform a closed-loop update on the six-dimensional digital profile to form a full life cycle management link.

[0115] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0116] In the above embodiments of the present invention, 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.

[0117] 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 can be 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 device, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0118] 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0119] Furthermore, the functional units in the various embodiments of the present invention 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.

[0120] 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 invention, 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 of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0121] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for full life-cycle management of construction project workers, characterized in that, include: Based on convolutional neural networks, Transformer networks, and graph neural networks, multi-source heterogeneous data of laborers are fused and processed to construct a six-dimensional digital profile that includes basic information, skill information, health information, safety information, performance information, and credit information. Based on the six-dimensional digital profile and the fully connected neural network model, a job matching analysis is performed on the project's job requirements and labor characteristics to generate matching degree ranking results and the optimal personnel allocation plan. Based on the matching degree ranking results and the optimal personnel configuration scheme, the six-dimensional digital profile is updated in a closed loop to form a full lifecycle management link.

2. The method for full life-cycle management of construction project workers according to claim 1, characterized in that, The six-dimensional digital profile is constructed by fusing the multi-source heterogeneous data of the laborers based on the convolutional neural network, the Transformer network, and the graph neural network, including the following steps: The workers' identification information was extracted using OCR technology to obtain structured identification data. The facial features of the workers are collected based on facial recognition to obtain facial feature vectors; Based on the health data of the workers collected by the intelligent physical examination equipment, basic health indicators were obtained. Based on the Transformer network, semantic parsing is performed on the text data of the laborers to obtain semantic feature vectors; The structured document data, the facial feature vector, the basic health indicators, and the semantic feature vector are standardized and fused to generate the six-dimensional digital profile.

3. The method for full life-cycle management of construction project laborers according to claim 1, characterized in that, Based on the six-dimensional digital profile and the fully connected neural network model, the job requirements and labor characteristics of the project are analyzed to generate the matching degree ranking results and the optimal personnel allocation plan, including the following steps: Extract the characteristics of job requirements for the project to obtain a job requirement vector; The characteristics of the laborers are extracted from the six-dimensional digital profile to obtain the personnel feature vector; The job requirement vector and the personnel feature vector are input into a pre-trained fully connected neural network model to calculate the job-person matching degree and obtain a matching degree score. Based on the ranking results of the matching scores, the optimal personnel allocation plan is generated in combination with the team structure.

4. The method for full life-cycle management of construction project workers according to claim 1, characterized in that, It also includes handling the attendance and safety behavior of the aforementioned workers, including the following steps: Real-time processing of construction site video images is performed using convolutional neural networks to identify undetected attendance records and unsafe behavior characteristics. The first-level early warning signal is triggered based on the unsafe behavior characteristics, and on-site alarm information is obtained; Based on the Long Short-Term Memory Network, the location data and physiological data collected by the smart wearable device are analyzed to determine the working status and health status of the workers and obtain the status assessment results. Based on the status assessment results, a second-level early warning signal is triggered, generating an early warning notification that is pushed to the management terminal.

5. The method for full life-cycle management of construction project workers according to claim 4, characterized in that, It also includes scheduling and dispatching the work of the laborers, including the following steps: Based on the status assessment results and the contactless attendance records, the available time periods and real-time workload of the workers are extracted to obtain personnel availability parameters; Based on a deep reinforcement learning network, with the optimization goals of shortest construction period, lowest cost and highest personnel utilization, the personnel availability parameters, project schedule, personnel skill level and historical work efficiency are input into the deep reinforcement learning network to generate daily shift schedule and task allocation scheme. In the event of an emergency, the emergency parameters are re-inputted into the deep reinforcement learning network to obtain an adjusted scheduling plan.

6. The method for full life-cycle management of construction project workers according to claim 1, characterized in that, It also includes calculating the performance and compensation of the aforementioned workers, including the following steps: The actual work efficiency is obtained by collecting the workload data of the workers using IoT devices. The construction quality is automatically evaluated based on a convolutional neural network to obtain a quality score. The quality score, along with safety records, attendance, and training participation, are input into the performance evaluation model to generate a comprehensive performance evaluation result. Based on the comprehensive performance evaluation results and salary standards, the payable salary of the workers is automatically calculated.

7. The method for full life-cycle management of construction project workers according to claim 6, characterized in that, It also includes training and recommending the aforementioned workers, including the following steps: Constructing a knowledge graph for construction industry training; Based on the comprehensive performance evaluation results and the six-dimensional digital profile, the skill deficiencies of the workers are identified, and skill deficiency characteristics are obtained. Based on the skill deficiency features, the construction industry training knowledge graph is retrieved to obtain a set of candidate training knowledge points; The candidate training knowledge point set is sorted and filtered based on the recommendation algorithm to generate a personalized training course list; Assessment questions are automatically generated for each course in the personalized training course list based on natural language processing technology.

8. The method for full life-cycle management of construction project laborers according to claim 1, characterized in that, It also includes evaluating the creditworthiness of the workers, including the following steps: From the aforementioned six-dimensional digital profile, performance status, safety records, performance, training participation, and attendance are extracted to obtain multi-dimensional credit factors; The multidimensional credit factors are input into a credit evaluation model constructed by a graph neural network to calculate the credit score of the laborer and obtain an initial credit score. Credit ratings are determined based on the initial credit score, resulting in credit rating labels. The credit score and the credit rating label are associated with the credit information dimension of the six-dimensional digital profile.

9. A device for managing the entire life cycle of construction project workers, characterized in that, include: The module is used to fuse multi-source heterogeneous data of laborers based on convolutional neural networks, Transformer networks and graph neural networks to build a six-dimensional digital profile that includes basic information, skill information, health information, safety information, performance information and credit information. The generation module is used to perform job matching analysis on project job requirements and labor characteristics based on the six-dimensional digital profile and the fully connected neural network model, and generate matching degree ranking results and optimal personnel configuration plan; The forming module is used to perform closed-loop updates on the six-dimensional digital profile based on the matching degree ranking results and the optimal personnel configuration scheme, forming a management link for the entire life cycle.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 8.

Images