Engineering safety risk management method and system

Abstract

The present invention provides a method and system for engineering safety risk management. The method includes: determining a target vector based on collected on-site engineering data; the target vector includes at least on-site engineering characteristic elements and on-site risk characteristic elements; and determining early warning information related to engineering safety risks based on the target vector and a safety risk model, and transmitting it to the user. This method comprehensively considers the causal relationship between various risk factors and accidents, enabling automated comprehensive risk factor analysis. This makes the generation of risk warning results faster and more accurate, facilitating effective subsequent risk factor management by relevant personnel.

Description

Technical Field

[0001] This specification relates to the field of engineering safety management, and in particular to an engineering safety risk management method and system. Background Art

[0002] In recent years, with the rapid growth of engineering technology complexity and project scale, engineering safety risks have become increasingly prominent. From design oversights to construction hazards, from natural disasters to human error, risk factors permeate the entire project lifecycle. Once uncontrolled, these factors can trigger serious accidents with dire consequences. Most existing engineering safety assessment methods are limited to assessing safety risks based on inspection results and conducting risk management based on these findings. However, this approach fails to comprehensively consider the various factors affecting construction safety, nor can it accurately determine the causal relationship between these factors and construction accidents. Furthermore, it is susceptible to the subjective influence of inspectors and assessors.

[0003] Therefore, it is hoped to provide an engineering safety risk management method and system to accurately and comprehensively identify and analyze risk factors that affect engineering safety, thereby conducting effective risk management and control. Summary of the Invention

[0004] In order to solve the problem of how to accurately analyze safety risks and effectively manage them, this specification provides an engineering safety risk management method and system.

[0005] The invention includes a method for engineering safety risk management. The method comprises: determining a target vector based on collected on-site engineering data, the target vector comprising at least on-site engineering characteristic elements and on-site risk characteristic elements; and determining, based on the target vector and a safety risk model, early warning information related to the engineering safety risk and transmitting it to a user.

[0006] In some embodiments, based on a safety responsibility list, risk characteristic data of functional domains under multiple safety risk control types is determined, and the risk characteristic data corresponds to risk hazards; historical accident data of multiple historical accidents is obtained; based on the historical accident data and the risk characteristic data, multiple accident vectors corresponding to the multiple historical accidents are determined, and the accident vectors include at least engineering characteristic elements, risk characteristic elements, and accident characteristic elements; based on the multiple accident vectors, an accident vector library is constructed; based on the engineering characteristic elements, the accident vector library is divided into multiple sub-vector libraries; for each sub-vector library, an association relationship between multiple risk characteristic elements and multiple accident characteristic elements in the sub-vector library is determined; based on the association relationship, the risk level corresponding to the multiple risk characteristic elements and multiple accident characteristic elements in the sub-vector library is determined; based on the multiple risk characteristic elements and multiple accident characteristic elements in the sub-vector library, the association relationship, and the risk level, a safety risk model is constructed.

[0007] In some embodiments, determining the risk characteristic data of functional domains under multiple security risk control types includes determining a risk characteristic tree based on a security responsibility list, and the risk characteristic tree includes at least three layers, wherein the first layer includes multiple security risk control types, the second layer includes the functional domains included in each security risk control type, and the third layer includes the risk characteristic data corresponding to each functional domain, and the risk characteristic data includes at least risk characteristic values ​​and value conditions corresponding to the risk characteristic values.

[0008] In some embodiments, determining the association relationship between multiple risk feature elements and multiple accident feature elements in the sub-vector library includes determining, for each risk feature element and each accident feature element in the sub-vector library, a risk feature value corresponding to the risk feature element and an accident feature value corresponding to the accident feature element; classifying the accident vectors in the sub-vector library based on the risk feature value to determine a first classification result; classifying the accident vectors in the sub-vector library based on the accident feature value to determine a second classification result; and determining the association relationship between the risk feature elements and the accident feature elements based on the first classification result and the second classification result.

[0009] In some embodiments, determining the association relationship between multiple risk feature elements and multiple accident feature elements in the sub-vector library includes screening a preset number of risk feature elements from the multiple risk feature elements in the sub-vector library to form a risk feature element combination, and determining a risk feature value combination corresponding to the risk feature element combination; for each accident feature element in the sub-vector library, determining the accident feature value corresponding to the accident feature element; based on the risk feature value combination, classifying the accident vectors in the sub-vector library to determine a third classification result; based on the accident feature value, classifying the accident vectors in the sub-vector library to determine a fourth classification result; and based on the third classification result and the fourth classification result, determining the association relationship between the risk feature element combination and the accident feature element.

