An intelligent construction management system and method based on data analysis

Abstract

The invention discloses an intelligent construction management system and method based on data analysis, belonging to the field of information management technology. The invention calculates each extracted feature data to generate a decision tree, merges the decision data of all feature data to construct a classifier, uses the classifier to judge all personnel data involved in the construction, and obtains the department category of the on-site construction; constructs the on-site construction process chain; finds the position of each construction department in the construction process chain and marks it; calculates the correlation index between adjacent departments in the construction process chain; constructs an on-site construction system using the calculated correlation index and the construction process chain; uses the accident threshold to monitor the construction site of each department in real time and issue an early warning; and the superior process department conducts a comprehensive analysis of all early warning departments.

Description

Technical Field

[0001] The present invention relates to the technical field of information management, and in particular to an intelligent construction management system and method based on data analysis. Background Art

[0002] The development of computers in the second half of the 20th century has greatly improved data processing and storage capabilities. Construction management has gradually shifted from manual recording and management to computerized management. With the popularization of the Internet, real-time transmission and sharing of information have been promoted, and construction management systems have been online to support remote collaboration and management; the development of Internet of Things technology has enabled a large number of sensors to be installed on construction sites to monitor environmental conditions, equipment operating status and personnel activities; the application of real-time data collection and transmission technology has been realized, making construction management more refined and intelligent. With the expansion of the scale and complexity of construction projects, a large amount of structural, environmental, personnel and equipment data has been accumulated during the construction process, which undoubtedly increases the difficulty of construction process management and the time for processing construction information; during the construction process, since the construction site may involve a wide range, the personnel and departments required in each field construction project are relatively complicated, resulting in insufficient information sharing between different departments or teams, resulting in the inability to fully integrate safety monitoring information; after the discovery of safety hazards, information transmission and communication may not be timely, affecting the speed of problem solving. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent construction management system and method based on data analysis to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] An intelligent construction management method based on data analysis, the method comprising the following steps:

[0006] S100, collecting personnel data of each department in the historical construction process, extracting characteristic data of the construction department in the history, calculating each extracted characteristic data to generate a decision tree, merging the decision data of all characteristic data to construct a classifier, and using the classifier to judge the data of all personnel involved in the construction during the on-site construction to obtain the department category of the on-site construction;

[0007] Furthermore, the specific steps for classifying on-site construction departments according to the characteristics of construction personnel are as follows:

[0008] S101. Collect the personnel data of each department in the historical construction process. , Indicates collecting the 1st, 2nd, 3rd, ..., nth personnel data of each department in the historical construction process, where n is a positive integer; comparing the collected personnel data of each department in the history, extracting the personnel data existing in each department as feature data, and constructing the feature data set as , represents the 1st, 2nd, 3rd, ..., mth characteristic data of the extracted construction department, where m is a positive integer;

[0009] S102. After determining the characteristic data of the construction department, collect the same characteristic data values ​​of each department in history. , Indicates the same characteristic data values ​​of the 1st, 2nd, 3rd, ..., jth departments in the collected history, where j is a positive integer; collect the same characteristic data values ​​of each department in the history k times, and extract the maximum value of the k same characteristic data of each department in the collected history and minimum value ; Sort the maximum and minimum values ​​of the extracted j departments from small to large to get , Indicates the maximum and minimum values ​​of the same characteristic data of the 1st, 2nd, 3rd, ..., jth departments after sorting from small to large; compares the previous maximum value and the next minimum value in adjacent departments. When the adjacent departments have overlapping values ​​of the same characteristic data, Indicates the maximum value of the same characteristic data of the previous department in the adjacent department. Indicates the minimum number of households with the same characteristics in the next department among adjacent departments; When , it is determined that there is no overlap in the same characteristic data values ​​of adjacent departments;

[0010] S103. When it is determined that adjacent departments have overlapping characteristic data, the maximum value of the previous department and the minimum value of the next department in the adjacent departments are replaced. The formula is:

[0011] ,

[0012] In the formula, Th represents the replacement value for replacing the maximum value of the previous department and the minimum value of the next department in the adjacent department; the data interval of the adjacent departments that are interlaced after replacement is and ; represents the interval after the replacement of the previous department in the adjacent department, Represents the interval after replacement of the next department in the adjacent departments; Use the calculated replacement value to replace the maximum and minimum values ​​of all departments with overlaps, and finally obtain the interval of the same feature data in all departments; Collect and analyze the intervals of all feature data in all departments in the feature data set to obtain the interval of each feature data in all departments; Use the interval of each feature data in all departments to design a decision tree, and construct m decision trees as follows , Indicates the 1st, 2nd, 3rd, ..., mth decision trees constructed using feature data; the m constructed decision trees are superimposed to establish a classifier;

[0013] S104, in the on-site construction, real-time feature data of each personnel data in the feature data set is collected in real time, and the real-time feature data of each personnel data collected is judged by using a classifier. Suppose the real-time feature data of each personnel data collected is , represents the 1st, 2nd, 3rd, ..., Lth feature data in each collected personnel data, where L is a positive integer; the L types of real-time feature data collected are judged by the classifier and the L types of classification results are obtained: , It means that the 1st, 2nd, 3rd, ..., Lth classification results are judged by the classifier on L kinds of real-time feature data, and the L classification results are comprehensively analyzed. The formula is:

[0014] ,

[0015] In the formula, Re represents the comprehensive probability of each classification result after classification, H represents the number of the same classification results, and Ts represents the collected real-time feature data. Represents the interval boundary value of the department in each result after classification, include and ; Calculate the difference between the collected real-time feature data and the boundary value, and select the minimum value of the two differences; Calculate the comprehensive probability of each judgment result, and select the result with the largest comprehensive probability as the judgment result of the corresponding personnel data; Use the classifier to judge each personnel data in the on-site construction and obtain all the department types. , Represents the 1st, 2nd, 3rd, ..., gth department in on-site construction, where g is a positive integer.

