An electromechanical construction management system and method based on an AI algorithm

Abstract

The application discloses an electromechanical construction management system and method based on an AI algorithm, relates to the technical field of electromechanical construction management, and is based on historical project data, analyzes installation sequence constraints between electromechanical modules, and constructs a directed graph containing strong dependency relationships; analyzes the construction proficiency of each engineer on different electromechanical modules, and forms a personnel skill matrix; uses the directed graph to simulate task progress and discrete events, outputs a comprehensive bottleneck electromechanical module list and corresponding construction period delay risk value indicators; compares and simulates by virtually replacing engineers with different construction proficiency indicators, and screens implicit bottleneck electromechanical modules; in new project construction, analyzes construction period execution deviation indicators, generates an early warning signal and sends the early warning signal to an administrator when the construction period execution deviation indicators continuously exceed a threshold value; and after the project ends, updates the personnel skill matrix by using actual installation data. The reliability and analysis accuracy of the model are improved.

Description

Technical Field

[0001] This invention relates to the field of electromechanical construction management technology, specifically an electromechanical construction management system and method based on AI algorithms. Background Technology

[0002] Traditional electromechanical construction management relies heavily on the project manager's personal experience and static planning tools. When developing a construction plan, managers typically determine the general installation sequence of electromechanical modules by manually reviewing past cases or general specifications, and assign tasks based on their rough understanding of the construction team's capabilities. This approach has significant limitations.

[0003] First, its identification of complex dependencies between processes is subjective and incomplete, often only capturing the most obvious mandatory sequential relationships, while a large number of implicit constraints based on process coordination, spatial conflicts, or resource allocation are easily overlooked, resulting in inherent flaws in the plan itself. The matching between personnel skills and task difficulty remains qualitative, failing to quantify the differences in construction efficiency of specific technicians for certain complex modules. The pre-set work hours in the plan deviate from actual personnel capabilities, leading to insufficient accuracy in schedule prediction. This kind of plan, based on broad-based experience and static tables, is slow to adjust to the ever-changing dynamic situation on-site, making it difficult to anticipate chain delays caused by fluctuations in personnel efficiency or unexpected waiting times. The so-called critical path analysis often quickly becomes ineffective after the project starts.

[0004] Therefore, this invention discloses an electromechanical construction management system and method based on AI algorithms to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide an electromechanical construction management system and method based on AI algorithms to solve the problems raised in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an electromechanical construction management method based on AI algorithms, the method comprising the following steps: S1: Based on historical project data, analyze the installation sequence constraints between electromechanical modules and construct a directed graph containing strong dependencies; analyze the construction proficiency of each engineer for different electromechanical modules to form a personnel skill matrix; S2: Use directed graphs to simulate task progress and discrete events, and output a comprehensive list of bottleneck electromechanical modules and corresponding project delay risk values. S3: For the comprehensive bottleneck electromechanical module list, a comparative simulation is conducted by virtually replacing engineers with different construction proficiency indicators to analyze the sensitivity of the electromechanical module construction period to personnel changes and to screen out hidden bottleneck electromechanical modules. S4: During the construction of new projects, collect actual installation data, analyze the schedule execution deviation index, and generate an early warning signal and send it to the administrator when the schedule execution deviation index continues to exceed the threshold; after the project is completed, update the personnel skill matrix using actual installation data.

[0007] S1 includes the following: S101: Extract the installation sequence time series of all electromechanical modules from the historical project construction record database; for any two electromechanical modules M in the set of electromechanical modules... i With M j Statistics on electromechanical modules M in all historical projects i The installation was completed earlier than that of the electromechanical module M. j The probability of dependency on installation start time, which is equal to the probability of dependency on electromechanical module M. i The installation was completed earlier than that of the electromechanical module M. j The number of projects with installation start time and simultaneously including M i With M j The ratio of the total number of projects; where i and j are the serial numbers of electromechanical modules, and i is not equal to j; S102: Determine the relationship between electromechanical modules by combining the preset strong dependency threshold θ1 and weak cooperation threshold θ2: If the dependency probability is greater than the strong dependency threshold θ1, then determine M. i With M j The relationships between them are strong dependencies, and the relationship weight of the strong dependency is equal to the dependency probability; if the dependency probability is greater than the weak collaboration threshold θ2 and less than or equal to the strong dependency threshold θ1, then M is determined. i With M j The two electromechanical modules have a weak cooperative relationship, and the relationship weight of the weak cooperative relationship is equal to the dependency probability. If the dependency probability is less than or equal to the weak cooperative threshold θ2, it is determined that there is no sequential dependency between the two electromechanical modules. For all electromechanical module pairs that are determined to have a strong dependency relationship, a directed edge is established; the weight of the directed edge is equal to the relation weight of the corresponding relationship; a directed graph is constructed based on all electromechanical module pairs with strong dependencies and the corresponding edge set; S103: Based on the historical performance of each engineer, analyze the construction proficiency index of each engineer for each electromechanical module; integrate the construction proficiency indexes of all engineers and all electromechanical modules to generate a personnel skill matrix; the row index of the personnel skill matrix corresponds to the engineer's number; the column index of the personnel skill matrix corresponds to the electromechanical module's number. This invention explicitly uses dependency probability as the basis for relationship determination, which makes the construction of directed graphs based on objective statistical data rather than subjective experience, thus improving the reliability of the model; the establishment of the personnel skill matrix concretizes the abstract personnel capabilities into quantifiable indicators, realizes the digital management of human resources, and provides a basis for refined task assignment.

