Risk management method and system

Abstract

The embodiment of the invention provides a risk management method and system. The method comprises the following steps: acquiring a target project parameter value of a target project at target time; determining a risk coefficient of the target project based on the target project parameter value; based on the risk coefficient and a target risk threshold value, determining whether the target project needs to be inspected at the target time, the target risk threshold value being determined based on the following modes: obtaining an initial risk threshold value and historical data of a first reference project; based on the historical data and the initial risk threshold, generating an inspection judgment model which is a trained machine learning model; and determining a target risk threshold based on the inspection judgment model and the initial risk threshold.

Description

Technical Field

[0001] The present invention relates to the field of engineering management, and in particular to a risk management method and system. Background Art

[0002] In the field of construction, risk management is crucial to ensure the safety, quality, progress and cost control of the project. However, traditional risk management methods mainly rely on historical experience or subjective inferences of experts, which may make the risk management of the project inconsistent with the actual situation, thus affecting the quality of the project. The complexity of construction projects, the variability of the construction environment, and the high mobility of construction projects make the risk assessment of the project very difficult and complicated, increasing the difficulty of risk control. How to scientifically and accurately manage the risk of the project is a topic worth studying.

[0003] Therefore, a risk management method and system are provided to improve the efficiency and accuracy of risk assessment. Summary of the invention

[0004] One of the embodiments of the present invention provides a risk management method, including: obtaining a target project parameter value of a target project at a target time; determining a risk coefficient of the target project based on the target project parameter value; determining whether the target project needs inspection at the target time based on the risk coefficient and a target risk threshold, wherein the target risk threshold is determined based on the following method: obtaining an initial risk threshold and historical data of a first reference project; generating an inspection judgment model based on the historical data and the initial risk threshold, wherein the inspection judgment model is a trained machine learning model; determining the target risk threshold based on the inspection judgment model and the initial risk threshold.

[0005] One of the embodiments of the present invention provides a risk management system, comprising: an acquisition module, configured to acquire a target project parameter value of a target project at a target time; a risk assessment module, configured to determine the risk coefficient of the target project based on the target project parameter value; an inspection determination module, configured to determine whether the target project needs inspection at the target time based on the risk coefficient and a target risk threshold; a risk threshold determination module, configured to: acquire an initial risk threshold and historical data of a first reference project; generate an inspection judgment model based on the historical data and the initial risk threshold, the inspection judgment model being a trained machine learning model; and determine the target risk threshold based on the inspection judgment model and the initial risk threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The present invention will be further described in the form of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0007] Figure 1 is a schematic diagram of an application scenario of a risk management system according to some embodiments of the present invention;

[0008] Figure 2 is a schematic diagram of modules of a risk management system according to some embodiments of the present invention;

[0009] Figure 3 is an exemplary schematic diagram of a risk management method according to some embodiments of the present invention;

[0010] Figure 4 is an exemplary flow chart of a method for generating an inspection judgment model according to some embodiments of the present invention;

[0011] Figure 5 is an exemplary flow chart of a method for determining a target risk threshold according to some embodiments of the present invention;

[0012] Figure 6 is an exemplary flow chart of a method for determining recommended inspection items according to some embodiments of the present invention. DETAILED DESCRIPTION

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of the present invention. For ordinary technicians in this field, the present invention can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0014] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0015] Unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0016] The present invention uses a flow chart to illustrate the operations performed by the system according to an embodiment of the present invention. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, the various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 is a schematic diagram of an application scenario of a risk management system according to some embodiments of the present invention.

[0018] like Figure 1 As shown, the application scenario 100 of the risk management system may include a processing device 110 , a network 120 , a terminal 130 , a storage device 140 and an engineering information database 150 .

[0019] The processing device 110 can process the data and / or information obtained from the terminal 130, the storage device 140 and the project information database 150. For example, the processing device 110 can obtain the project-related information from the project information database 150, and determine the risk factor of the target project based on the project-related information. For another example, the processing device 110 can determine whether the target project needs to be inspected at the target time based on the risk factor and the target risk threshold of the target project. More information about the risk factor and the target risk threshold can also be found elsewhere in the present invention (e.g. Figure 3 ).

[0020] In some embodiments, processing device 110 can be a single server or a server group. In some embodiments, processing device 110 can be local or remote. Processing device 110 can be directly connected to terminal 130, storage device 140 and engineering information database 150 to access stored or acquired information and / or data. In some embodiments, processing device 110 can be implemented on a cloud platform. As an example only, a cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc. or any combination thereof. In some embodiments, processing device 110 can be a distributed server group, which can include multiple server nodes.

[0021] The network 120 may include any suitable network that facilitates information and / or data exchange for the risk management system application scenario 100. In some embodiments, one or more components of the risk management system application scenario 100 (e.g., the terminal 130, the processing device 110, the storage device 140, or the task information database 150) may communicate information and / or data with one or more other components of the risk management system application scenario 100 via the network 120. For example, the processing device 110 may obtain project-related information from the storage device 140 and / or the project information database 150 via the network 120.

[0022] In some embodiments, the network 120 may be any one or more of a wired network or a wireless network. In some embodiments, the network may be a point-to-point, shared, centralized, or other topological structure or a combination of multiple topological structures.

[0023] The terminal 130 may include a mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, etc., or any combination thereof. In some embodiments, the terminal 130 may interact with other components in the application scenario 100 of the risk management system through the network 120. In some embodiments, the terminal 130 may receive information and / or instructions input by a user, and send the received information and / or instructions to the processing device 110 via the network 120. For example, the terminal 130 may receive instructions from a user (such as an engineering manager or an inspector), obtain engineering-related information from the storage device 140 and / or the engineering information database 150 through the network 120, and present it to the user. In some embodiments, the terminal 130 may also send and / or present risk warning information to the user based on the risk factor of the target project. Risk warning information includes, but is not limited to, voice, text, image, video and other information.

[0024] In some embodiments, the terminal 130 also includes various types of collection devices (not shown in the figure), which are configured to collect various types of data related to the project. For example, it may include but is not limited to video equipment (such as cameras, infrared cameras, etc.), environmental monitoring equipment (such as temperature collection equipment, humidity collection equipment), personnel self-service check-in equipment (such as access control equipment, card check-in equipment), etc. The collection equipment can be deployed according to actual needs. For example, it can be deployed at the construction site or office location of the target project.