[0016] By analyzing the personnel data of each department in history, extracting feature data to build a decision tree, and using decision trees to superimpose and build a classifier, the data of on-site construction personnel can be classified, the type of on-site construction department can be obtained, and the type of on-site construction department can be judged using automated technical means to avoid the uncertainty caused by human subjective judgment; and the method of superimposing multiple decision trees to build a classifier can reduce the probability of classifier classification errors, and use multiple decision trees for comprehensive judgment to eliminate the errors in the judgment of a single decision tree.

[0017] S200, after classifying the on-site construction departments, collect the process data of the historical construction and build the on-site construction process chain; analyze the process data of each construction department, find the position of each construction department in the construction process chain and mark it;

[0018] Furthermore, the specific steps to find the position of each construction department in the construction process chain and mark it are:

[0019] S201. Collect the process data of each department during construction in history. , Represents the process data of the 1st, 2nd, 3rd, ..., jth departments in the collected history. The collected process data of each department are compared, and the department corresponding to the minimum value of the process data is extracted as the first construction department. Then, each department in the collected history is sorted in order from small to large according to the process data to build a construction process chain as , represents the 1st, 2nd, 3rd, ..., fth process in the construction process of the build, where f is a positive integer;

[0020] S202, collect the process data of each department in the on-site construction in real time, use the collected process data to find the process position of each department in the on-site construction in the construction process chain and mark it in each department. , It means marking the positions of the 1st, 2nd, 3rd, ..., gth departments in the construction process chain during on-site construction.

[0021] S300, after marking the position of each department in the construction process chain, calculating the correlation index between departments at adjacent positions in the construction process chain; using the calculated correlation index and the construction process chain to construct an on-site construction system;

[0022] Furthermore, the specific steps of constructing the on-site construction system using the calculated correlation index and construction process chain are as follows:

[0023] S301, when judging the number of departments in each process in the construction process chain, collecting the number of departments in each process in the construction process chain , Indicates the number of departments in the 1st, 2nd, 3rd, ..., fth processes in the collected construction process chain; the number of departments in each process is judged. , determine whether there is information isolation risk in the corresponding process; make a judgment on each process in the construction process chain to obtain all processes with information isolation risk; calculate the correlation index between all departments of the process with information isolation risk and each department in the previous process, assuming that the number of departments in the previous process is Y, and collect the number and amount of information exchanged between every two departments in history. The formula is:

[0024] ,

[0025] In the formula, Ga represents the correlation index between each department in the process with information isolation risk and each department at the previous level, X represents the number of information exchanges between the two departments, and Q represents the amount of information when the two departments exchange information; the correlation index between each department in the process with information isolation risk and Y departments in the previous level process is calculated, and the previous level department corresponding to the maximum correlation index is selected as the management department of the department in the process with information isolation risk. The calculation is performed for each department in the process with information isolation risk, and the process department with information isolation risk managed by each of the Y departments at the previous level is obtained as follows: , Indicates the 1st, 2nd, 3rd, ..., dth department in the process with information isolation risk managed by each department at the previous level;

[0026] S302. Calculate all departments in the process with information isolation risk, find the management department, and build a tree-structured on-site construction system in combination with the construction process chain. The first construction department in the facility worker process chain serves as the first management level. Then the management levels are determined in sequence according to the construction process chain. The last process in the facility worker process chain serves as the construction base level.

[0027] Use the process data of each department in history to determine the construction process chain, and mark the position of the on-site construction department in the construction process chain. Then find the processes with information isolation risks, calculate the correlation index between departments, and build an on-site construction system in combination with the construction process chain. When using the on-site construction system to analyze and manage the data during construction, circulate the information of all departments in the on-site construction system, use the upper-level department to manage the lower-level department, and use the tree-like management method to eliminate the information isolation problem of individual departments during existing construction, so that the information of all departments can be disseminated in the on-site construction system.

[0028] S400, installing sensors in each department of the on-site construction system, and using the sensors to collect on-site environmental data of each department during construction in real time; calculating accident thresholds based on environmental data of accidents that occurred during construction in the past, and using the accident thresholds to monitor and warn the construction site of each department in real time;

[0029] Furthermore, the specific steps of using accident thresholds to conduct real-time monitoring and early warning of the construction site of each department are as follows:

[0030] S401, collect the environmental data of accidents that occurred during construction in history, extract the values ​​of each environmental data before and after the accident, and calculate the difference between the values ​​of each environmental data before and after the accident , Ec represents the difference of each environmental data before and after the accident. represents the environmental data value before the accident, and E represents the environmental data value after the accident. The maximum value of all environmental data differences is extracted as the characteristic environmental data when the accident occurs in the construction;

[0031] S402, collecting characteristic environmental data values ​​when accidents occurred during construction in history, and calculating the accident threshold during construction, the formula is: ; In the formula, Ey represents the accident threshold during construction, It represents the average value of characteristic environmental data values ​​collected in the history when construction accidents occurred. It represents the standard deviation of the characteristic environmental data values ​​when accidents occurred in construction in the collected history;

[0032] S403, set up sensors in each department of the on-site construction system, use the sensors to collect real-time on-site environmental data of each department during construction, extract characteristic environmental data values ​​ESS from the on-site environmental data, use accident thresholds to judge the on-site characteristic environmental data values, and when When the corresponding department is at risk of an accident, an early warning is issued; When the corresponding department is judged to have no accident risk, the characteristic environmental data of all departments in the on-site construction system are judged, and the results of all early warning departments are , Indicates the 1st, 2nd, 3rd, ..., pth warning departments that have the risk of accidents.