[0010] In some embodiments, constructing a safety risk model includes determining the contribution parameter of each risk characteristic element to the historical accident for each risk characteristic element in the accident vector corresponding to each historical accident; and determining the key risk characteristic element of the historical accident based on each risk characteristic element and the corresponding contribution parameter.

[0011] In some embodiments, determining the contribution parameters of each risk characteristic element to historical accidents includes, for each risk characteristic element, determining a first association parameter between each risk characteristic element and the historical accident; for each risk characteristic element combination in the accident vector corresponding to the historical accident, determining a second association parameter between the risk characteristic element combination and the corresponding historical accident; and based on the first association parameter and the second association parameter, determining the contribution parameter of each risk characteristic element to the historical accident.

[0012] In some embodiments, determining the key risk characteristic elements of historical accidents includes: determining whether the contribution parameter of each risk characteristic element to the historical accident meets a first preset condition; in response to the contribution parameter meeting the first preset condition, determining that the risk characteristic element is a key risk characteristic element.

[0013] In some embodiments, based on the target vector and the safety risk model, determining the warning information related to the engineering safety risk and sending it to the user includes determining the safety risk model based on the on-site engineering characteristic elements in the target vector; querying the safety risk model based on the on-site risk characteristic elements in the target vector, and determining the target accident characteristic elements and the risk level corresponding to the target accident characteristic elements whose association relationship with the on-site risk characteristic elements meets the second preset condition; and determining the warning information based on the target accident characteristic elements, the risk level and the warning rules and sending it to the user.

[0014] In some embodiments, an accident occurs at a construction site, and the target vector site further includes a site accident characteristic element. Based on the target vector and the safety risk model, early warning information related to the engineering safety risk is determined and sent to the user, including: determining the safety risk model based on the site engineering characteristic elements in the target vector; querying the safety risk model based on the target vector to determine the key risk characteristic elements of historical accidents; and determining the early warning information based on the key risk characteristic elements and early warning rules and sending it to the user.

[0015] The invention includes an engineering safety risk management system, which includes a processor configured to execute a method as described in any one of the embodiments of this specification.

[0016] The beneficial effects brought about by the above invention include but are not limited to: (1) it can comprehensively consider the causal relationship between various risk factors and accidents, and determine a safety risk model with the highest possible accuracy, thereby making the generation process of risk warning results faster and more accurate, facilitating the relevant personnel to effectively deal with risk factors in the future and reducing the probability of safety accidents; (2) it can efficiently, comprehensively and accurately identify various risk characteristics and conduct risk level assessment based on the actual situation of the construction site, and quickly identify various risks and causes on site, taking into account the comprehensive consideration of multiple influencing factors, so that the risk identification results are more accurate and reliable; (3) it can determine the safety performance obligations of each unit and each position based on the safety performance obligations of each unit and each position. Comprehensive safety risk control types, functional domains and risk characteristic data are intuitively displayed in a tree structure, which is convenient for subsequent accident attribution based on risk characteristic data and the determination of accurate risk factors; (4) It can comprehensively consider the possibility of all risk characteristic element combinations, and determine the most accurate weight distribution possible based on various parameters such as the scale of each combination, the contribution of the elements in each combination, and the importance of the elements themselves, and determine the contribution parameters of the risk characteristic elements, and then screen out the key risk characteristic elements with larger contribution parameters, so as to facilitate the subsequent determination of appropriate early warning information, timely and effective processing, and improve the overall risk management capabilities. Compared with simple judgment based on correlation relationships, the key characteristic elements obtained in this way are more reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, like numbers represent like structures, wherein:

[0018] Figure 1 is a schematic diagram of an application scenario of an exemplary engineering safety risk management system according to some embodiments of this specification;

[0019] Figure 2 is a module diagram of an exemplary engineering safety risk management system according to some embodiments of this specification;

[0020] Figure 3A is a flowchart of an exemplary engineering safety risk management method according to some embodiments of this specification;

[0021] Figure 3B is a flowchart of an exemplary construction of a security risk model according to some embodiments of this specification;