[0025] In some embodiments, the collection device can send the collected data related to the project to the project information database 150, the storage device 140 and / or the processing device 110 through the network 120 for storage, analysis and / or processing. In some embodiments, the collected data of the collection device can be used to generate the target project parameter value (such as employee attendance rate) of the target project. For example, the collection device (such as a self-service check-in device for personnel) can send the check-in information (such as card swiping record / attendance record) of each employee on that day to the project information database 150 through the network 120 to generate the employee attendance record of the target project. The processing device 110 can analyze and / or process (such as statistical processing) the personnel check-in information and / or employee attendance records to determine the personnel attendance information (such as the number of employees present, the number of latecomers, the employee attendance rate, the lateness rate, etc.). More content about the target project and the target project parameter value can be found elsewhere in the present invention (for example, Figure 3 ).

[0026] In some embodiments of the present invention, data related to the project can be automatically and real-time collected through the collection device, thereby providing a reliable and effective data basis for risk assessment of the target project.

[0027] The storage device 140 may store data and / or instructions. In some embodiments, the storage device 140 may store data obtained from the processing device 110, the terminal 130 and / or the engineering information database 150. For example, the storage device 140 may store data (such as engineering related information) obtained from the engineering information database 150, etc. In some embodiments, the storage device 140 may store data and / or instructions used by the processing device 110 to execute the exemplary methods described in the present invention. For example, the storage device 140 may store instructions for the processing device 110 to execute the methods shown in each flowchart. In some embodiments, the storage device 140 may include a mass storage device, a removable storage device, a volatile read-write memory, a read-only memory (ROM), etc., or any combination thereof. In some embodiments, the storage device 140 may be implemented on a cloud platform. In some embodiments, the storage device 140 may be part of the processing device 110.

[0028] The project information database 150 refers to a source for providing data related to the project, for example, it may be a database of an operating entity (such as an enterprise, a construction unit, etc.) and / or a third-party service platform (such as an information service provider).

[0029] In some embodiments, the project information database 150 is used to provide project-related information of the target project. Project-related information includes, but is not limited to, the basic project information of the target project and various types of project management information. For example, the basic project information includes the name of the target project, the construction unit, the region, the construction period, etc. The project management information includes, but is not limited to, the task allocation information of the target project, the task acceptance information of the employees (such as the task acceptance rate of the workers), the attendance information of the personnel, the quality and safety records of the task orders, the work record of the team leader for the employees, the payment of employee salaries, and the fatigue work of the employees. It should be noted that the project-related information can be information in various forms such as text, graphics, sound, and video.

[0030] In some embodiments, the project information database 150 can interact with other components in the application scenario 100 of the risk management system through the network 120. For example, it can send project-related information to the processing device 110 through the network 120, so that the processing device 110 performs risk analysis and / or processing of the target project. For another example, the project information database 150 can send project-related information to the terminal 130 through the network 120, so that the terminal 130 presents the project-related information to the user. In some embodiments, the project information database 150 can be integrated or deployed in the storage device 140.

[0031] The above description is for illustrative purposes only, and actual application scenarios may vary.

[0032] It should be noted that the application scenario 100 is provided for illustrative purposes only and is not intended to limit the scope of the present invention. For those of ordinary skill in the art, various modifications or variations can be made based on the description of the present invention. However, these variations and modifications will not deviate from the scope of the present invention.

[0033] Figure 2 is a schematic diagram of modules of a risk management system according to some embodiments of the present invention.

[0034] like Figure 2 As shown, the risk management system 200 may include an acquisition module 210, a risk assessment module 220, a patrol determination module 230, and a risk threshold determination module 240. In some embodiments, the acquisition module 210, the risk assessment module 220, the patrol determination module 230, and the risk threshold determination module 240 may be implemented by the processing device 110.

[0035] The acquisition module 210 is configured to acquire a target project parameter value of a target project at a target time.

[0036] The risk assessment module 220 is configured to determine the risk factor of the target project based on the parameter value of the target project.

[0037] In some embodiments, the risk assessment module 220 is also configured to standardize the reference engineering parameter values ​​of the second reference engineering to determine the standardized values ​​corresponding to the reference engineering parameter values; based on the standardized values, determine the indicator weights corresponding to each engineering parameter; based on the target engineering parameter values ​​and indicator weights corresponding to each engineering parameter, determine the risk coefficient.

[0038] In some embodiments, the target project and the second reference project are in the same target project stage, and the indicator weight corresponds to the target project stage.

[0039] The inspection determination module 230 is configured to determine whether the target project needs inspection at the target time based on the risk coefficient and the target risk threshold.

[0040] The risk threshold determination module 240 is configured to obtain an initial risk threshold and historical data of a first reference project; generate an inspection judgment model based on the historical data and the initial risk threshold, where the inspection judgment model is a trained machine learning model; and determine a target risk threshold based on the inspection judgment model and the initial risk threshold.

[0041] In some embodiments, the historical data includes historical inspection records and historical engineering parameter records of a first reference project, and the risk threshold determination module 240 is configured to: determine training labels based on historical engineering parameter records and initial risk thresholds; determine training inputs based on historical inspection records; and obtain an inspection judgment model by training an initial model based on the training labels and training inputs.

[0042] In some embodiments, the risk threshold determination module 240 is configured to: determine the historical engineering parameter values ​​of the first reference project at the historical time based on the historical engineering parameter records; determine the training label based on the historical engineering parameter values ​​and the initial risk threshold; determine the historical inspection parameter values ​​of the first reference project in the historical period before the historical time based on the historical inspection records; and determine the training input based on the historical inspection parameter values.

[0043] In some embodiments, the historical inspection parameter value includes a first historical inspection parameter value corresponding to a first historical period before a historical time and a second historical inspection parameter value corresponding to a second historical period, and the training input is the difference between the first historical inspection parameter value and the second historical inspection parameter value.

[0044] In some embodiments, the risk threshold determination module 240 is further configured to: obtain test inspection data of the target project in the test time period, the test inspection data including inspection judgment results corresponding to multiple test time points in the test time period and inspection records corresponding to at least some of the test time points, wherein the inspection judgment results corresponding to the test time points are determined based on the inspection judgment model; determine whether the initial risk threshold needs to be adjusted based on the test inspection data; in response to determining that the initial risk threshold needs to be adjusted, adjust the initial risk threshold to determine the target risk threshold.

[0045] In some embodiments, the risk threshold determination module 240 is further configured to: determine the test inspection parameter value based on the test inspection data; determine the adjustment range of the initial risk threshold based on the test inspection parameter value; adjust the initial risk threshold based on the adjustment range to determine the target risk threshold.

[0046] In some embodiments, the risk threshold determination module 240 is further configured to: determine the target project stage in response to determining that inspection is required at the target time; obtain the reference value and indicator weight corresponding to each project parameter in the target project stage; and determine the recommended inspection items based on the target project parameter value, reference value and indicator weight of each project parameter.