[0033] S500. Determine the department that issues the warning. When there are multiple warning departments at the same process position in the on-site construction system and belonging to the same superior process, send all warning information to the superior process department, and the superior process department conducts a comprehensive analysis of all warning departments.

[0034] Furthermore, the specific steps for the superior process department to conduct a comprehensive analysis of all early warning departments are as follows:

[0035] S501. Find the number of warning departments Gy in the next level of departments managed by each department in all management levels in the on-site construction system. When the warning information is transmitted from the next level warning department to the corresponding management department at the previous level, the management department receives the characteristic environmental data value of each warning department at the next level. When the characteristic environmental data values ​​received from all warning departments are the same, the management department determines that the warning accident is contagious, and issues a preventive warning to all the departments at the next level under its management.

[0036] S502, when When an accident occurs, the next level early warning department does not need to transmit the early warning information to the upper level management department, and the early warning department handles the accident alone.

[0037] By using the upper-level management department to judge each early warning department, and then conducting a comprehensive analysis, on the one hand, the upper-level management department can take timely precautions against accidents with spreadability to avoid causing greater losses; on the other hand, when the next-level early warning department is judged to be 1, the early warning department can handle the accident on its own, reducing the system's information processing volume and enhancing the system's work efficiency.

[0038] An intelligent construction management system based on data analysis, the intelligent construction management system includes a data collection module, a department classification module, a field construction system building module, an early warning module and a comprehensive analysis module;

[0039] The data collection module is used to collect personnel data of each department during historical construction and environmental data when accidents occur;

[0040] The department classification module is used to analyze the personnel data of each department in history, build a classifier, and use the classifier to classify the personnel data during on-site construction to obtain the department type during on-site construction;

[0041] The on-site construction system building module is used to build a construction process chain based on historical data, calculate the correlation index between each department, and build an on-site construction system based on the construction process chain and the correlation index;

[0042] The early warning module is used to search for characteristic environmental data of accidents that occurred during construction in the past, and calculate the accident threshold during construction using the characteristic environmental data;

[0043] The comprehensive analysis module is used to determine the number of warning departments under each management department after determining all the warning departments. The management department determines the warning information of all the warning departments it manages and issues preventive warnings to all the lower-level departments it manages.

[0044] The department classification module includes a classifier building unit and a real-time classification unit;

[0045] The classifier construction unit is used to analyze the personnel data of each department in the historical construction, extract feature data, use the feature data to build a decision tree, and superimpose all decision trees to build a classifier;

[0046] The real-time classification unit is used to collect personnel data of on-site construction, extract on-site feature data from the personnel data, use a classifier to judge the on-site feature data, and obtain the department type of the on-site construction.

[0047] The on-site construction system building module includes a construction process chain unit and a correlation index calculation unit;

[0048] The construction process chain unit is used to analyze the process data of each department in history, build a construction process chain, collect the process data of the on-site construction department, determine the position of the on-site construction department in the construction process chain and mark it;

[0049] The correlation index calculation unit is used to judge the process with information isolation risk, calculate the correlation index of each department and the upper-level department in the process with information isolation risk, and build an on-site construction system in combination with the construction process chain.

[0050] The early warning module includes a threshold calculation unit and a real-time early warning unit;

[0051] The threshold calculation unit is used to analyze the environmental data of construction failures in history, find characteristic environmental data, and calculate the construction accident threshold using the characteristic environmental data;

[0052] The real-time early warning unit is used to judge the characteristic environmental data in each department of the on-site construction, and issue an early warning when it is judged that there is a risk of an accident in the on-site construction department.

[0053] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0054] 1. The present invention utilizes decision tree superposition to construct a classifier, realizes the classification of on-site construction personnel data, obtains the type of on-site construction department, and utilizes automated technical means to judge the type of on-site construction department, thereby avoiding the uncertainty caused by human subjective judgment; and adopts a method of superimposing multiple decision trees to construct a classifier, which can reduce the probability of classifier classification errors, and uses multiple decision trees for comprehensive judgment, thereby eliminating the errors in the judgment of a single decision tree.

[0055] 2. The present invention constructs an on-site construction system based on historical data, and uses the on-site construction system to analyze and process information during construction. Through a tree-like management approach, the problem of information isolation of a single department during existing construction is eliminated, so that information from all departments can be disseminated in the on-site construction system.