[0008] S2 includes the following: S201: The directed graph is used as a process logic diagram to simulate the task progress. The simulation process follows the basic principle of the critical path method and includes forward traversal and backward traversal. The forward traversal includes: starting from all electromechanical modules that do not have strong dependencies on predecessor tasks, analyzing the earliest start time and earliest finish time of each electromechanical module; the earliest start time of an electromechanical module is equal to the earliest finish time of all its predecessor tasks; then the earliest finish time of the corresponding electromechanical module is equal to the earliest start time of the electromechanical module plus the corresponding standard working hours; in the directed graph, if there are no other electromechanical modules between the first and second electromechanical modules, and the corresponding directed edge is from the first electromechanical module to the second electromechanical module, then the first electromechanical module is the predecessor task of the second electromechanical module; The reverse traversal includes: taking the earliest completion time of the last electromechanical module in the directed graph as the total project duration, and assigning the total project duration to the latest completion time of the last electromechanical module; analyzing the latest completion time and latest start time of each electromechanical module; the latest completion time of an electromechanical module is equal to the latest start time of all its successors; then the latest start time of the corresponding electromechanical module is equal to the latest completion time of the electromechanical module minus the corresponding standard working hours; if there are no other electromechanical modules between the first and second electromechanical modules in the directed graph, and the corresponding directed edge is from the first electromechanical module to the second electromechanical module, then the second electromechanical module is the successor of the first electromechanical module; The difference between the latest and earliest completion times of the electromechanical modules is denoted as the floating time index; the membership index of all electromechanical modules with a floating time index of zero is marked as one, and the membership index of other electromechanical modules is marked as zero; the betweenness centrality index of each electromechanical module is analyzed. The floating time index, membership index, and betweenness centrality index are normalized, and the normalized floating time index, membership index, and betweenness centrality index are weighted and fused to generate the electromechanical module criticality index. S202: For the target engineer assigned to each electromechanical module in the current construction plan, obtain the construction proficiency index of the corresponding electromechanical module from the personnel skill matrix; Couple the electromechanical module criticality index with the construction proficiency index to calculate the personnel electromechanical module mismatch factor between the target engineer and the assigned electromechanical module. The value of the personnel electromechanical module mismatch factor is positively correlated with the electromechanical module criticality index and negatively correlated with the construction proficiency index. S203: Adjust the planned working hours of each electromechanical module based on the personnel-electromechanical module mismatch factor; Using the revised set of planned working hours for the electromechanical modules, a discrete event simulation is performed in the directed graph. The discrete event simulation includes: setting all electromechanical module states to an unstarted state; resetting the simulation clock to zero; extracting all electromechanical modules whose preceding strongly dependent tasks have been completed and generating a ready queue; where the directed edge connects the third electromechanical module to the fourth electromechanical module, then the third electromechanical module is a preceding strongly dependent task of the fourth electromechanical module. When the electromechanical module M i If all prerequisite tasks with strong dependencies have been completed and the assigned target engineer is available, then the electromechanical module M can proceed. i Start installation timing and record its actual start time AS i And mark the corresponding target engineer as busy; Electromechanical Module M i The simulation duration is the corrected planned man-hours (Tsim). i After the installation timer starts, Tsim will be used. i Time triggers the end event; records its actual completion time AF. i =AS i +Tsim i Release the target engineer as available and check the electromechanical module M. i Does this mean that all subsequent strongly dependent tasks have the starting condition? If there is a directed edge from the third electromechanical module to the fourth electromechanical module, then the fourth electromechanical module is a subsequent strongly dependent task of the third electromechanical module. Traverse each electromechanical module in the directed graph. In discrete event simulation, two event chains that cause delays are traced: Resource Waiting: The electromechanical module cannot start immediately because its assigned target engineer is performing other tasks, even if all the strong prerequisites of the electromechanical module have been completed. The resulting waiting time is recorded as personnel waiting delay. Process waiting: Due to insufficient skills in a certain pre-required task, the revised planned working time is extended, causing the subsequent electromechanical modules to be unable to start at the earliest start time obtained from the task progress simulation. The resulting waiting time is recorded as process propagation delay. Analyze the expected start delay for each electromechanical module; the expected start delay is equal to the difference between the actual start time and the earliest start time obtained from simulation. After discrete event simulation, identify electromechanical modules whose floating time index becomes zero or whose expected start delay exceeds a threshold, and generate a comprehensive bottleneck electromechanical module list; analyze the schedule delay risk value index of each electromechanical module in the comprehensive bottleneck electromechanical module list; This invention couples module topology criticality with personnel skill mismatch factors, achieving joint analysis of "task characteristics" and "executor capabilities," making bottleneck assessment more closely resemble actual construction scenarios. By personalizing planned work hours, the simulation input better matches the actual skill level of specific engineers, improving the accuracy of simulation predictions. Clearly tracking two event chains—resource waiting and process waiting—allows for differentiation of delays due to different causes, providing a detailed basis for developing targeted control strategies. Based on an event-triggered discrete simulation process, it clearly simulates the logic of task state transitions, resource occupation, and release, realistically reproducing the dynamics and uncertainties of construction. By explicitly considering the availability of engineer resources, it can identify unplanned waiting caused by resource competition, which traditional Gantt charts or CPM methods cannot capture.