[0047] In some embodiments, the risk threshold determination module 240 is further configured to: determine the difference between the risk coefficient and the target risk threshold; determine the inspection urgency based on the difference; and initiate a risk warning in response to the inspection urgency being greater than the urgency threshold.

[0048] It should be noted that the above description of the risk management system 200 and its modules is for convenience of description only and does not limit the present invention to the scope of the embodiments. It is understandable that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the various modules, or form a subsystem to connect with other modules without deviating from this principle. For example, the acquisition module 210, the risk assessment module 220, the inspection determination module 230 and the risk threshold determination module 240 may also be different modules in the system, or one module may implement the functions of two or more of the above modules. For example, each module may share a storage module, or each module may have its own storage module. Such variations are all within the scope of protection of the present invention.

[0049] Figure 3 is an exemplary schematic diagram of a risk management method according to some embodiments of the present invention.

[0050] In some embodiments, process 300 may be performed by a risk management system. Figure 3 As shown, process 300 includes the following steps.

[0051] Step S310, obtaining target project parameter values ​​of a target project at a target time.

[0052] The target project refers to a project that needs to be risk managed. For example, the target project may be one or more projects that are in progress (eg, under construction, in progress) and need risk assessment and / or risk control.

[0053] The target engineering parameter value refers to the parameter value of various engineering parameters at the target time. Among them, engineering parameters include various preset engineering indicators or evaluation indicators. For example, engineering parameters include but are not limited to task acceptance rate, employee attendance rate, task order acceptance rate, task order work record rate, employee salary payment rate, safety rectification order ratio, quality rectification order ratio, fatigue worker ratio, etc.

[0054] The target time refers to the time when risk assessment or risk analysis is currently required, for example, the current moment. It should be noted that, considering that various engineering parameters have certain differences (such as timeliness, different statistical methods, etc.), the target time can also be a time period ending at the current moment, for example, a construction cycle ending at the current moment (such as today, this week).

[0055] As an example, the attendance rate at the target time indicates the proportion of employees who have been present on the day as of the current time (e.g., 11:00:00 on January 10, 2025). For another example, for a target project with weekly payroll, the employee payroll rate at the target time indicates the proportion of employees who have received payroll on the most recent payday as of the current time.

[0056] In some embodiments, the risk management system can determine the target project parameter value based on the project-related information corresponding to the target project. For example, the employee attendance rate can be determined based on the employee attendance information of the target project at the target time. For another example, the safety rectification order ratio and / or quality rectification order ratio can be determined based on the quality and safety records of the employee task orders of the target project at the target time. For more information about project-related information, see Figure 1 and its description.

[0057] Step S320, determining the risk factor of the target project based on the target project parameter value.

[0058] The risk factor is used to reflect the degree of risk of the target project at the target time. For example, the risk factor can be a value in the interval (0,1), and the larger the value, the greater the risk of the target project. The risk factor can also be expressed in other forms, such as risk level or risk score. The higher the risk level / risk score, the greater the risk.

[0059] In some embodiments, the risk management system can determine the engineering parameter abnormality based on multiple target engineering parameter values, and determine the risk coefficient of the target engineering based on the engineering parameter abnormality. The engineering parameter abnormality is used to reflect the degree of abnormality of the engineering parameters in the target engineering, which can be determined based on the number or proportion of abnormal engineering parameters. The larger the number or proportion of abnormal engineering parameters, the greater the engineering parameter abnormality corresponding to the target engineering, and the greater the risk coefficient.

[0060] Abnormal engineering parameters refer to engineering parameters whose target engineering parameter values ​​satisfy the abnormal conditions of their corresponding indicators. Different engineering parameters may correspond to different abnormal conditions of indicators, which can be determined according to actual needs. In some embodiments, the risk management system can determine abnormal engineering parameters based on the target engineering parameter values ​​and their corresponding reference values. As an example only, when the employee attendance rate of the target project (such as 70%) is lower than the reference value (such as 95%), the employee attendance rate is an abnormal engineering parameter. For more information about reference values, see Figure 6 and its description.

[0061] In some embodiments, the risk management system may determine the degree of engineering parameter abnormality (eg, 0.2) based on the ratio of the number of abnormal engineering parameters to the total number of engineering parameters (eg, 0.2), thereby determining the risk factor of the target engineering (eg, 0.2).

[0062] In some embodiments, the risk management system can standardize the reference engineering parameter values ​​of the second reference project to determine the standardized values ​​corresponding to the reference engineering parameter values; based on the standardized values, determine the indicator weights corresponding to each engineering parameter; and then determine the risk coefficient based on the target engineering parameter values ​​and indicator weights corresponding to each engineering parameter.

[0063] The second reference project refers to a project of the same project type as the target project. The project type can be various preset types, such as construction projects, road and bridge projects, water conservancy projects, greening projects, etc. When the target project is a construction project (such as an industrial construction project, a structure project, etc.), the second reference project can be a construction project type project.

[0064] The parameter values ​​corresponding to the various project parameters of the second reference project are the reference project parameter values, which can be the parameter values ​​of the various project parameters at historical time and / or target time. For example, the employee attendance rate, safety rectification order ratio, quality rectification order ratio, etc. of the second reference project at a certain historical time.

[0065] The indicator weight reflects the importance of each engineering parameter in the risk assessment of the target project. It can be used to measure the contribution (or influence) of different engineering parameters on the risk coefficient (value). For example, the indicator weight can be a value in the interval [0,1). The larger the indicator weight corresponding to a certain engineering parameter, the greater the influence of the engineering parameter on the risk coefficient of the target project relative to other engineering parameters.

[0066] In some embodiments, the target project and the second reference project are in the same target project stage, and the indicator weight corresponds to the target project stage.

[0067] The target project stage refers to the stage that the target project is currently in, for example, the construction stage (such as the start of construction, mid-construction, final construction, etc.), the acceptance stage (such as item-by-item acceptance stage, comprehensive acceptance stage, delivery stage, etc.), etc., which can be determined based on the actual situation of the target project.

[0068] In some embodiments, the target project is in different engineering stages, and the corresponding indicator weights of different engineering parameters may be different. As an example only, for the construction stage that pays more attention to the construction progress, low employee attendance may slow down the construction progress. Therefore, the impact of this engineering parameter on the risk factor is greater, and the corresponding indicator weight is greater; for the acceptance stage that pays more attention to the quality of the project, the greater the proportion of safety rectification orders and / or quality rectification orders, the target project may have serious quality problems, and therefore the corresponding indicator weight is greater.

[0069] In view of the different requirements of different engineering stages (such as construction progress, quality requirements, etc.), in some embodiments of the present invention, the indicator weights corresponding to the engineering parameters in different engineering stages are set differently, so that the subsequent risk assessment of the target project is more in line with actual needs and more accurate.