[0056] 3. The present invention utilizes the upper-level management department to judge each early warning department, and then conducts a comprehensive analysis; on the one hand, the upper-level management department takes timely precautions against accidents with spreadability to avoid causing greater losses; on the other hand, when the next-level early warning department is judged to be 1, the early warning department can handle the accident on its own, reducing the amount of information processing in the system and enhancing the system's work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0058] Figure 1 It is a module distribution diagram of an intelligent construction management system based on data analysis of the present invention;

[0059] Figure 2 It is a schematic diagram of the steps of an intelligent construction management method based on data analysis of the present invention;

[0060] Figure 3 It is a schematic diagram of a field construction system of an intelligent construction management method based on data analysis of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] See also Figure 1-Figure 3 , the present invention provides a technical solution:

[0063] An intelligent construction management method based on data analysis, the method comprising the following steps:

[0064] S100, collecting personnel data of each department in the historical construction process, extracting characteristic data of the construction department in the history, calculating each extracted characteristic data to generate a decision tree, merging the decision data of all characteristic data to construct a classifier, and using the classifier to judge the data of all personnel involved in the construction during the on-site construction to obtain the department category of the on-site construction;

[0065] The specific steps for classifying on-site construction departments according to the characteristics of construction personnel are as follows:

[0066] S101. Collect the personnel data of each department in the historical construction process. , Indicates collecting the 1st, 2nd, 3rd, ..., nth personnel data of each department in the historical construction process, where n is a positive integer; comparing the collected personnel data of each department in the history, extracting the personnel data existing in each department as feature data, and constructing the feature data set as , represents the 1st, 2nd, 3rd, ..., mth characteristic data of the extracted construction department, where m is a positive integer;

[0067] S102. After determining the characteristic data of the construction department, collect the same characteristic data values ​​of each department in history. , Indicates the same characteristic data values ​​of the 1st, 2nd, 3rd, ..., jth departments in the collected history, where j is a positive integer; collect the same characteristic data values ​​of each department in the history k times, and extract the maximum value of the k same characteristic data of each department in the collected history and minimum value ; Sort the maximum and minimum values ​​of the extracted j departments from small to large to get , Indicates the maximum and minimum values ​​of the same characteristic data of the 1st, 2nd, 3rd, ..., jth departments after sorting from small to large; compares the previous maximum value and the next minimum value in adjacent departments. When the adjacent departments have overlapping values ​​of the same characteristic data, Indicates the maximum value of the same characteristic data of the previous department in the adjacent department. Indicates the minimum number of households with the same characteristics in the next department among adjacent departments; When , it is determined that there is no overlap in the same characteristic data values ​​of adjacent departments;

[0068] S103. When it is determined that adjacent departments have overlapping characteristic data, the maximum value of the previous department and the minimum value of the next department in the adjacent departments are replaced. The formula is:

[0069] ,

[0070] In the formula, Th represents the replacement value for replacing the maximum value of the previous department and the minimum value of the next department in the adjacent department; the data interval of the adjacent departments that are interlaced after replacement is and ; represents the interval after the replacement of the previous department in the adjacent department, Represents the interval after replacement of the next department in the adjacent departments; Use the calculated replacement value to replace the maximum and minimum values ​​of all departments with overlaps, and finally obtain the interval of the same feature data in all departments; Collect and analyze the intervals of all feature data in all departments in the feature data set to obtain the interval of each feature data in all departments; Use the interval of each feature data in all departments to design a decision tree, and construct m decision trees as follows , Indicates the 1st, 2nd, 3rd, ..., mth decision trees constructed using feature data; the m constructed decision trees are superimposed to establish a classifier;

[0071] S104, in the on-site construction, real-time feature data of each personnel data in the feature data set is collected in real time, and the real-time feature data of each personnel data collected is judged by using a classifier. Suppose the real-time feature data of each personnel data collected is , represents the 1st, 2nd, 3rd, ..., Lth feature data in each collected personnel data, where L is a positive integer; the L types of real-time feature data collected are judged by the classifier and the L types of classification results are obtained: , It means that the 1st, 2nd, 3rd, ..., Lth classification results are judged by the classifier on L kinds of real-time feature data, and the L classification results are comprehensively analyzed. The formula is:

[0072] ,

[0073] In the formula, Re represents the comprehensive probability of each classification result after classification, H represents the number of the same classification results, and Ts represents the collected real-time feature data. Represents the interval boundary value of the department in each result after classification, include and ; Calculate the difference between the collected real-time feature data and the boundary value, and select the minimum value of the two differences; Calculate the comprehensive probability of each judgment result, and select the result with the largest comprehensive probability as the judgment result of the corresponding personnel data; Use the classifier to judge each personnel data in the on-site construction and obtain all the department types. , Represents the 1st, 2nd, 3rd, ..., gth department in on-site construction, where g is a positive integer.

[0074] By analyzing the personnel data of each department in history, extracting feature data to build a decision tree, and using decision trees to superimpose and build a classifier, the data of on-site construction personnel can be classified, the type of on-site construction department can be obtained, and the type of on-site construction department can be judged using automated technical means to avoid the uncertainty caused by human subjective judgment; and the method of superimposing multiple decision trees to build a classifier can reduce the probability of classifier classification errors, and use multiple decision trees for comprehensive judgment to eliminate the errors in the judgment of a single decision tree.

[0075] S200, after classifying the on-site construction departments, collect the process data of the historical construction and build the on-site construction process chain; analyze the process data of each construction department, find the position of each construction department in the construction process chain and mark it;

[0076] The specific steps to find the position of each construction department in the construction process chain and mark it are:

[0077] S201. Collect the process data of each department during construction in history. , Represents the process data of the 1st, 2nd, 3rd, ..., jth departments in the collected history. The collected process data of each department are compared, and the department corresponding to the minimum value of the process data is extracted as the first construction department. Then, each department in the collected history is sorted in order from small to large according to the process data to build a construction process chain as , represents the 1st, 2nd, 3rd, ..., fth process in the construction process of the build, where f is a positive integer;

[0078] S202, collect the process data of each department in the on-site construction in real time, use the collected process data to find the process position of each department in the on-site construction in the construction process chain and mark it in each department. , It means marking the positions of the 1st, 2nd, 3rd, ..., gth departments in the construction process chain during on-site construction.