[0022] Figure 4 is a schematic diagram of an exemplary risk feature tree according to some embodiments of this specification;

[0023] Figure 5is a schematic diagram of an exemplary security risk model according to some embodiments of this specification;

[0024] Figure 6 is a flowchart of an exemplary determination of association relationships according to some embodiments of this specification;

[0025] Figure 7A is a schematic diagram of exemplary accident vectors according to some embodiments of this specification;

[0026] Figure 7B is a schematic diagram of an exemplary classified accident vector according to some embodiments of this specification;

[0027] Figure 8 is a flowchart of an exemplary determination of association relationships according to some embodiments of this specification;

[0028] Figure 9 This is a flowchart of an exemplary method for determining key risk feature elements according to some embodiments of this specification. DETAILED DESCRIPTION

[0029] To more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of this specification. Those skilled in the art can apply this specification to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0030] It should be understood that the terms "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 terms can achieve the same purpose, the terms may be replaced by other expressions.

[0031] As used in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not refer to the singular but also include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0032] Flowcharts are used throughout this specification to illustrate the operations performed by systems according to embodiments of this specification. It should be understood that preceding or following operations do not necessarily need to be performed in exact order. Instead, the steps may be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0033] Figure 1 This is a schematic diagram of an application scenario of an exemplary engineering safety risk management system according to some embodiments of this specification. Figure 1 As shown, the engineering safety risk management system 100 may include a data acquisition device 110 , a server 120 , a user terminal 130 , a storage device 140 and a network 150 .

[0034] The data acquisition device 110 can collect information such as engineering site data within the engineering site. The engineering site refers to a site used for carrying out construction and other related engineering activities. Engineering activities may include one or more of the new construction, reconstruction, maintenance and demolition of buildings, railways, etc. In some embodiments, the data acquisition device 110 may include measuring instruments, sensors, etc. In some embodiments, the engineering site data acquired by the data acquisition device 110 can be transmitted to the server 120, the user terminal 130, the storage device 140, etc. based on the network 150 to realize data exchange and communication. In some embodiments, the user can be in the engineering site and acquire the engineering site data. The above-mentioned description of the engineering site is for illustrative purposes only and is not intended to limit the scope of this description.

[0035] Server 120 is used to manage resources and process data and / or information from at least one component of the engineering safety risk management system 100 or an external data source. In some embodiments, the server can be a single server or a server group. The server group can be centralized or distributed, dedicated, or simultaneously served by other devices or systems. In some embodiments, the server can be regional or remote. In some embodiments, the server can be implemented on a cloud platform or provided virtually.

[0036] In some embodiments, the server 120 may include a processor. The processor may be used to process information and / or data related to the engineering safety risk management system 100. In some embodiments, the processor may process data, information and / or processing results obtained from other devices or system components, and execute program instructions based on these data, information and / or processing results to perform one or more functions described in this specification. For example, the processor may determine a target vector based on data collected from on-site engineering. For another example, the processor may determine early warning information related to engineering safety risks based on the target vector and the safety risk model and send it to the user through the user terminal 130. For more information, please refer to Figure 3A-3B and its related descriptions.

[0037] In some embodiments, the processor may include one or more sub-processing devices (e.g., a single-core processing device or a multi-core multi-core processing device). Exemplarily, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a reduced instruction set computer (RISC), a microprocessor, or any combination thereof.

[0038] User terminal 130 refers to one or more terminal devices or software used by a user. A user refers to a user or manager of the engineering safety risk management method and system, such as a contractor, security personnel, manager, or construction worker.

[0039] In some embodiments, a user can interact with other components of the engineering safety risk management system 100 (such as server 120) via network 150 via user terminal 130. For example, server 120 can send early warning information related to engineering safety risks to user terminal 130 via network 150. For another example, a user can interact with server 120 via user terminal 130 to obtain risk levels corresponding to multiple risk characteristic elements and multiple accident characteristic elements.

[0040] In some embodiments, the user terminal 130 may include a mobile device 130-1, a computer 130-2, a laptop computer 130-3, or any combination thereof. In some embodiments, the user terminal 130 may also include a virtual reality device, such as a VR device or an AR device. In some embodiments, the user terminal 130 may also include a display screen or a speaker installed at the construction site to display warning information to users at the construction site.