[0070] The standardization process can be used to process the parameter values ​​corresponding to different engineering parameters (such as reference engineering parameter values) into a standard range. For example, the standardized value corresponding to an engineering parameter of a second reference engineering can be the proportion of the reference engineering parameter value of the engineering parameter to the sum of the reference engineering parameter values ​​of the engineering parameters of all second reference engineerings (see formula (1) below).

[0071] In some embodiments, the risk management system may preprocess the reference engineering parameter values ​​before standardization. The preprocessing may include, but is not limited to, dimensionalization processing by various algorithms such as forward, reverse, and intervalization to obtain dimensionalized data, so that the data structure, unit, number system or format of the reference engineering parameter values ​​corresponding to each engineering parameter remain unified, so that subsequent unified operations (such as addition operations) can be performed.

[0072] Exemplarily, the dimensioned data is shown in Table 1 below: Table 1. Engineering parameters and dimensionalized data of the second reference project

[0073] It should be noted that the reference project parameter values ​​can be obtained from various data sources (such as acquisition equipment, project information database 150). The multiple projects (such as project 1, project 2, etc.) shown in Table 1 above represent multiple different second reference projects. The various project parameters (such as task acceptance rate, employee attendance rate, etc.) are only exemplary. The actual project parameters can be determined according to actual needs. For example, it can be a part of the multiple project parameters shown in Table 1, or it can also include other project parameters. The project parameters in Table 1 corresponding to different target project stages may also be different.

[0074] In some embodiments, the risk management system may normalize the dimensionalized data (such as the data in Table 1) based on the formula (1) shown below to determine a normalized value.

[0075] Among them, r in formula (1) ij represents the normalized value of the jth engineering parameter in the ith second reference engineering, x ij represents the reference engineering parameter value (such as original data or dimensionalized data) of the j-th engineering parameter in the ith second reference engineering, and m represents the total number of second reference engineering (such as 5).

[0076] In some embodiments, the risk management system may further process the standardized value based on formula (2) shown below to obtain an entropy value corresponding to each engineering parameter.

[0077] The entropy value corresponding to each engineering parameter is used to reflect the uncertainty of the engineering parameter. The smaller the entropy value, the greater the amount of information of the engineering parameter, the smaller the uncertainty, and the greater the corresponding indicator weight. j represents the entropy value corresponding to the jth engineering parameter; r ij Represents the standardized value of the jth engineering parameter in the i-th second reference engineering.

[0078] Exemplarily, the entropy values ​​corresponding to various engineering parameters are shown in Table 2 below: Table 2. Entropy values ​​corresponding to engineering parameters

[0079] In some embodiments, the risk management system may process the entropy values ​​corresponding to the engineering parameters based on formula (3) shown below to obtain the indicator weights corresponding to the engineering parameters.

[0080] Among them, w in formula (3) j represents the index weight corresponding to the jth engineering parameter, e j represents the entropy value corresponding to the jth engineering parameter, and n represents the total number of engineering parameters (such as 8).

[0081] For example, the indicator weights corresponding to each engineering parameter are shown in Table 3 below: Table 3. Index weights corresponding to engineering parameters Engineering parameters <![CDATA[Entropy value e j > <![CDATA[Weight w j > Task acceptance rate 0.95 0.0417 Employee attendance rate 0.90 0.0833 Task order acceptance rate 0.92 0.0667 Task order work rate 0.91 0.075 Employee salary rate 0.85 0.0583 Proportion of safety rectification orders 0.93 0.125 Proportion of quality rectification orders 0.88 0.1 Percentage of employees working under fatigue 0.94 0.05

[0082] In some embodiments, the risk management system may process the target engineering parameter values ​​of the target engineering and the indicator weights corresponding to each engineering parameter based on the following formula (4) to obtain the risk coefficient of the target engineering:

[0083] Where S in formula (4) represents the risk factor of the target project, x i represents the target engineering parameter value corresponding to the i-th engineering parameter, w i Represents the indicator weight corresponding to the i-th engineering parameter.

[0084] Step S330, based on the risk coefficient and the target risk threshold, determine whether the target project needs to be inspected at the target time.

[0085] Inspection refers to the action of inspection personnel conducting engineering inspections on target projects at target times. For example, inspection personnel can conduct on-site inspections of target projects (such as inspections of equipment, materials, worker status, construction progress, etc.) and follow-up visits to construction personnel (such as personnel interviews, questionnaires, etc.) to collect actual engineering-related information of the target project at the target time, and generate inspection results or inspection records. Among them, the inspection results may include a combination of one or more of a variety of information such as text, images, audio and video. More information about inspections and inspection results can be found elsewhere in the present invention (e.g. Figure 4 or Figure 6 ).

[0086] The target risk threshold refers to the critical risk value used to determine whether the target project needs to be inspected. It can be used to reflect the threshold for the target project to be inspected. The representation of the target risk threshold is consistent with the representation of the risk coefficient (for example, represented by a numerical value in the interval (0,1)). The smaller the target risk threshold corresponding to the target project, the easier it is to trigger the inspection personnel to inspect the target project. When the risk coefficient of the target project is greater than or equal to the target risk threshold, it means that the target project needs to be inspected, otherwise, it means that no inspection is required.

[0087] In some embodiments, the target risk threshold may be determined based on historical data. For example, the risk management system may be determined based on historical risk thresholds (such as maximum / minimum values, average values, etc.) corresponding to one or more second reference projects.

[0088] In some embodiments, the target risk threshold is related to the target project stage, and different target project stages have different corresponding target risk thresholds. For example, for the project acceptance stage, the target risk threshold can be set lower to promptly trigger the inspection personnel to inspect the target project.

[0089] In some embodiments, the risk management system determines the target risk threshold based on the inspection judgment model and the initial risk threshold. Figure 4 For more information on determining target risk thresholds, see Figure 5 .

[0090] In some embodiments, the risk management system may also determine the difference between the risk coefficient and the target risk threshold; determine the inspection urgency based on the difference; and initiate a risk warning in response to the inspection urgency being greater than the urgency threshold.

[0091] The difference between the risk factor and the target risk threshold can reflect the severity of the risk. When the risk factor is greater than the target risk threshold, the greater the difference, the more serious the risk of the target project.

[0092] Inspection urgency is used to measure the urgency of inspecting the target project. It can be a numerical value or other forms of values ​​(such as minor level, general level, severe level, etc.). The greater the inspection urgency, the more urgent it is to inspect the target project in a timely manner at the target time (such as quickly going to the project site).

[0093] In some embodiments, the risk management system can set different inspection urgency levels according to different difference amplitudes to initiate different levels of risk warnings to the inspectors, and the risk warnings include but are not limited to information in multiple forms such as text, images, audio and video. The risk management system can send risk warnings to the terminal devices of the inspectors to remind or urge the inspectors to conduct inspections.