[0079] S300, after marking the position of each department in the construction process chain, calculating the correlation index between departments at adjacent positions in the construction process chain; using the calculated correlation index and the construction process chain to construct an on-site construction system;

[0080] The specific steps of building a field construction system using the calculated correlation index and construction process chain are as follows:

[0081] S301, when judging the number of departments in each process in the construction process chain, collecting the number of departments in each process in the construction process chain , Indicates the number of departments in the 1st, 2nd, 3rd, ..., fth processes in the collected construction process chain; the number of departments in each process is judged. , determine whether there is information isolation risk in the corresponding process; make a judgment on each process in the construction process chain to obtain all processes with information isolation risk; calculate the correlation index between all departments of the process with information isolation risk and each department in the previous process, assuming that the number of departments in the previous process is Y, and collect the number and amount of information exchanged between every two departments in history. The formula is:

[0082] ,

[0083] In the formula, Ga represents the correlation index between each department in the process with information isolation risk and each department at the previous level, X represents the number of information exchanges between the two departments, and Q represents the amount of information when the two departments exchange information; the correlation index between each department in the process with information isolation risk and Y departments in the previous level process is calculated, and the previous level department corresponding to the maximum correlation index is selected as the management department of the department in the process with information isolation risk. The calculation is performed for each department in the process with information isolation risk, and the process department with information isolation risk managed by each of the Y departments at the previous level is obtained as follows: , Indicates the 1st, 2nd, 3rd, ..., dth department in the process with information isolation risk managed by each department at the previous level;

[0084] S302. Calculate all departments in the process with information isolation risk, find the management department, and build a tree-structured on-site construction system in combination with the construction process chain. The first construction department in the facility worker process chain serves as the first management level. Then the management levels are determined in sequence according to the construction process chain. The last process in the facility worker process chain serves as the construction base level.

[0085] Use the process data of each department in history to determine the construction process chain, and mark the position of the on-site construction department in the construction process chain. Then find the processes with information isolation risks, calculate the correlation index between departments, and build an on-site construction system in combination with the construction process chain. When using the on-site construction system to analyze and manage the data during construction, circulate the information of all departments in the on-site construction system, use the upper-level department to manage the lower-level department, and use the tree-like management method to eliminate the information isolation problem of individual departments during existing construction, so that the information of all departments can be disseminated in the on-site construction system.

[0086] S400, installing sensors in each department of the on-site construction system, and using the sensors to collect on-site environmental data of each department during construction in real time; calculating accident thresholds based on environmental data of accidents that occurred during construction in the past, and using the accident thresholds to monitor and warn the construction site of each department in real time;

[0087] The specific steps for using accident thresholds to conduct real-time monitoring and early warning of the construction site of each department are as follows:

[0088] S401, collect the environmental data of accidents that occurred during construction in history, extract the values ​​of each environmental data before and after the accident, and calculate the difference between the values ​​of each environmental data before and after the accident , Ec represents the difference of each environmental data before and after the accident. represents the environmental data value before the accident, and E represents the environmental data value after the accident. The maximum value of all environmental data differences is extracted as the characteristic environmental data when the accident occurs in the construction;

[0089] S402, collecting characteristic environmental data values ​​when accidents occurred during construction in history, and calculating the accident threshold during construction, the formula is: ; In the formula, Ey represents the accident threshold during construction, It represents the average value of characteristic environmental data values ​​collected in the history when construction accidents occurred. It represents the standard deviation of the characteristic environmental data values ​​when accidents occurred in construction in the collected history;

[0090] S403, set up sensors in each department of the on-site construction system, use the sensors to collect real-time on-site environmental data of each department during construction, extract characteristic environmental data values ​​ESS from the on-site environmental data, use accident thresholds to judge the on-site characteristic environmental data values, and when When the corresponding department is at risk of an accident, an early warning is issued; When the corresponding department is judged to have no accident risk, the characteristic environmental data of all departments in the on-site construction system are judged, and the results of all early warning departments are , Indicates the 1st, 2nd, 3rd, ..., pth warning departments that have the risk of accidents.

[0091] S500. Determine the department that issues the warning. When there are multiple warning departments at the same process position in the on-site construction system and belonging to the same superior process, send all warning information to the superior process department, and the superior process department conducts a comprehensive analysis of all warning departments.

[0092] The specific steps for the superior process department to conduct a comprehensive analysis of all early warning departments are:

[0093] S501. Find the number of warning departments Gy in the next level of departments managed by each department in all management levels in the on-site construction system. When the warning information is transmitted from the next level warning department to the corresponding management department at the previous level, the management department receives the characteristic environmental data value of each warning department at the next level. When the characteristic environmental data values ​​received from all warning departments are the same, the management department determines that the warning accident is contagious, and issues a preventive warning to all the departments at the next level under its management.

[0094] S502, when When an accident occurs, the next level early warning department does not need to transmit the early warning information to the upper level management department, and the early warning department handles the accident alone.

[0095] By using the upper-level management department to judge each early warning department, and then conducting a comprehensive analysis, on the one hand, the upper-level management department can take timely precautions against accidents with spreadability to avoid causing greater losses; on the other hand, when the next-level early warning department is judged to be 1, the early warning department can handle the accident on its own, reducing the system's information processing volume and enhancing the system's work efficiency.