[0041] The storage device 140 is used to store data, instructions and / or any other information. The storage device 140 may include one or more storage components, each of which may be an independent device or a part of another device.

[0042] In some embodiments, storage device 140 may include random access memory (RAM), read-only memory (ROM), removable memory, or any combination thereof. In some embodiments, storage device 140 may be connected to network 150 to enable communication with one or more components of engineering safety risk management system 100. In some embodiments, storage device 140 may be part of server 120.

[0043] Network 150 may comprise any suitable network capable of facilitating information and / or data exchange within engineering safety risk management system 100. In some embodiments, one or more components of engineering safety risk management system 100 (e.g., data acquisition device 110, server 120, user terminal 130, storage device 140, etc.) may exchange information and / or data with one or more components of engineering safety risk management system 100 via network 150. In some embodiments, network 150 may comprise any one or more of any wired or wireless network. In some embodiments, network 150 may comprise one or more network access points.

[0044] It should be noted that the engineering safety risk management system 100 is provided for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art will readily appreciate that various modifications and variations can be made based on the description herein. For example, the engineering safety risk management system 100 can be implemented on other devices to achieve similar or different functionality. However, such variations and modifications will not depart from the scope of this specification.

[0045] Figure 2 2 is a block diagram of an exemplary engineering safety risk management system according to some embodiments of this specification. In some embodiments, engineering safety risk management system 200 may include a target vector determination module 210 and an early warning module 220. In some embodiments, each module in engineering safety risk management system 200 may be implemented by server 120 (e.g., a processor in server 120).

[0046] The target vector determination module 210 can be used to determine the target vector based on the field engineering collected data. For more information on how to determine the target vector, please refer to the relevant description of step 310.

[0047] The early warning module 220 can determine early warning information related to the engineering safety risk based on the target vector and the safety risk model and send it to the user. For more information on how to determine the early warning information, please refer to the relevant description of step 320.

[0048] In some embodiments, the engineering safety risk management system 200 may further include a safety risk model construction module 230. The safety risk model construction module 230 may construct a safety risk model. For example, the safety risk model construction module 230 may determine risk characteristic data for functional domains under multiple safety risk control types based on the safety responsibility list, where the risk characteristic data corresponds to potential risks. For another example, the safety risk model construction module 230 may obtain historical accident data for multiple historical accidents and, based on the historical accident data and risk characteristic data, determine multiple accident vectors corresponding to the multiple historical accidents, where the accident vectors include at least engineering characteristic elements, risk characteristic elements, and accident characteristic elements. For another example, the safety risk model construction module 230 may construct an accident vector library based on multiple accident vectors and divide the accident vector library into multiple sub-vector libraries based on the engineering characteristic elements. For each sub-vector library, the safety risk model construction module 230 can determine the association relationships between multiple risk characteristic elements and multiple accident characteristic elements in the sub-vector library; based on the association relationships, determine the corresponding risk levels of the multiple risk characteristic elements and the multiple accident characteristic elements in the sub-vector library; and construct a safety risk model based on the multiple risk characteristic elements and the multiple accident characteristic elements in the sub-vector library, the association relationships, and the risk levels. For more information on how to construct the safety risk model, please refer to the description of steps 331-335.

[0049] In some embodiments, two or more modules in the engineering safety risk management system 200 can be combined into a single module that can implement the functions of the two or more modules. For example, the early warning module 220 and the safety risk model construction module 230 can be combined into a single module that can be used to construct a safety risk model and determine early warning information. In some embodiments, one or more modules in the engineering safety risk management system 200 can be deleted, or one or more modules can be added to the engineering safety risk management system 200. For example, the engineering safety risk management system 200 can exclude the safety risk model construction module 230. The safety risk model can be provided by a supplier and stored in a storage device, and the early warning module 220 can obtain the safety risk model from the storage device.

[0050] Figure 3A is a flowchart of an exemplary engineering safety risk management method according to some embodiments of this specification. Figure 3A As shown, process 300 includes the following steps. In some embodiments, process 300 may be executed by an engineering safety risk management system (eg, server 120 of engineering safety risk management system 100 or various modules of engineering safety risk management system 200).

[0051] Step 310 : Determine the target vector based on the field engineering collected data. In some embodiments, step 310 may be performed by the target vector determination module 210 .