[0009] The simulation process is logically rigorous and can be converted into a computer program for efficient execution, providing feasibility for large-scale, multi-scenario simulation analysis.

[0010] In S3, the following is included: For each electromechanical module in the comprehensive bottleneck electromechanical module list, in the personnel skill matrix, find the virtual engineer with the highest construction proficiency index and the median construction proficiency index, replace the target engineer with the virtual engineer, and repeat the discrete event simulation to obtain two new expected start delays; computer electrical module M i The personnel dependence elasticity coefficient E i E i =|D i -Dmedian i | / (Dmax i -Dmedian i ), where D i For the electromechanical module M in the original discrete event simulation i The expected start delay; Dmedian i Indicates electromechanical module M i The projected start delay is generated by using virtual engineers with the median construction proficiency index for discrete event simulation; Dmax i Indicates electromechanical module M i The estimated start delay is generated by using a virtual engineer with the highest construction proficiency index to simulate discrete events. Electromechanical modules with personnel dependence elasticity coefficients exceeding the coefficient threshold are marked as implicit bottleneck electromechanical modules; This invention quantifies the dependence of specific modules on personnel skills through comparative simulation using controlled variables (virtual replacement engineers), providing data support for personnel allocation optimization. The "personnel dependence elasticity coefficient" directly reflects the sensitivity of module duration to personnel changes, helping managers identify hidden risks arising from "personnel lock-in." This method isolates personnel factors from the complex construction system for specialized analysis, improving the accuracy of problem attribution.

[0011] S4 includes the following: S401: During the construction of a new project, the actual start time, actual end time, and actual responsible engineer information of each electromechanical module in the integrated bottleneck electromechanical module list are collected in real time; based on the collected actual installation data, the actual construction period of each electromechanical module is calculated and compared with the corrected planned working hours and the expected start delay, generating a construction period execution deviation index; the construction period execution deviation index is equal to the ratio of the difference between the actual construction period and the corrected planned working hours to the expected start delay; when several consecutive construction period execution deviation indices exceed the preset warning threshold, a warning signal is generated and sent to the administrator; S402: After the project is completed, for each electromechanical module actually constructed and the corresponding actual responsible engineer, analyze the ratio of the actual construction period to the standard working hours, and record it as the efficiency performance value of the engineer's current construction; based on the efficiency performance value, perform a weighted smooth update on the construction proficiency index corresponding to the personnel skill matrix. The updated construction proficiency index is a weighted fusion of the historical construction proficiency index and the current efficiency performance value.

[0012] This invention, based on a continuous deviation-based threshold early warning mechanism, effectively filters out random fluctuations, reduces false alarms, and ensures the seriousness and effectiveness of early warning signals. Real-time monitoring and dynamic comparison of predicted values ​​enable "digital twin" management of the construction process, allowing managers to promptly identify and correct deviations. The use of weighted smoothing to update the personnel skill matrix respects historical performance while incorporating the latest data, enabling the personnel skill model to evolve dynamically and maintain its timeliness. The entire system forms a complete closed loop of "data-driven decision-making – execution feedback – model optimization," embodying the core characteristics of an adaptive intelligent system.

[0013] Another aspect of this application provides an AI-based electromechanical construction management system, which is applied to the above-mentioned AI-based electromechanical construction management method. The system includes an electromechanical module personnel basic analysis module, an electromechanical module list analysis module, an electromechanical module filtering module, and a real-time analysis and update module. The electromechanical module personnel basic analysis module is used to analyze historical project data, analyze the installation sequence constraints between electromechanical modules, construct a directed graph containing strong dependencies, and analyze the construction proficiency of each engineer for different electromechanical modules to form a personnel skill matrix. The electromechanical module list analysis module is used to perform task progress simulation and discrete event simulation using directed graphs, and outputs a comprehensive bottleneck electromechanical module list and corresponding project delay risk value indicators. The electromechanical module screening module is used to compare and simulate the comprehensive bottleneck electromechanical module list by virtually replacing engineers with different construction proficiency indicators, analyze the sensitivity of the electromechanical module construction period to personnel changes, and screen out hidden bottleneck electromechanical modules. The real-time analysis and update module is used to collect actual installation data and analyze the schedule execution deviation index during the construction of new projects. When the schedule execution deviation index continues to exceed the threshold, an early warning signal is generated and sent to the administrator. After the project is completed, the personnel skill matrix is ​​updated using the actual installation data.

[0014] The electromechanical module personnel basic analysis module includes a dependency probability analysis unit, a directed graph construction unit, and a personnel analysis unit; The dependency probability analysis unit is used to extract the installation sequence time series of all electromechanical modules from the historical project construction record database; and to analyze the dependency probability of any two electromechanical modules in the set of electromechanical modules. The directed graph construction unit is used to determine the relationship between electromechanical modules based on the dependency probability between two electromechanical modules; for all electromechanical module pairs determined to have a strong dependency relationship, a directed edge is established; and a directed graph is constructed based on all electromechanical module pairs with strong dependencies and the corresponding edge set. The personnel analysis unit is used to analyze the construction proficiency index of each engineer for each electromechanical module based on the historical performance of each engineer; and to integrate the calculation results of all engineers and all electromechanical modules to generate a personnel skill matrix.