[0096] An intelligent construction management system based on data analysis, the intelligent construction management system includes a data collection module, a department classification module, a field construction system building module, an early warning module and a comprehensive analysis module;

[0097] The data collection module is used to collect personnel data of each department during historical construction and environmental data when accidents occur;

[0098] The department classification module is used to analyze the personnel data of each department in history, build a classifier, and use the classifier to classify the personnel data during on-site construction to obtain the department type during on-site construction;

[0099] The on-site construction system building module is used to build a construction process chain based on historical data, calculate the correlation index between each department, and build an on-site construction system based on the construction process chain and the correlation index;

[0100] The early warning module is used to search for characteristic environmental data of accidents that occurred during construction in the past, and calculate the accident threshold during construction using the characteristic environmental data;

[0101] The comprehensive analysis module is used to determine the number of warning departments under each management department after determining all the warning departments. The management department determines the warning information of all the warning departments it manages and issues preventive warnings to all the lower-level departments it manages.

[0102] The department classification module includes a classifier building unit and a real-time classification unit;

[0103] The classifier construction unit is used to analyze the personnel data of each department in the historical construction, extract feature data, use the feature data to build a decision tree, and superimpose all decision trees to build a classifier;

[0104] The real-time classification unit is used to collect personnel data of on-site construction, extract on-site feature data from the personnel data, use a classifier to judge the on-site feature data, and obtain the department type of the on-site construction.

[0105] The on-site construction system building module includes a construction process chain unit and a correlation index calculation unit;

[0106] The construction process chain unit is used to analyze the process data of each department in history, build a construction process chain, collect the process data of the on-site construction department, determine the position of the on-site construction department in the construction process chain and mark it;

[0107] The correlation index calculation unit is used to judge the process with information isolation risk, calculate the correlation index of each department and the upper-level department in the process with information isolation risk, and build an on-site construction system in combination with the construction process chain.

[0108] The early warning module includes a threshold calculation unit and a real-time early warning unit;

[0109] The threshold calculation unit is used to analyze the environmental data of construction failures in history, find characteristic environmental data, and calculate the construction accident threshold using the characteristic environmental data;

[0110] The real-time early warning unit is used to judge the characteristic environmental data in each department of the on-site construction, and issue an early warning when it is judged that there is a risk of an accident in the on-site construction department.

[0111] Example 1: Now manage the construction in a certain construction site, collect the personnel data of each department in the history, and extract the characteristic data including location data, work site temperature data and workload data;

[0112] Assume that the construction units in history are project management department, design department, and construction department; calculate the maximum and minimum values ​​of the three characteristic data of each department in history, for example: the maximum and minimum values ​​of the workload data in the three departments are (2-15), (12-20), and (20-30) respectively; determine that there is an overlap between the project management department and the design department, calculate the replacement value to be 13.5, and obtain the new workload data intervals of the three departments as (2-13.5), (13.5-20), and (20-30); similarly calculate the intervals of the other two characteristic data and construct a decision tree, and superimpose the three decision trees to construct a classifier;

[0113] Collect the location data, work site temperature data and workload data of the on-site construction personnel data in real time; use the classifier to judge, and judge which department the on-site construction personnel data belongs to according to each decision tree; after judging the three characteristic data of a construction personnel data on site, the results are project management department, design department, and design department respectively; calculate the comprehensive probability of each classification result, the formula is:

[0114] ,

[0115] Assume that in the three classification judgments of the personnel data, the minimum difference between the real-time feature data and the interval is 6, 3, and 4 respectively, and the comprehensive probabilities of the two judgment results are calculated to be 6 and 7 respectively; finally, it is judged that the personnel data belongs to the design department; the same method is used to judge all the personnel data in the on-site construction;

[0116] A classifier is constructed, which includes a location decision tree, a temperature decision tree and a workload decision tree;

[0117] The personnel data in the on-site construction was extracted to obtain the location data and workload data. The classifier was used to determine the types of on-site departments, namely the planning department, design department, construction department and security inspection department.

[0118] Collect the process data of each department in history to obtain the construction process chain for the project management department Planning Department, Design Department Construction Department, Quality Inspection Department, Security Inspection Department Finance Department;

[0119] Collect the process data of the on-site construction department and mark the positions in the construction process chain as 2, 3, and 5; calculate and judge that the process with information isolation risk is process 3; calculate the correlation index as 10 for the planning department-construction department, 12 for the planning department-security department, 20 for the design department-construction department, and 15 for the design department-security department; after judgment, it is found that the planning department manages the security department, and the design department manages the construction department; construct a tree-like on-site construction system such as Figure 3 As shown;

[0120] The accident threshold calculated through historical data is 80, and the characteristic environmental data collected in real time from the four departments are 78, 79, 83, and 82; it is judged that the next-level early warning departments managed by the planning department and the design department are both 1, so the early warning department does not need to transmit the early warning information to the management department.