[0052] On-site engineering data refers to project-related data collected at the project site. Examples include on-site engineering characteristic data and on-site risk characteristic data. On-site engineering characteristic data refers to engineering characteristic data acquired at the project site. Engineering characteristic data refers to project-related data, such as project type (e.g., transportation projects, building construction projects, water conservancy projects), project size (e.g., 00,000-100,000 cubic meters, 100,000-200,000 cubic meters), and so on. On-site risk characteristic data refers to risk characteristic data acquired at the project site. For more information on risk characteristic data, please refer to the following description, such as step 331.

[0053] In some embodiments, the processor can obtain on-site engineering data through various methods. For example, the processor can determine risk characteristics and their value conditions using a risk characteristic tree. The processor can then directly obtain real-time data collected from the engineering site related to the corresponding risk characteristics using a data acquisition device (such as data acquisition device 110). Based on this real-time data and the corresponding value conditions, the processor can automatically determine and assign values ​​to the risk characteristics, thereby determining risk characteristic values ​​and thereby determining on-site risk characteristic data. For another example, the processor can determine on-site engineering characteristic data using engineering data stored in a storage device internal to or external to the engineering safety risk management system 200. For more information on risk characteristic data, value conditions, and risk characteristic values, please refer to the relevant description below.

[0054] The target vector refers to a feature vector formed by field engineering data collected. In some embodiments, the target vector may include at least field engineering feature elements and field risk feature elements. Field engineering feature elements refer to elements corresponding to field engineering feature data, and field risk feature elements refer to elements corresponding to field risk feature data. For example, the target vector may be {[Project Category A, Project Scale B], [(Risk Feature 1: Risk Feature Value 1), (Risk Feature 2: Risk Feature Value 2), …, (Risk Feature n: Risk Feature Value n)]}, etc. For more information on risk feature data and risk feature values, please refer to the relevant description below.

[0055] In some embodiments, the processor can determine a target vector based on the accident vectors in the accident vector library and the field engineering data. For example, the processor can determine a matching sub-vector library from the accident vector library based on the field engineering feature data in the field engineering data. Based on the accident vectors in the sub-vector library, the processor can obtain the standard format of the accident vector and populate the risk feature element portion of the standard format of the accident vector with the field risk feature data in the field data to determine the target vector. For more information about the accident vector library, sub-vector library, and accident vectors, please refer to the relevant description below.

[0056] Step 320 : Based on the target vector and the safety risk model, determine warning information related to the engineering safety risk and send it to the user. In some embodiments, step 320 may be performed by the warning module 220 .

[0057] The safety risk model refers to a related model for determining the safety risk of an engineering project. In some embodiments, the safety risk model may include data related to accidents and risk characteristics. Figure 5 is a schematic diagram of an exemplary security risk model according to some embodiments of this specification, such as Figure 5 As shown, the safety risk model may include accident types (such as accident type 1-accident type 3, etc.), accident severity (i.e., accident characteristic elements and their corresponding accident characteristic values. It is understandable that the accident severity corresponds to the accident type, such as accident severity 1.1-accident severity 1.2 corresponding to accident type 1, accident severity 2.1-accident severity 2.3 corresponding to accident type 2, accident severity 3.1-accident severity 3.3 corresponding to accident type 3, etc.), risk level, different preset numbers of risk characteristic elements and their corresponding risk characteristic values, etc. For the construction of the safety risk model, please refer to Figure 3B and its related descriptions.

[0058] Early warning information refers to warnings about project safety risks. These warnings can be presented in various forms. For example, they can be text, sound, or light. Examples of early warning messages include a text message stating, "The foundation anchorage of the tower crane is at risk of failure and may overturn. Please evacuate the site," or a siren signaling a fire.

[0059] In some embodiments, the processor can determine, based on the target vector and the safety risk model, warning information related to the engineering safety risk using a third preset rule and send it to the user. The third preset rule can be preset based on experience or demand. An exemplary third preset rule can be querying the safety risk model based on the target vector. If the risk level of the on-site risk characteristic element and a certain accident characteristic element in the target vector exceeds the risk level threshold, the relevant information is processed to generate warning information and then sent to the user. The risk level threshold can be preset based on experience or demand. The specific processing and generation method can be a machine learning model, a preset algorithm, etc., and the specific sending method can be sending it to the user terminal via the network, etc.