[0015] The electromechanical module list analysis module includes a criticality index analysis unit, a mismatch factor analysis unit, and an anomaly module analysis unit. The criticality index analysis unit is used to simulate task progress using a directed graph and analyze the criticality index of each electromechanical module. The mismatch factor analysis unit is used to couple the criticality index of electromechanical modules with the construction proficiency index to calculate the personnel-electromechanical module mismatch factor between the target engineer and the assigned electromechanical modules. The anomaly module analysis unit is used to correct the planned working hours of each electromechanical module based on the personnel electromechanical module mismatch factor; using the corrected set of planned working hours of the electromechanical modules, discrete event simulation is performed in the directed graph; during the discrete event simulation, two event chains that cause delays are tracked; the two event chains that cause delays include resource waiting and process waiting; unplanned waiting event chains are identified, and a comprehensive list of bottleneck electromechanical modules and corresponding schedule delay risk values ​​are output.

[0016] The electromechanical module screening module includes a replacement analysis unit and a marking unit; The replacement analysis unit is used to find, in the personnel skill matrix, the virtual engineer with the highest construction proficiency index and the median construction proficiency index for each electromechanical module in the comprehensive bottleneck electromechanical module list, and replace the target engineer with the virtual engineer. The discrete event simulation is repeated to obtain two new expected start delays. The marking unit is used for the personnel dependence elasticity coefficient of the computer electrical module, and the electromechanical module whose personnel dependence elasticity coefficient exceeds the coefficient threshold is marked as a hidden bottleneck electromechanical module.

[0017] The real-time analysis and update module includes a real-time early warning unit and an attribute update unit; The real-time early warning unit is used to collect the actual installation data of each electromechanical module in the comprehensive bottleneck electromechanical module list in real time during the construction of new projects, calculate the actual construction period of each electromechanical module, and compare it with the corrected planned working hours and the expected start delay to generate a construction period execution deviation index; when several consecutive construction period execution deviation indices exceed the preset early warning threshold, an early warning signal is generated and sent to the administrator. The attribute update unit is used to analyze the ratio of the actual construction period to the standard working hours for each electromechanical module actually constructed and the corresponding actual responsible engineer after the project is completed, and record it as the efficiency performance value of the engineer's current construction; based on the efficiency performance value, the construction proficiency index corresponding to the personnel skill matrix is ​​updated with weighted smoothness.