[0121] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0122] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent construction management method based on data analysis, characterized in that: The method comprises the following steps: S100, collecting personnel data of each department in the historical construction process, extracting characteristic data of the construction department in the history, calculating each extracted characteristic data to generate a decision tree, merging the decision data of all characteristic data to construct a classifier, and using the classifier to judge the data of all personnel involved in the construction during the on-site construction to obtain the department category of the on-site construction; The specific steps for classifying on-site construction departments according to the characteristics of construction personnel are as follows: S101. Collect the personnel data of each department in the historical construction process. , Indicates collecting the 1st, 2nd, 3rd, ..., nth personnel data of each department in the historical construction process, where n is a positive integer; comparing the collected personnel data of each department in the history, extracting the personnel data existing in each department as feature data, and constructing the feature data set as , represents the 1st, 2nd, 3rd, ..., mth characteristic data of the extracted construction department, where m is a positive integer; S102. After determining the characteristic data of the construction department, collect the same characteristic data values ​​of each department in history. , Indicates the same characteristic data values ​​of the 1st, 2nd, 3rd, ..., jth departments in the collected history, where j is a positive integer; collect the same characteristic data values ​​of each department in the history k times, and extract the maximum value of the k same characteristic data of each department in the collected history and minimum value ; Sort the maximum and minimum values ​​of the extracted j departments from small to large to get , Indicates the maximum and minimum values ​​of the same characteristic data of the 1st, 2nd, 3rd, ..., jth departments after sorting from small to large; compares the previous maximum value and the next minimum value in adjacent departments. When the adjacent departments have overlapping values ​​of the same characteristic data, Indicates the maximum value of the same characteristic data of the previous department in the adjacent department. Indicates the minimum number of households with the same characteristics in the next department among adjacent departments; When , it is determined that there is no overlap in the same characteristic data values ​​of adjacent departments; S103. When it is determined that adjacent departments have overlapping characteristic data, the maximum value of the previous department and the minimum value of the next department in the adjacent departments are replaced. The formula is: ， In the formula, Th represents the replacement value for replacing the maximum value of the previous department and the minimum value of the next department in the adjacent department; the data interval of the adjacent departments that are staggered after replacement is and ; represents the interval after the replacement of the previous department in the adjacent department, Represents the interval after replacement of the next department in the adjacent departments; Use the calculated replacement value to replace the maximum and minimum values ​​of all departments with overlaps, and finally obtain the interval of the same feature data in all departments; Collect and analyze the intervals of all feature data in all departments in the feature data set to obtain the interval of each feature data in all departments; Use the interval of each feature data in all departments to design a decision tree, and construct m decision trees as follows , Indicates the 1st, 2nd, 3rd, ..., mth decision trees constructed using feature data; the m constructed decision trees are superimposed to establish a classifier; S104, in the on-site construction, real-time feature data of each personnel data in the feature data set is collected in real time, and the real-time feature data of each personnel data collected is judged by using a classifier. Suppose the real-time feature data of each personnel data collected is , represents the 1st, 2nd, 3rd, ..., Lth feature data in each collected personnel data, where L is a positive integer; the L types of real-time feature data collected are judged by the classifier and the L types of classification results are obtained: , It means that the 1st, 2nd, 3rd, ..., Lth classification results are judged by the classifier on L kinds of real-time feature data, and the L classification results are comprehensively analyzed. The formula is: ， In the formula, Re represents the comprehensive probability of each classification result after classification, H represents the number of the same classification results, and Ts represents the collected real-time feature data. Represents the interval boundary value of the department in each result after classification, include and ; Calculate the difference between the collected real-time feature data and the boundary value, and select the minimum value of the two differences; Calculate the comprehensive probability of each judgment result, and select the result with the largest comprehensive probability as the judgment result of the corresponding personnel data; Use the classifier to judge each personnel data in the on-site construction and obtain all the department types. , represents the 1st, 2nd, 3rd, ..., gth type of departments in the on-site construction, where g is a positive integer; S200, after classifying the on-site construction departments, collect the process data of the historical construction and build the on-site construction process chain; analyze the process data of each construction department, find the position of each construction department in the construction process chain and mark it; S300, after marking the position of each department in the construction process chain, calculating the correlation index between departments at adjacent positions in the construction process chain; using the calculated correlation index and the construction process chain to construct an on-site construction system; S400, installing sensors in each department of the on-site construction system, and using the sensors to collect on-site environmental data of each department during construction in real time; calculating accident thresholds based on environmental data of accidents that occurred during construction in the past, and using the accident thresholds to monitor and warn the construction site of each department in real time; S500. Determine the department that issues the warning. When there are multiple warning departments at the same process position in the on-site construction system and belonging to the same superior process, send all warning information to the superior process department, and the superior process department conducts a comprehensive analysis of all warning departments.

2. The intelligent construction management method based on data analysis according to claim 1 is characterized by: The specific steps of finding the position of each construction department in the construction process chain and marking it in S200 are: S201. Collect the process data of each department during construction in history. , Represents the process data of the 1st, 2nd, 3rd, ..., jth departments in the collected history. The collected process data of each department are compared, and the department corresponding to the minimum value of the process data is extracted as the first construction department. Then, each department in the collected history is sorted in order from small to large according to the process data to build a construction process chain as , represents the 1st, 2nd, 3rd, ..., fth process in the construction process of the build, where f is a positive integer; S202, collect the process data of each department in the on-site construction in real time, use the collected process data to find the process position of each department in the on-site construction in the construction process chain and mark it in each department. , It means marking the positions of the 1st, 2nd, 3rd, ..., gth departments in the construction process chain during on-site construction.