[0094] In some embodiments of the present invention, by quantitatively analyzing various engineering parameters of the target project, the potential risks of the target project can be evaluated and analyzed more objectively and accurately, while also making the risk assessment results verifiable, and issuing inspection reminders to inspection personnel in a timely and accurate manner.

[0095] Figure 4 is an exemplary schematic diagram of a method for generating an inspection judgment model according to some embodiments of the present invention.

[0096] In some embodiments, the risk management system may obtain an initial risk threshold and historical data of the first reference project, and generate an inspection judgment model based on the historical data and the initial risk threshold. The inspection judgment model is a trained machine learning model.

[0097] In some embodiments, the first reference project is the target project itself, and the historical data of the first reference project is the historical data of the target project in a specific historical time period before the target moment (for example, historical data of the initial construction period of the target project). In some embodiments, the first reference project refers to a project similar to the target project. For example, a project with the same or similar engineering information as the target project, such as the project type, project scale, and project cycle, can be used as the first reference project. It should be noted that the first reference project may also include a second reference project. For more information about the second reference project, see Figure 3 and its description.

[0098] The initial risk threshold may be an initially set threshold, for example, a value set based on experience or manually. For example, it may be set to 0.5. More information about the initial risk threshold may be found elsewhere in the present invention (e.g. Figure 5 ).

[0099] The inspection judgment model can be used to verify whether the target project needs inspection. In some embodiments, the inspection judgment model can be a logistic regression model, which can be trained based on historical data and an initial risk threshold.

[0100] In some embodiments, the historical data includes historical inspection records and historical engineering parameter records of the first reference project. The risk management system can determine training labels based on the historical engineering parameter records and the initial risk threshold; determine training inputs based on the historical inspection records; and obtain an inspection judgment model by training the initial model based on the training labels and the training inputs.

[0101] The historical data of the first reference project can be determined based on the historical project related information of the first reference project at the historical engineering stage or historical time point. The risk management system can perform information extraction and other processing on the historical project related information to obtain the historical engineering parameter record of the first reference project. Exemplarily, the historical engineering parameter record includes the historical parameter values ​​of various engineering parameters at multiple historical times. For example, the historical task acceptance rate, employee attendance rate, task order acceptance rate, task order work record rate, employee salary payment rate, safety rectification order ratio, quality rectification order ratio, fatigue operation personnel ratio, etc.

[0102] The historical inspection record refers to the inspection record formed after the inspection personnel related to the first reference project inspected the first reference project at one or more historical times (historical project stages, historical moments). The inspection record may include information such as the inspection time and the inspection area. In some embodiments, the inspection record includes a plurality of preset inspection parameters and their corresponding parameter values. For example, the inspection parameters may include but are not limited to the number of inspections, the inspection coverage rate, and the quality ratio of the inspection results. In the following, for the convenience of description, the number of inspections, the inspection coverage rate, and the quality ratio of the inspection results will refer to the inspection parameters, and also refer to their corresponding values ​​or parameter values.

[0103] Among them, the number of inspections is used to reflect the number of inspections performed by the inspected personnel or the number of inspected items in the project. The more inspections there are, the greater the potential risk of the project. For example, when inspectors frequently check inspection items related to employee attendance, it means that the project may be at risk of lagging in construction progress due to employee attendance not meeting the standard.

[0104] The inspection coverage rate is used to reflect the proportion of inspection items that have been inspected in the project. It can be determined based on the ratio of the number of inspection items that have been inspected to the total number of inspection items. The higher the inspection coverage rate, the greater the probability that potential problems of the project will be solved, and the subsequent risk factor may be reduced; the lower the inspection coverage rate, the more likely there are problems such as neglected inspection items, one or more inspection items taking too long, or unreasonable setting of the target risk threshold.

[0105] In some embodiments, the inspection items are related to the engineering stage, and the inspected inspection items are valid inspection items. Different engineering stages correspond to different valid inspection items. For example, the inspection items in the construction stage do not include the inspection items in the acceptance stage or part thereof.

[0106] The inspection result quality ratio is used to reflect the ratio of qualified results to unqualified results after the inspection items are inspected. For example, it may include the number or ratio of qualified results, the number / ratio of unqualified results, etc.

[0107] More information about the number of inspections, inspection coverage, and the quality ratio of inspection results can be found elsewhere in the present invention (e.g. Figure 5 ).

[0108] In some embodiments, Figure 4As shown, the risk management system can determine the historical engineering parameter value 422 of the first reference project at the historical time based on the historical engineering parameter record 421, and determine the training label 420 based on the historical engineering parameter value 422 and the initial risk threshold 423. Among them, the risk management system can determine the historical risk coefficient of the first reference project at the historical time according to the historical engineering parameter value 422 based on the method of step S320, and then determine the training label 420 according to the size relationship between the historical risk coefficient and the initial risk threshold 423. For example, when the historical risk coefficient is greater than or equal to the initial risk threshold 423, the value of the training label 420 is 1, indicating that inspection is required. Otherwise, the value of the training label 420 is 0, indicating that inspection is not required.

[0109] In some embodiments, Figure 4 As shown, the risk management system can also determine the historical inspection parameter value 412 of the first reference project in the historical period before the historical time based on the historical inspection record 411; and determine the training input 410 based on the historical inspection parameter value 412. The risk management system can generate a set of training samples 430 by combining the training input 410 with the training label 420.

[0110] The historical inspection parameter values ​​412 corresponding to a set of training samples 430 may be in the form of a vector matrix with M (e.g., 3) rows and N (e.g., 4) columns. For example, each row corresponds to a first reference project, and each column has M elements. i N j Indicates the jth historical inspection parameter value (such as the number of historical inspections, inspection coverage, and inspection result quality ratio) corresponding to the i-th project. In some embodiments, the training label 420 may also be generated by manual annotation or other methods.

[0111] The risk management system can input multiple groups of training samples 430 with training labels 420 into the initial model 440 respectively, and perform multiple rounds of iterative training on the initial model 440 to obtain a trained inspection judgment model 450. During training, the value of the loss function can be determined based on the difference between the output of the initial model 440 (such as the inspection probability) and the training label 430. The parameters of the initial model 440 can be iteratively updated based on the value of the loss function until the training termination condition is met (such as the loss function converges, a specific number of iterations are performed, etc.). The updated initial model 440 can be used as the trained inspection judgment model 450. In some embodiments, the parameters of the initial model 440 can be expressed based on the expression (6) shown below:

[0112] Among them, in expression (6), P represents the probability of inspection, which can be a value in the interval [0,1], e represents the base of the natural logarithm function, which is a constant (approximately 2.71828). S represents the historical risk factor, F represents the number of inspections, R represents the proportion of unqualified results, and C represents the inspection coverage rate. β 0 , β 1 , β 2 , β 3 and β 4 The parameters of the initial model 440 need to be updated.