[0018] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention explicitly uses dependency probability as the basis for relationship determination, enabling the construction of directed graphs to be based on objective statistical data rather than subjective experience, thus improving the reliability of the model. The establishment of the personnel skill matrix concretizes abstract personnel capabilities into quantifiable indicators, realizing digital management of human resources and providing a basis for refined task allocation. This invention couples module topological criticality with personnel skill mismatch factors, achieving joint analysis of "task characteristics" and "executor capabilities," making bottleneck assessment closer to actual construction scenarios. By personalizing the planned working hours, the simulation input better matches the actual level of specific engineers, improving the accuracy of simulation predictions. Clearly tracking two event chains—resource waiting and process waiting—can distinguish delays caused by different reasons, providing a detailed basis for subsequent targeted control strategies. Based on an event-triggered discrete simulation process, the logic of task state transitions, resource occupation and release is clearly simulated, realistically reproducing the dynamics and uncertainties in construction. Clearly considering the availability of engineer resources, it can identify unplanned waiting caused by resource competition, which traditional Gantt charts or CPM methods cannot capture. This invention quantifies the dependence of specific modules on personnel skills through comparative simulation of controlled variables, providing data support for personnel allocation optimization. The "personnel dependence elasticity coefficient" directly reflects the sensitivity of module duration to personnel changes, helping managers identify hidden risks arising from "personnel lock-in." This method isolates personnel factors from the complex construction system for specialized analysis, improving the accuracy of problem attribution. The invention's over-threshold early warning mechanism based on continuous deviation effectively filters out random fluctuations, reduces false alarms, and ensures the seriousness and effectiveness of early warning signals. Real-time monitoring and dynamic comparison of predicted values ​​enable "digital twin" management of the construction process, allowing managers to promptly identify and correct deviations. The use of weighted smoothing to update the personnel skill matrix respects historical performance while incorporating the latest data, enabling the personnel skill model to evolve dynamically and maintain timeliness. The entire system forms a complete "data-driven decision-making – execution feedback – model optimization" closed loop, embodying the core characteristics of an adaptive intelligent system. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating an AI-based electromechanical construction management method according to the present invention. Figure 2 This is a schematic diagram of the structure of an electromechanical construction management system based on AI algorithms according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0021] Please see Figure 1 The present invention provides a technical solution: an electromechanical construction management method based on AI algorithms, the method comprising the following steps: S1: Based on historical project data, analyze the installation sequence constraints between electromechanical modules and construct a directed graph containing strong dependencies; analyze the construction proficiency of each engineer for different electromechanical modules to form a personnel skill matrix; S1 includes the following: S101: Extract the installation sequence time series of all electromechanical modules from the historical project construction record database; for any two electromechanical modules M in the set of electromechanical modules... i With M j Statistics on electromechanical modules M in all historical projects i The installation was completed earlier than that of the electromechanical module M. j The dependency probability of installation start time is equal to that of electromechanical module M. i The installation was completed earlier than that of the electromechanical module M. j The number of projects with installation start time and simultaneously including M i With M j The ratio of the total number of projects; S102: Determine the relationship between electromechanical modules by combining the preset strong dependency threshold θ1 and weak cooperation threshold θ2: If the dependency probability is greater than the strong dependency threshold θ1, then determine M. i With M j The relationships between them are strong dependencies, and the relationship weight of a strong dependency is equal to the dependency probability. If the dependency probability is greater than the weak collaboration threshold θ2 and less than or equal to the strong dependency threshold θ1, then M is determined. i With M j The two electromechanical modules have a weak cooperative relationship, and the relationship weight of the weak cooperative relationship is equal to the dependency probability. If the dependency probability is less than or equal to the weak cooperative threshold θ2, it is determined that there is no sequential dependency between the two electromechanical modules. For all electromechanical module pairs that are determined to have a strong dependency relationship, a directed edge is established; the weight of the directed edge is equal to the relation weight of the corresponding relationship; a directed graph is constructed based on all electromechanical module pairs with strong dependencies and their corresponding edge sets; S103: Based on the historical performance of each engineer, analyze the construction proficiency index of each engineer for each electromechanical module; integrate the construction proficiency index of all engineers and all electromechanical modules to generate a personnel skill matrix; the row index of the personnel skill matrix corresponds to the engineer's number; the column index of the personnel skill matrix corresponds to the electromechanical module's number. Example 1: In this example, we query all instances in historical projects where the engineer was independently responsible for or led the completion of the same electromechanical module, and calculate the average actual construction time; the ratio of the standard construction time corresponding to the electromechanical module to the average actual construction time is recorded as the construction proficiency index. S2: Use directed graphs to simulate task progress and discrete events, and output a comprehensive list of bottleneck electromechanical modules and corresponding project delay risk values. S2 includes the following: S201: The directed graph is used as a process logic diagram to simulate the task progress. The simulation process follows the basic principle of the critical path method and includes forward traversal and backward traversal. The forward traversal includes: starting from all electromechanical modules that do not have strong dependencies on predecessor tasks, analyzing the earliest start time and earliest finish time of each electromechanical module; the earliest start time of an electromechanical module is equal to the earliest finish time of all its predecessor tasks; then the earliest finish time of the corresponding electromechanical module is equal to the earliest start time of the electromechanical module plus the corresponding standard working hours; in the directed graph, if there are no other electromechanical modules between the first and second electromechanical modules, and the corresponding directed edge is from the first electromechanical module to the second electromechanical module, then the first electromechanical module is the predecessor task of the second electromechanical module; The reverse traversal includes: taking the earliest completion time of the last electromechanical module in the directed graph as the total project duration, and assigning the total project duration to the latest completion time of the last electromechanical module; analyzing the latest completion time and latest start time of each electromechanical module; the latest completion time of an electromechanical module is equal to the latest start time of all its successors; then the latest start time of the corresponding electromechanical module is equal to the latest completion time of the electromechanical module minus the corresponding standard working hours; if there are no other electromechanical modules between the first and second electromechanical modules in the directed graph, and the corresponding directed edge is from the first electromechanical module to the second electromechanical module, then the second electromechanical module is the successor of the first electromechanical module; The difference between the latest and earliest completion times of the electromechanical modules is denoted as the floating time index; the membership index of all electromechanical modules with a floating time index of zero is marked as one, and the membership index of other electromechanical modules is marked as zero; the betweenness centrality index of each electromechanical module is analyzed. The floating time index, membership index, and betweenness centrality index are normalized, and the normalized floating time index, membership index, and betweenness centrality index are weighted and fused to generate the electromechanical module criticality index. S202: For the target engineer assigned to each electromechanical module in the current construction plan, obtain the construction proficiency index of the corresponding electromechanical module from the personnel skill matrix; Couple the criticality index of electromechanical modules with the construction proficiency index to calculate the personnel-electromechanical module mismatch factor between the target engineer and the assigned electromechanical module. The value of the personnel-electromechanical module mismatch factor is positively correlated with the criticality index of electromechanical modules and negatively correlated with the construction proficiency index. Example 2: In this example, D （i，p） =C i ×(1 / R （i，p） ) or D （i，p） =C i ×(Ravg) i / R （i，p） ); where D （i，p） Indicates electromechanical module M i Regarding the target engineer Eng p Personnel electromechanical module mismatch factor; C i Indicates electromechanical module M i Key performance indicators of electromechanical modules; R （i，p） Indicates motor module M i Regarding the target engineer Eng p Construction proficiency index; Ravg i Indicates motor module M i The average value of the construction proficiency index for each target engineer; S203: Adjust the planned working hours of each electromechanical module based on the personnel-electromechanical module mismatch factor; Example 3: In this example, the electromechanical module M i Revised Planned Work Hours Tsim i =Tstd i ×(1+γ×D （i，p） ); where Tstd i Indicates electromechanical module M i Standard working hours; γ represents the preset construction period impact coefficient; Using the revised set of planned working hours for the electromechanical modules, a discrete event simulation is performed in the directed graph. The discrete event simulation includes: setting all electromechanical module states to an unstarted state; resetting the simulation clock to zero; extracting all electromechanical modules whose preceding strongly dependent tasks have been completed and generating a ready queue; where the directed edge connects the third electromechanical module to the fourth electromechanical module, the third electromechanical module is a preceding strongly dependent task of the fourth electromechanical module. When the electromechanical module M i If all prerequisite tasks with strong dependencies have been completed and the assigned target engineer is available, then the electromechanical module M can proceed. iStart installation timing and record its actual start time AS i And mark the corresponding target engineer as busy; Electromechanical Module M i The simulation duration is the corrected planned man-hours (Tsim). i After the installation timer starts, Tsim will be used. i Time triggers the end event; records its actual completion time AF. i =AS i +Tsim i Release the target engineer as available and check the electromechanical module M. i Does this mean that all subsequent strongly dependent tasks have the starting condition? If there is a directed edge from the third electromechanical module to the fourth electromechanical module, then the fourth electromechanical module is a subsequent strongly dependent task of the third electromechanical module. Traverse each electromechanical module in the directed graph. In discrete event simulation, two event chains that cause delays are traced: Resource Waiting: The electromechanical module cannot start immediately because its assigned target engineer is performing other tasks, even if all the strong prerequisites of the electromechanical module have been completed. The resulting waiting time is recorded as personnel waiting delay. Process waiting: Due to insufficient skills in a certain pre-required task, the revised planned working time is extended, causing the subsequent electromechanical modules to be unable to start at the earliest start time obtained from the task progress simulation. The resulting waiting time is recorded as process propagation delay. Analyze the expected start delay for each electromechanical module; the expected start delay is equal to the difference between the actual start time and the earliest start time obtained from the simulation. After discrete event simulation, identify electromechanical modules whose floating time index becomes zero or whose expected start delay exceeds a threshold, and generate a comprehensive bottleneck electromechanical module list; analyze the schedule delay risk value index of each electromechanical module in the comprehensive bottleneck electromechanical module list; Example 4: In this example, the project delay risk value index is equal to the product of the difference between the actual start time of the electromechanical module and the earliest start time generated by the task progress simulation, and the criticality index of the electromechanical module. S3: For the comprehensive bottleneck electromechanical module list, compare and simulate the process by virtually replacing engineers with different construction proficiency indicators, analyze the sensitivity of the electromechanical module construction period to personnel changes, and screen out hidden bottleneck electromechanical modules. In S3, the following is included: For each electromechanical module in the comprehensive bottleneck electromechanical module list, in the personnel skill matrix, find the virtual engineer with the highest construction proficiency index and the median construction proficiency index, replace the target engineer with the virtual engineer, and repeat the discrete event simulation to obtain two new expected start delays; computer electrical module M i The personnel dependence elasticity coefficient Ei E i =|D i -Dmedian i | / (Dmax i -Dmedian i ), where D i For the electromechanical module M in the original discrete event simulation i The expected start delay; Dmedian i Indicates electromechanical module M i The projected start delay is generated by using virtual engineers with the median construction proficiency index for discrete event simulation; Dmax i Indicates electromechanical module M i The estimated start delay is generated by using a virtual engineer with the highest construction proficiency index to simulate discrete events. Electromechanical modules with personnel dependence elasticity coefficients exceeding the coefficient threshold are marked as implicit bottleneck electromechanical modules; S4: During the construction of new projects, collect actual installation data, analyze the schedule execution deviation index, and generate an early warning signal and send it to the administrator when the schedule execution deviation index continues to exceed the threshold; after the project is completed, update the personnel skill matrix using actual installation data.