[0060] In some embodiments, the processor can determine the safety risk model based on the on-site engineering feature elements in the target vector. It is understandable that the processor can set different safety risk models based on different sub-vector libraries, that is, based on accident vectors of different engineering categories and different engineering scales, set safety risk models corresponding to the engineering category and engineering scale respectively. In this way, the complex safety risk data can be classified based on engineering category and engineering scale, enhancing the pertinence of risk information queries and effectively improving query efficiency. Exemplarily, the processor can determine the safety risk model corresponding to the engineering category and engineering scale based on the engineering category and engineering scale in the on-site engineering feature elements in the target vector.

[0061] Afterwards, the processor can query the safety risk model based on the on-site risk characteristic elements in the target vector, and determine the target accident characteristic elements whose correlation with the on-site risk characteristic elements meets the second preset condition and the risk level corresponding to the target accident characteristic elements. The target accident characteristic elements refer to accident characteristic elements that have safety risks and need to be paid attention to. The second preset condition refers to the condition for screening the target accident characteristic elements based on the correlation relationship. The second preset condition can be set based on experience or demand. Exemplarily, the second preset condition can be to determine the accident characteristic elements whose accident characteristic values ​​in the safety risk model exceed the accident characteristic value threshold and whose correlation with the on-site risk characteristic elements exceeds the correlation relationship threshold as target accident characteristic elements. The correlation relationship threshold can be set based on experience or demand.

[0062] Finally, the processor can determine and send a warning message to the user based on the target accident characteristic elements, risk level, and warning rules. Warning rules refer to the specific rules for generating warning information. For example, the warning rule may determine that when the risk level exceeds a first risk level threshold (set based on experience or needs), the warning message is determined to be a red-marked target accident characteristic element and sent to the user. For another example, the warning rule may determine that when the risk level exceeds a second risk level threshold (set based on experience or needs) but does not exceed the risk level threshold, the warning message is determined to be a yellow-marked target accident characteristic element and sent to the user.

[0063] In some embodiments of the present specification, by querying the safety risk model based on the on-site risk characteristic elements in the target vector, the target accident characteristic elements and the risk level corresponding to the target accident characteristic elements whose correlation relationship with the on-site risk characteristic elements meets the second preset condition are determined; based on the target accident characteristic elements, risk level and warning rules, warning information is determined and sent to the user, which can accurately and efficiently identify accident characteristics with engineering safety risks that need to be focused on in the implementation engineering data, generate appropriate warning information and send it to the user, and reduce safety hazards.

[0064] In some embodiments, the processor may determine a safety risk model based on the field engineering feature elements in the target vector. For more information on determining the safety risk model, please refer to the above description.

[0065] Afterwards, the processor can query the safety risk model based on the target vector to determine the key risk characteristic elements of the accident. Key risk characteristic elements refer to risk characteristic elements that are more critical to the degree of impact on the accident. For example, risk characteristic elements with the strongest correlation, etc. In some embodiments, the processor can determine the key risk characteristic elements of the accident based on a fourth preset rule. The fourth preset rule can be preset based on experience or demand. An exemplary fourth preset rule can be to query the safety risk model and determine the risk characteristic elements of the on-site risk characteristic elements of the target vector that have the strongest correlation with the accident characteristic elements of the accident as key risk characteristic elements. For more information about key risk characteristic elements, please refer to Figure 9 and its related descriptions.

[0066] Finally, the processor can determine and send warning information to the user based on the key risk characteristic elements and warning rules. An exemplary warning rule can be to process relevant information about the key risk characteristic elements to generate warning information and then send it to the user. Specific processing methods can include machine learning models, preset algorithms, and so on. For more information about warning rules and warning information, please refer to the relevant description above.

[0067] In some embodiments of the present specification, a safety risk model is determined based on the on-site engineering characteristic elements in the target vector; based on the target vector, the safety risk model is queried to determine the key risk characteristic elements of the accident; based on the key risk characteristic elements and warning rules, warning information is determined and sent to the user. This can prioritize the identification of key risk characteristics that have the greatest impact on the accident, facilitate the subsequent determination of appropriate warning information, and handle it in a timely and effective manner, thereby improving the overall risk management and control capabilities.

[0068] In some embodiments, the processor can monitor the target vector and the safety risk model and their data changes in real time, dynamically update the early warning information related to the engineering safety risk and send it to the user.