3. The intelligent construction management method based on data analysis according to claim 2 is characterized by: The specific steps of constructing the on-site construction system using the calculated correlation index and the construction process chain in S300 are: S301, when judging the number of departments in each process in the construction process chain, collecting the number of departments in each process in the construction process chain , It represents the number of departments of the 1st, 2nd, 3rd, ..., fth processes in the collected construction process chain; Determine the number of departments in each process. When the corresponding process is identified, it is determined that there is information isolation risk in the corresponding process; each process in the construction process chain is judged to obtain all processes with information isolation risk; Calculate the correlation index between all departments in the process with information isolation risk and each department in the previous process. Assume that the number of departments in the previous process is Y. Collect the number and amount of information exchanged between every two departments in history. The formula is: ， In the formula, Ga represents the correlation index between each department and each department at the previous level in the process with information isolation risk, X represents the number of information exchanges between the two departments, and Q represents the amount of information when the two departments exchange information; Calculate the correlation index between each department in the process with information isolation risk and Y departments in the previous process, select the previous department corresponding to the maximum correlation index as the management department of the department in the process with information isolation risk, calculate each department in the process with information isolation risk, and get the process department with information isolation risk managed by each of the Y departments in the previous level. , Indicates the 1st, 2nd, 3rd, ..., dth department in the process with information isolation risk managed by each department at the previous level; S302. Calculate all departments in the process with information isolation risk, find the management department, and build a tree-structured on-site construction system in combination with the construction process chain. The first construction department in the facility worker process chain serves as the first management level. Then the management levels are determined in sequence according to the construction process chain. The last process in the facility worker process chain serves as the construction base level.

4. The intelligent construction management method based on data analysis according to claim 3 is characterized by: The specific steps of using the accident threshold to monitor and warn the construction site of each department in real time in S400 are: S401, collect the environmental data of accidents that occurred during construction in history, extract the values ​​of each environmental data before and after the accident, and calculate the difference between the values ​​of each environmental data before and after the accident , Ec represents the difference of each environmental data before and after the accident. represents the environmental data value before the accident, and E represents the environmental data value after the accident. The maximum value of all environmental data differences is extracted as the characteristic environmental data when the accident occurs in the construction; S402, collecting characteristic environmental data values ​​when accidents occurred during construction in history, and calculating the accident threshold during construction, the formula is: ; In the formula, Ey represents the accident threshold during construction, It represents the average value of characteristic environmental data values ​​collected in the history when construction accidents occurred. It represents the standard deviation of the characteristic environmental data values ​​when accidents occurred in construction in the collected history; S403, set up sensors in each department of the on-site construction system, use the sensors to collect real-time on-site environmental data of each department during construction, extract characteristic environmental data values ​​ESS from the on-site environmental data, use accident thresholds to judge the on-site characteristic environmental data values, and when When the corresponding department is at risk of an accident, an early warning is issued; When the corresponding department is judged to have no accident risk, the characteristic environmental data of all departments in the on-site construction system are judged, and the results of all early warning departments are , Indicates the 1st, 2nd, 3rd, ..., pth warning departments that have the risk of accidents.

5. The intelligent construction management method based on data analysis according to claim 4 is characterized in that: The specific steps for the superior process department in S500 to conduct a comprehensive analysis of all early warning departments are: S501. Find the number of warning departments Gy in the next level of departments managed by each department in all management levels in the on-site construction system. When the warning information is transmitted from the next level warning department to the corresponding management department at the previous level, the management department receives the characteristic environmental data value of each warning department at the next level. When the characteristic environmental data values ​​received from all warning departments are the same, the management department determines that the warning accident is contagious, and issues a preventive warning to all the departments at the next level under its management. S502, when When an accident occurs, the next level early warning department does not need to transmit the early warning information to the upper level management department, and the early warning department handles the accident alone.

6. An intelligent construction management system based on data analysis using an intelligent construction management method based on data analysis as claimed in any one of claims 1 to 5, characterized in that: The intelligent construction management system includes data collection module, department classification module, on-site construction system building module, early warning module and comprehensive analysis module; The data collection module is used to collect personnel data of each department during historical construction and environmental data when accidents occur; The department classification module is used to analyze the personnel data of each department in history, build a classifier, and use the classifier to classify the personnel data during on-site construction to obtain the department type during on-site construction; The on-site construction system building module is used to build a construction process chain based on historical data, calculate the correlation index between each department, and build an on-site construction system based on the construction process chain and the correlation index; The early warning module is used to search for characteristic environmental data of accidents that occurred during construction in the past, and calculate the accident threshold during construction using the characteristic environmental data; The comprehensive analysis module is used to determine the number of warning departments under each management department after determining all the warning departments. The management department determines the warning information of all the warning departments it manages and issues preventive warnings to all the lower-level departments it manages.

7. The intelligent construction management system based on data analysis according to claim 6 is characterized by: The department classification module includes a classifier construction unit and a real-time classification unit; The classifier construction unit is used to analyze the personnel data of each department in the historical construction, extract feature data, use the feature data to build a decision tree, and superimpose all decision trees to build a classifier; The real-time classification unit is used to collect personnel data of on-site construction, extract on-site feature data from the personnel data, use a classifier to judge the on-site feature data, and obtain the department type of the on-site construction.

8. The intelligent construction management system based on data analysis according to claim 6 is characterized by: The on-site construction system building module includes a construction process chain unit and a correlation index calculation unit; The construction process chain unit is used to analyze the process data of each department in history, build a construction process chain, collect the process data of the on-site construction department, determine the position of the on-site construction department in the construction process chain and mark it; The correlation index calculation unit is used to judge the process with information isolation risk, calculate the correlation index of each department and the upper-level department in the process with information isolation risk, and build an on-site construction system in combination with the construction process chain.

9. The intelligent construction management system based on data analysis according to claim 6 is characterized by: The early warning module includes a threshold calculation unit and a real-time early warning unit; The threshold calculation unit is used to analyze the environmental data of construction failures in history, find characteristic environmental data, and calculate the construction accident threshold using the characteristic environmental data; The real-time early warning unit is used to judge the characteristic environmental data in each department of the on-site construction, and issue an early warning when it is judged that there is a risk of an accident in the on-site construction department.

Images