[0022] S4 includes the following: S401: During the construction of a new project, the actual start time, actual end time, and information of the responsible engineer for each electromechanical module in the integrated bottleneck electromechanical module list are collected in real time. Based on the collected actual installation data, the actual construction period of each electromechanical module is calculated and compared with the corrected planned working hours and the expected start delay, generating a construction period execution deviation index. The construction period execution deviation index is equal to the ratio of the difference between the actual construction period and the corrected planned working hours to the expected start delay. When several consecutive construction period execution deviation indices exceed the preset warning threshold, a warning signal is generated and sent to the administrator. S402: After the project is completed, for each electromechanical module actually constructed and the corresponding responsible engineer, analyze the ratio of the actual construction period to the standard working hours, and record it as the engineer's current construction efficiency performance value; based on the efficiency performance value, perform a weighted smooth update on the construction proficiency index in the personnel skill matrix. The updated construction proficiency index is a weighted fusion of the historical construction proficiency index and the current efficiency performance value.

[0023] Please see Figure 2 The present invention provides a technical solution: an electromechanical construction management system based on AI algorithm, which includes an electromechanical module personnel basic analysis module, an electromechanical module list analysis module, an electromechanical module filtering module, and a real-time analysis and update module; The electromechanical module personnel basic analysis module uses historical project data to analyze the installation sequence constraints between electromechanical modules, construct a directed graph containing strong dependencies, and analyze the construction proficiency of each engineer for different electromechanical modules to form a personnel skill matrix. The electromechanical module list analysis module is used to simulate task progress and discrete events using directed graphs, and outputs a comprehensive list of bottleneck electromechanical modules and corresponding project delay risk values. The electromechanical module screening module is used to compare and simulate the comprehensive bottleneck electromechanical module list by virtually replacing engineers with different construction proficiency indicators, analyze the sensitivity of the electromechanical module construction period to personnel changes, and screen out hidden bottleneck electromechanical modules. The real-time analysis and update module is used to collect actual installation data and analyze schedule execution deviation indicators during the construction of new projects. When the schedule execution deviation indicators continue to exceed the threshold, an early warning signal is generated and sent to the administrator. After the project is completed, the personnel skill matrix is ​​updated using actual installation data.