[0113] In some embodiments, the historical inspection parameter value includes a first historical inspection parameter value corresponding to a first historical period before a historical time and a second historical inspection parameter value corresponding to a second historical period, and the training input is the difference between the first historical inspection parameter value and the second historical inspection parameter value.

[0114] The first historical time period and the second historical time period may be two consecutive historical time periods with inspection records. For example, the first historical time period is the first historical time period month before the historical time (such as the first month), and the second historical time period is the second historical time period before the historical time (such as the second month).

[0115] The difference between the first historical inspection parameter value and the second historical inspection parameter value refers to the difference between the values ​​corresponding to the same type of inspection parameters in the first historical inspection parameter value and the second historical inspection parameter value, for example, it includes the inspection number difference △F, the unqualified result ratio difference △R and the inspection coverage difference △C.

[0116] In other embodiments, continue to combine Figure 4 , the risk management system can generate a set of training samples 440 based on the inspection number difference △F, the unqualified result ratio difference △R and the inspection coverage difference △C to serve as the training input of the initial model 440. In addition, the training labels 420 corresponding to the training samples 440 can be based on the historical time T 1 The corresponding historical risk coefficient is determined by the magnitude relationship between the initial risk threshold 423 (such as 0 or 1). 1 , the first historical period T` 2 , the second historical period T` 3 Multiple groups of training samples 440 may be generated, so as to perform multiple rounds of iterative training on the initial model 440 to obtain a trained inspection judgment model 450 .

[0117] In some embodiments of the present invention, a training sample is constructed by using the difference between the first historical inspection parameter value and the second historical inspection parameter value, so that the training process of the inspection judgment model can learn the relationship between the inspection parameter changes in multiple consecutive time periods and the inspection requirements, so that the judgment result of whether the inspection is needed is more in line with the actual situation. In addition, since the first historical time period and the second historical time period are earlier than the historical time, the inspection judgment model can learn from the input historical data whether the inspection is needed at the current moment or in the future, thereby realizing a forward-looking inspection demand judgment.

[0118] Figure 5 is an exemplary flow chart of a method for determining a target risk threshold according to some embodiments of the present invention.

[0119] In some embodiments, process 500 is performed by a risk management system. Figure 5 As shown, process 500 includes the following steps.

[0120] Step S510, obtain the test inspection data of the target project in the test time period, the test inspection data includes the inspection judgment results corresponding to multiple test time points in the test time period and the inspection records corresponding to at least some of the test time points, wherein the inspection judgment results corresponding to the test time points are determined based on the inspection judgment model.

[0121] The test period refers to the period after the inspection judgment model is generated (such as after training), which can be a historical period (hereinafter referred to as the historical test period) or a future period (hereinafter referred to as the future test period). For example, the historical test period can be one month before the current moment after the model is generated, and the future test period can be one week after the current moment, etc. For more information about the inspection judgment model, see Figure 4 and its description.

[0122] The test inspection data refers to data used to test or verify the inspection judgment model, which can be the historical test inspection data corresponding to the historical test period of the target project and / or the predicted test inspection data corresponding to the future test period. Among them, the test inspection data includes the inspection records of the test time period.

[0123] The multiple test time points in the test time period can be historical test time points and / or future test time points based on a time series (such as continuous). For example, a test time point can be set for each day in the test time period. The inspection judgment result corresponding to the test time point refers to the judgment result of whether the inspection is required at the test time point determined by the inspection judgment model. Specifically, for each test time point, its corresponding model input can be determined, and after the model input is input into the inspection judgment model, an output value of 0 or 1 can be obtained, where 0 means that no inspection is required and 1 means that inspection is required.

[0124] For the historical test time point, the model input is similar to the determination method of the training input 410. For example, the model input is the difference in parameter values ​​of the inspection parameters of the first historical time period and the second historical time period past the historical test time point. For another example, the model input is the parameter value of the inspection parameter at the test time point. At each historical test time point, when the inspection judgment model determines that the historical test time point needs to be inspected, the risk management system can notify the inspection personnel to go to the target project for inspection, and the result obtained after the inspection is the inspection record corresponding to the test time point.

[0125] For future test time points, the model input can be determined based on the parameter value or parameter value difference of the inspection parameters after the inspection personnel conduct inspections at future test time points (such as every day in the next month), and then determine whether inspections are required at the future test time points based on the model input and the inspection judgment model (i.e., the inspection judgment result) to verify or determine whether the initial risk threshold needs to be adjusted and / or whether the parameters of the inspection judgment model need to be adjusted. See below for details.

[0126] In other embodiments, for future test time points, model inputs and inspection results are generated by means of data prediction. For example, the risk management system can predict the inspection parameters for each day of the next month in turn, and determine the model input based on the inspection parameters, and then predict whether an inspection is required based on the inspection judgment model (i.e., obtain the inspection judgment result). When the inspection judgment model determines that an inspection is required at a future test time point, the risk management system can combine historical data to predict the inspection record corresponding to the future time point.

[0127] Step S520: Determine whether the initial risk threshold needs to be adjusted based on the test inspection data.

[0128] It should be noted that the initial risk threshold here refers to the risk threshold before adjustment (update). It can be an initially preset value or a random value, or an intermediate result or temporary value during the adjustment (update) process. It is understandable that the risk management system can perform multiple rounds of iterative adjustment processing on the initial risk threshold.

[0129] In some embodiments, the risk management system may determine whether the initial risk threshold needs to be adjusted based on the following steps S521 to S523 to determine the target risk threshold.

[0130] Step S521, determining a test inspection parameter value based on the test inspection data.

[0131] The test inspection parameter value refers to the test value corresponding to the inspection parameter in the test inspection record. For example, the number of inspections, inspection coverage, and the proportion of unqualified results. Among them, the test inspection record refers to the inspection record corresponding to each test time point in the test time period.

[0132] In some embodiments, the risk management system can set a preset test parameter value, and determine whether the initial risk threshold needs to be adjusted based on the test inspection parameter value and the preset test parameter value. The preset test parameter value may include a preset number of inspections, a preset inspection coverage rate, and a preset proportion of unqualified results. Exemplarily, the preset number of inspections can be set to 10 inspections within a preset inspection cycle (such as a week), the preset inspection coverage rate is 90% coverage, and the preset proportion of unqualified results is 20%.

[0133] The risk management system can determine whether the initial risk threshold needs to be adjusted based on the comparison results of each test inspection parameter value with the corresponding preset test parameter. For example, if the test inspection parameter value (such as the number of inspections) is greater than the preset test parameter (such as the preset number of inspections), it means that the initial risk threshold is set too low, causing the inspection personnel to inspect too frequently, and it can be adjusted higher. For another example, if the proportion of unqualified results is greater than the preset proportion of unqualified results, it means that the initial risk threshold is set high, and it can be lowered to make it easier to trigger the inspection.

[0134] In some embodiments, the preset test parameter value may be determined based on historical statistical values. For example, the average historical inspection times in a preset inspection cycle over a period of time (such as half a year) in the past may be counted as the preset inspection times.

[0135] When it is determined that the initial risk threshold does not need to be adjusted, it can be directly used as the target risk threshold. When it is determined that the initial risk threshold needs to be adjusted, steps S522 and S523 can be performed to determine the target risk threshold.

[0136] Step S522: determining the adjustment range of the initial risk threshold based on the test inspection parameter value.

[0137] The adjustment range refers to the range by which the initial risk threshold is reduced or increased. It can be a positive or negative value. For example, 0.21 means that the initial risk threshold is increased by 0.21, and -0.3 means that the initial risk threshold is reduced by 0.3, etc.

[0138] In some embodiments, in response to the initial risk threshold requiring adjustment, the risk management system may determine the adjustment range based on the difference between the test inspection parameter value and the preset threshold. The larger the difference, the larger the adjustment range.

[0139] In some embodiments, the risk management system can also adjust the parameters of the inspection judgment model based on the results of whether the inspection is reasonable and the judgment results of whether the inspection is needed. The adjustment of the parameters of the inspection judgment model can be achieved in many ways. For example, more inspection sample data can be constructed to continue training the inspection judgment model. It is also possible to increase the variables of the inspection judgment model and their corresponding parameters so that the inspection judgment model can consider more factors, thereby making its prediction results more refined and accurate.

[0140] In some embodiments, considering that it is in the testing or verification stage, the parameters of the inspection judgment model can be fine-tuned. 0 , β 1 , β 2 , β 3 and β 4 For example, reducing β 3 , so as to reduce the weight of the unqualified result ratio, thereby reducing the contribution of the unqualified result ratio in the inspection parameter value to the judgment result (prediction result) of the inspection judgment model. 2 value to increase the weight of the inspection times, etc.

[0141] Step S523: Adjust the initial risk threshold based on the adjustment range to determine the target risk threshold.

[0142] The risk management system can perform a summation process based on the initial threshold and the adjustment range to obtain a target risk threshold.

[0143] It should be noted that after determining the target risk threshold, the risk management system can use the target risk threshold to determine whether the target project needs to be inspected at the target time based on steps S310 to S330 in subsequent applications.

[0144] In some embodiments, the risk management system may periodically (eg, monthly) detect whether the target risk threshold needs to be adjusted based on steps S520 to S523.

[0145] In some embodiments of the present invention, by adjusting the initial risk threshold using the inspection judgment model during the test period, the initial risk threshold can be adjusted while evaluating the accuracy of the model parameters. In addition, through the inspection judgment model, the current target risk threshold can also be monitored regularly or irregularly based on the inspection records of the inspection personnel, so that the setting of the target risk threshold is consistent with the actual situation of the target project, reducing the deviation of manual judgment, and avoiding the low inspection efficiency caused by improper setting of the risk threshold.

[0146] Figure 6is an exemplary flow chart of a method for determining recommended inspection items according to some embodiments of the present invention.

[0147] In some embodiments, process 600 may be performed by a risk management system. Figure 6 As shown, process 600 includes the following steps.

[0148] Step S610, in response to determining that inspection is required at the target time, determining the target project stage where the target project is located.

[0149] In some embodiments, when the risk factor of the target project at the target time is greater than the target risk threshold, the risk management system determines that an inspection is required at the target time and issues a warning to the inspectors.

[0150] For more information on target projects, risk factors, target risk thresholds, and target project phases, see Figure 3 And its description will not be repeated here.

[0151] Step S620, obtaining the reference value and indicator weight corresponding to each engineering parameter in the target engineering stage.

[0152] The reference value refers to the value that the parameter value corresponding to each engineering parameter (engineering parameter) is expected to reach. The reference values ​​corresponding to different engineering parameters may be different. The reference values ​​corresponding to different engineering parameters at different engineering stages may also be different.

[0153] In some embodiments, the reference value corresponding to each engineering parameter at the target engineering stage can be determined according to a preset reference table corresponding to the engineering stage and the reference value of each engineering parameter. The preset reference table can be set according to the actual requirements of different engineering stages (such as construction progress, construction budget, construction safety, etc.).

[0154] Taking the project parameter of employee attendance rate as an example, when the target project is in the middle of the construction phase, the reference value corresponding to the employee attendance rate can be set higher (such as 95%) due to the needs of the construction progress. When the target project is at the end of the construction phase and enters the acceptance phase, the employee attendance rate has less impact on the target project, and the reference value corresponding to the employee attendance rate can be set lower (such as 80%).

[0155] The risk management system can obtain reference values ​​corresponding to various project parameters from a preset reference table according to the target project stage in which the target project is currently located.

[0156] Step S630, determining recommended inspection items based on the target engineering parameter value, reference value and indicator weight of each engineering parameter.

[0157] Inspection items refer to inspection items related to engineering parameters. For example, for the engineering parameter of employee attendance rate, the inspection items can be to check whether the employees are on duty, the number of employees on duty, etc. For the engineering parameter of safety rectification order ratio, the inspection items can be to check whether the safety rectification order has been rectified, the rectification completion ratio, etc. Recommended inspection items refer to the inspection items that need to be inspected by the inspectors first.

[0158] In some embodiments, the risk management system can determine the first candidate inspection item based on the target engineering parameter values ​​corresponding to each engineering parameter and their corresponding reference values, and determine the second candidate inspection item based on the indicator weights of the engineering parameters corresponding to the target engineering parameter values, and then determine the recommended inspection item based on the first candidate inspection item and / or the second candidate inspection item.

[0159] In some embodiments, the risk management system can respectively determine the difference between each target engineering parameter value and its corresponding reference value, and sort the engineering parameters corresponding to each target engineering parameter value based on the size of the difference (such as in descending order). If the difference is larger (such as larger than a first preset threshold), it means that the corresponding engineering parameter has a higher degree of abnormality (also called an abnormal engineering parameter); if the difference is smaller (such as smaller than a first preset threshold), it means that the corresponding engineering parameter has a lower degree of abnormality or is more normal (also called a normal engineering parameter). The risk management system can determine the first candidate inspection item based on the sorted engineering parameters. For example, the inspection item corresponding to the abnormal engineering parameter or the inspection item corresponding to the top-ranked engineering parameter can be used as the first candidate inspection item.