[0024] The electromechanical module personnel basic analysis module includes a dependency probability analysis unit, a directed graph construction unit, and a personnel analysis unit; The dependency probability analysis unit is used to extract the installation sequence time series of all electromechanical modules from the historical project construction record database; and to analyze the dependency probability of any two electromechanical modules in the set of electromechanical modules. The directed graph construction unit is used to determine the relationship between electromechanical modules based on the dependency probability between two electromechanical modules; for all electromechanical module pairs determined to have a strong dependency relationship, directed edges are established; and a directed graph is constructed based on all electromechanical module pairs with strong dependencies and their corresponding edge sets. The personnel analysis unit is used to analyze the construction proficiency index of each engineer for each electromechanical module based on the historical performance of each engineer; it integrates the calculation results of all engineers and all electromechanical modules to generate a personnel skill matrix.

[0025] The electromechanical module list analysis module includes a criticality index analysis unit, a mismatch factor analysis unit, and an abnormal module analysis unit. The criticality index analysis unit is used to simulate task progress using directed graphs and analyze the criticality index of each electromechanical module. The mismatch factor analysis unit is used to couple the criticality index of electromechanical modules with the construction proficiency index to calculate the personnel-electromechanical module mismatch factor between the target engineer and the assigned electromechanical modules. The anomaly module analysis unit is used to correct the planned working hours of each electromechanical module based on the personnel electromechanical module mismatch factor; using the corrected set of planned working hours of electromechanical modules, discrete event simulation is performed in a directed graph; during the discrete event simulation, two event chains that cause delays are tracked; the two event chains that cause delays include resource waiting and process waiting; unplanned waiting event chains are identified, and a comprehensive list of bottleneck electromechanical modules and corresponding schedule delay risk values ​​are output.

[0026] The electromechanical module screening module includes a replacement analysis unit and a marking unit; The replacement analysis unit is used to find the virtual engineer with the highest construction proficiency index and the median construction proficiency index in the personnel skill matrix for each electromechanical module in the comprehensive bottleneck electromechanical module list. The virtual engineer is used to replace the target engineer, and the discrete event simulation is repeated to obtain two new expected start delays. The marking unit is used for the personnel dependence elasticity coefficient of computer electrical modules. Electromechanical modules with personnel dependence elasticity coefficients exceeding the coefficient threshold are marked as implicit bottleneck electromechanical modules.

[0027] The real-time analysis and update module includes a real-time early warning unit and an attribute update unit; The real-time early warning unit is used to collect the actual installation data of each electromechanical module in the comprehensive bottleneck electromechanical module list in real time during the construction of new projects, calculate the actual construction period of each electromechanical module, and compare it with the corrected planned working hours and the expected start delay to generate a construction period execution deviation index. When several consecutive construction period execution deviation indices exceed the preset early warning threshold, an early warning signal is generated and sent to the administrator. The attribute update unit is used to analyze the ratio of the actual construction period to the standard working hours for each actual electromechanical module and the corresponding actual responsible engineer after the project is completed, and record it as the efficiency performance value of the engineer's current construction; based on the efficiency performance value, the construction proficiency index in the personnel skill matrix is ​​updated with weighted smoothness.

[0028] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0029] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for electromechanical construction management based on AI algorithms, characterized in that, The method includes the following steps: S1: Based on historical project data, analyze the installation sequence constraints between electromechanical modules and construct a directed graph containing strong dependencies; analyze the construction proficiency of each engineer for different electromechanical modules to form a personnel skill matrix; S2: Use directed graphs to simulate task progress and discrete events, find electromechanical modules whose floating time index becomes zero or whose expected start delay is greater than the threshold after discrete event simulation, and generate a comprehensive bottleneck electromechanical module list; output the comprehensive bottleneck electromechanical module list and the corresponding schedule delay risk value index. S3: For the comprehensive bottleneck electromechanical module list, a comparative simulation is conducted by virtually replacing engineers with different construction proficiency indicators to analyze the sensitivity of the electromechanical module construction period to personnel changes and screen out hidden bottleneck electromechanical modules; the hidden bottleneck electromechanical modules are those whose personnel dependence elasticity coefficient exceeds the coefficient threshold. S4: During the construction of new projects, collect actual installation data, analyze the schedule execution deviation index, and generate an early warning signal and send it to the administrator when the schedule execution deviation index continues to exceed the threshold. After the project is completed, the personnel skills matrix will be updated using actual installation data.

2. The electromechanical construction management method based on AI algorithm according to claim 1, characterized in that: S1 includes the following: S101: Extract the installation sequence time series of all electromechanical modules from the historical project construction record database; analyze the dependency probability for any two electromechanical modules in the set of electromechanical modules; S102: Determine the relationship between the electromechanical modules based on the dependency probability between the two electromechanical modules; establish directed edges for all electromechanical module pairs that are determined to have a strong dependency relationship; construct a directed graph based on all electromechanical module pairs with strong dependencies and the corresponding edge set; S103: Based on the historical performance of each engineer, analyze the construction proficiency index of each engineer for each electromechanical module; integrate the construction proficiency index of all engineers and all electromechanical modules to generate a personnel skill matrix.