[0160] In some embodiments, the risk management system can sort (e.g., in descending order) the index weights of the engineering parameters corresponding to the target engineering parameter values, and select the inspection items corresponding to the top-ranked (e.g., top 10) engineering parameters as the second candidate inspection items. It can be understood that the larger the index weight, the greater the impact of the corresponding target engineering parameter value on the risk factor, and the more priority the corresponding inspection items need to be inspected. For more information on index weights, see Figure 3 and its description.

[0161] In some embodiments, the risk management system may determine multiple recommended inspection items based on the first candidate inspection item and the second candidate inspection item. For example, if an inspection item is included in both the first candidate inspection item and the second candidate inspection item, it will be used as a recommended inspection item.

[0162] In some embodiments, the risk management system may send recommended inspection items to the terminal device of the inspector to guide the inspection work of the inspector.

[0163] In some embodiments of the present invention, by analyzing the target project parameter values ​​and the index weights of the corresponding project parameters of the target project in different target project stages, the recommended inspection items can be determined more accurately, providing effective support for the inspection work of the inspection personnel, thereby improving the inspection efficiency.

[0164] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of the present invention. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and corrections to the present invention. Such modifications, improvements and corrections are suggested in the present invention, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present invention.

[0165] At the same time, the present invention uses specific words to describe the embodiments of the present invention. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present invention. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in the present invention does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of the present invention can be appropriately combined.

[0166] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in the present invention, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of the present invention. Although the above disclosure discusses some of the invention embodiments currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of the present invention. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0167] Similarly, it should be noted that in order to simplify the description of the present invention and thus facilitate the understanding of one or more embodiments of the invention, in the above description of the embodiments of the invention, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the subject matter of the invention requires more features than those mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0168] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of the present invention are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0169] Each patent, patent application, patent application disclosure, and other materials, such as articles, books, instructions, publications, documents, etc., cited in this invention are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the present invention, documents that limit the broadest scope of the claims of the present invention (currently or later attached to this invention) are also excluded. It should be noted that if the description, definition, and / or use of terms in the accompanying materials of this invention are inconsistent or conflicting with the contents of this invention, the description, definition, and / or use of terms of this invention shall prevail.

[0170] Finally, it should be understood that the embodiments described in the present invention are only used to illustrate the principles of the embodiments of the present invention. Other variations may also fall within the scope of the present invention. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present invention may be considered consistent with the teachings of the present invention. Accordingly, the embodiments of the present invention are not limited to the embodiments explicitly introduced and described in the present invention.

Claims

1. A method of risk management, characterized in that: include: Get the target project parameter value of the target project at the target time; Determining a risk factor of the target project based on the target project parameter value; Based on the risk coefficient and the target risk threshold, determine whether the target project needs inspection at the target time, wherein the target risk threshold is determined based on the following method: Obtain initial risk thresholds and historical data for the first reference project; Based on the historical data and the initial risk threshold, an inspection judgment model is generated, where the inspection judgment model is a trained machine learning model; The target risk threshold is determined based on the inspection judgment model and the initial risk threshold.

2. The method according to claim 1, characterized in that Determining the risk factor of the target project based on the target project parameter value includes: Standardizing the reference engineering parameter value of the second reference engineering to determine a standardized value corresponding to the reference engineering parameter value; Based on the standardized value, determining the indicator weight corresponding to each engineering parameter; The risk coefficient is determined based on the target engineering parameter value corresponding to each of the engineering parameters and the indicator weight.

3. The method according to claim 1, characterized in that The historical data includes historical inspection records and historical engineering parameter records of the first reference project, and generating an inspection judgment model based on the historical data and the initial risk threshold includes: Determining a training label based on the historical engineering parameter record and the initial risk threshold; Determining training input based on the historical inspection records; Based on the training labels and the training inputs, the inspection judgment model is obtained by training an initial model.

4. The method according to claim 3, characterized in that: Determining the training label includes: Determining historical engineering parameter values ​​of the first reference project at historical times based on the historical engineering parameter records; Determining the training label based on the historical engineering parameter value and the initial risk threshold; Determining the training input includes: Based on the historical inspection records, determining a historical inspection parameter value of the first reference project in a historical period before the historical time; The training input is determined based on the historical inspection parameter values.

5. The method according to claim 4, characterized in that The historical inspection parameter value includes a first historical inspection parameter value corresponding to a first historical period before the historical time and a second historical inspection parameter value corresponding to a second historical period. The training input is a difference between the first historical inspection parameter value and the second historical inspection parameter value.

6. The method according to claim 1, characterized in that The determining the target risk threshold based on the inspection judgment model and the initial risk threshold comprises: Acquire test inspection data of the target project in a test time period, wherein the test inspection data includes inspection judgment results corresponding to multiple test time points in the test time period and inspection records corresponding to at least part of the test time points, wherein the inspection judgment results corresponding to the test time points are determined based on the inspection judgment model; Based on the test inspection data, determining whether the initial risk threshold needs to be adjusted; In response to determining that the initial risk threshold needs to be adjusted, the initial risk threshold is adjusted to determine the target risk threshold.

7. The method according to claim 6, characterized in that Adjusting the initial risk threshold further includes: Determining test and inspection parameter values ​​based on the test and inspection data; Based on the test inspection parameter value, determining an adjustment range of the initial risk threshold; The initial risk threshold is adjusted based on the adjustment amplitude to determine the target risk threshold.

8. The method according to claim 1, characterized in that The method further comprises: In response to determining that the target time requires inspection, Determine the target project stage where the target project is located; Obtain the reference value and indicator weight corresponding to each engineering parameter at the target engineering stage; Based on the target engineering parameter value, the reference value and the indicator weight of each engineering parameter, a recommended inspection item is determined.

9. The method according to claim 1, characterized in that: The method further comprises: Determining a difference between the risk factor and the target risk threshold; Determining the inspection urgency based on the difference magnitude; In response to the inspection urgency being greater than an urgency threshold, a risk warning is initiated.

10. A risk management system, characterized in that: include: An acquisition module is configured to acquire a target project parameter value of a target project at a target time; A risk assessment module, configured to determine a risk factor of the target project based on the target project parameter value; An inspection determination module is configured to determine whether the target project needs inspection at the target time based on the risk coefficient and the target risk threshold; The risk threshold determination module is configured to: Obtain initial risk thresholds and historical data for the first reference project; Based on the historical data and the initial risk threshold, an inspection judgment model is generated, where the inspection judgment model is a trained machine learning model; The target risk threshold is determined based on the inspection judgment model and the initial risk threshold.

Images