3. The electromechanical construction management method based on AI algorithm according to claim 2, characterized in that: S2 includes the following: S201: Use directed graphs to simulate task progress and analyze the criticality indicators of each electromechanical module. S202: Couple the criticality index of electromechanical modules with the construction proficiency index to calculate the personnel-electromechanical module mismatch factor between the target engineer and the assigned electromechanical modules; S203: Correct the planned working hours of each electromechanical module based on the personnel mismatch factor; use the corrected set of planned working hours of the electromechanical modules to perform discrete event simulation in the directed graph; during the discrete event simulation, track two event chains that cause delays; the two event chains that cause delays include resource waiting and process waiting; identify unplanned waiting event chains, and output a comprehensive list of bottleneck electromechanical modules and corresponding schedule delay risk value indicators.

4. The electromechanical construction management method based on AI algorithm according to claim 3, characterized in that: The task progress simulation includes forward traversal and reverse traversal; the forward traversal includes: starting from all electromechanical modules that do not have strong dependencies on predecessor tasks, analyzing the earliest start time and earliest completion time of each electromechanical module; The reverse traversal includes: taking the earliest completion time of the last electromechanical module in the directed graph as the total project duration, and assigning the total project duration to the latest completion time of the last electromechanical module; and analyzing the latest completion time and latest start time of each electromechanical module.

5. The electromechanical construction management method based on AI algorithm according to claim 4, characterized in that: The analysis of the criticality indicators of each electromechanical module includes the following: The difference between the latest and earliest completion times of the electromechanical modules is recorded as the floating time index; the membership index of all electromechanical modules with a floating time index of zero is set to one, and the membership index of other electromechanical modules is set to zero; the betweenness centrality index of each electromechanical module is analyzed; the floating time index, membership index, and betweenness centrality index are normalized, and the normalized floating time index, membership index, and betweenness centrality index are weighted and fused to generate the criticality index of the electromechanical module.

6. The electromechanical construction management method based on AI algorithm according to claim 3, characterized in that: The discrete event simulation includes the following: Set all electromechanical modules to the "not started" state; reset the analog clock to zero; extract all electromechanical modules whose prerequisite strongly dependent tasks have been completed and generate a ready queue; where the directed edge connects the third electromechanical module to the fourth electromechanical module, the third electromechanical module is a prerequisite strongly dependent task of the fourth electromechanical module. Once all prerequisite tasks of the electromechanical module have been completed and the corresponding assigned target engineer is available, the electromechanical module begins installation timing. The actual start time of the electromechanical module is recorded, and the corresponding target engineer is marked as busy. The simulated duration of the electromechanical module is the corrected planned working hours. After the corrected planned working hours have elapsed since the installation timing began, an end event is triggered. The actual completion time of the electromechanical module is recorded, the target engineer is released as available, and the start conditions of all subsequent prerequisite tasks of the electromechanical module are checked. Each electromechanical module in the directed graph is traversed.

7. The electromechanical construction management method based on AI algorithm according to claim 3, characterized in that: In S3, the following is included: For each electromechanical module in the comprehensive bottleneck electromechanical module list, in the personnel skill matrix, find the virtual engineer with the highest construction proficiency index and the median construction proficiency index, replace the target engineer with the virtual engineer, repeat the discrete event simulation, and obtain two new expected start delay amounts; calculate the personnel dependency elasticity coefficient of the electromechanical module, and mark the electromechanical module with the personnel dependency elasticity coefficient exceeding the coefficient threshold as a hidden bottleneck electromechanical module.

8. The electromechanical construction management method based on AI algorithm according to claim 7, characterized in that: S4 includes the following: S401: During the construction of a new project, the actual installation data of each electromechanical module in the integrated bottleneck electromechanical module list is collected in real time. Based on the collected actual installation data, the actual construction period of each electromechanical module is calculated and compared with the corrected planned working hours and the expected start delay, generating a construction period execution deviation index. When several consecutive construction period execution deviation indices exceed the preset warning threshold, a warning signal is generated and sent to the administrator. S402: After the project is completed, for each electromechanical module actually constructed and the corresponding actual responsible engineer, analyze the ratio of the actual construction period to the standard working hours, and record it as the efficiency performance value of the engineer's current construction; based on the efficiency performance value, perform weighted smoothing update on the construction proficiency index corresponding to the personnel skill matrix.

9. An AI-based electromechanical construction management system, wherein the system is applied to the AI-based electromechanical construction management method described in any one of claims 1-8, characterized in that, The system includes a basic analysis module for electromechanical module personnel, a list analysis module for electromechanical modules, a filtering module for electromechanical modules, and a real-time analysis and update module. The electromechanical module personnel basic analysis module is used to analyze historical project data, analyze the installation sequence constraints between electromechanical modules, construct a directed graph containing strong dependencies, and analyze the construction proficiency of each engineer for different electromechanical modules to form a personnel skill matrix. The electromechanical module list analysis module is used to perform task progress simulation and discrete event simulation using directed graphs, and outputs a comprehensive bottleneck electromechanical module list and corresponding project delay risk value indicators. The electromechanical module screening module is used to compare and simulate the comprehensive bottleneck electromechanical module list by virtually replacing engineers with different construction proficiency indicators, analyze the sensitivity of the electromechanical module construction period to personnel changes, and screen out hidden bottleneck electromechanical modules. The real-time analysis and update module is used to collect actual installation data and analyze the schedule execution deviation index during the construction of new projects. When the schedule execution deviation index continues to exceed the threshold, an early warning signal is generated and sent to the administrator. After the project is completed, the personnel skill matrix is ​​updated using the actual installation data.

Images