[0069] In some embodiments of this specification, a target vector is determined based on data collected from on-site engineering projects; based on the target vector and the risk level safety risk model, early warning information related to engineering safety risks is determined and sent to the user; based on the actual situation of the construction site, various risk characteristics can be identified efficiently, comprehensively and accurately, and risk level assessments can be performed. While quickly identifying various risks and causes on site, the comprehensive consideration of multiple influencing factors is taken into account, making the risk identification results more accurate and reliable.

[0070] Figure 3B This is a flowchart of constructing a security risk model according to some embodiments of this specification. Figure 3B As shown, process 330 includes the following steps. In some embodiments, process 330 may be executed by an engineering safety risk management system (eg, server 120 of engineering safety risk management system 100 or various modules of engineering safety risk management system 200).

[0071] Step 331: Based on the security responsibility list, determine the risk characteristic data of functional domains under multiple security risk control types.

[0072] A safety responsibility list is a list of project safety-related responsibilities for multiple positions (e.g., responsible personnel, supervisors, safety personnel, construction personnel, and designers) within multiple engineering units (e.g., construction units, supervisors, construction units, designers, and survey units). This list can include industry standards, corporate policies, job responsibilities, and output deadlines.

[0073] The security risk control type refers to the type of method used to manage security risks. For example, the security risk control type may include self-prevention type, external control type, adjustment type, repair type, and improvement type. In some embodiments, the processor may directly obtain the security risk control type from the security responsibility list.

[0074] Functional domains refer to specific task modules divided based on the security risk control type. For example, when the security risk control type is self-defense, its lower-level functional domains may include awareness elements, behavioral elements, and instruction elements. For another example, when the security risk control type is improvement, its lower-level functional domains may include system optimization and technological innovation.

[0075] Risk signature data refers to information related to risk characteristics (i.e., characteristics that indicate safety risks). Examples include uploaded safety education materials, safety gear wearing information, and regular equipment inspection information. In some embodiments, risk signature data can correspond to potential risks. For example, safety gear wearing information can correspond to potential risks such as being struck by objects, cuts, or burns (i.e., potential risks that may arise from not wearing safety gear).

[0076] In some embodiments, users can mark the security risk control type, functional domain and corresponding risk characteristic data in the security responsibility list based on needs, experience, etc., and the processor can directly obtain the security risk control type, functional domain and risk characteristic data based on the marked content.

[0077] The risk signature tree refers to a tree-like hierarchical structure used to analyze risk signatures. Figure 4 is a schematic diagram of an exemplary risk feature tree according to some embodiments of this specification. Figure 4 As shown, the risk feature tree may include at least three layers, wherein the first layer includes multiple security risk control types (such as self-defense type, external control type, adjustment type, repair type and improvement type), the second layer includes the functional domains contained in each security risk control type (such as awareness elements, behavioral elements, instruction elements, organizational measures, management measures, technical measures, economic measures, hidden danger investigation, supervision and inspection, reporting and complaints, hierarchical management and control, special operations, supervision and rectification, system optimization, scientific and technological innovation, etc.), and the third layer includes the risk feature data corresponding to each functional domain (such as risk feature 1-risk feature n, etc.).

[0078] In some embodiments, the risk characteristic data may at least include a risk characteristic value and a value-taking condition corresponding to the risk characteristic value. The risk characteristic value refers to the numerical value after the risk characteristic is quantified. For example, when the risk characteristic is safety education materials, the risk characteristic value may be 0, 0.5 or 1. For another example, when the risk characteristic is regular equipment inspection, the risk characteristic value may be 0 or 1. The value-taking condition refers to the assignment rule of the risk characteristic value. For example, when the risk characteristic is safety education materials, the value-taking condition may be that if the construction personnel do not upload the safety education materials before the project construction, the risk characteristic value is 0; if the construction personnel upload the safety education materials during the project construction, the risk characteristic value is 0.5; if the construction personnel have uploaded the safety education materials before the project construction, the risk characteristic value is 1. For another example, when the risk characteristic is regular equipment inspection, the value-taking condition may be that if the safety personnel conduct regular equipment inspections, the risk characteristic value is 1; if the safety personnel do not conduct regular equipment inspections, the risk characteristic value is 0. In some embodiments, the value-taking conditions can be preset based on experience or demand, and the processor can automatically collect relevant data of the risk characteristics, and assign values ​​to the risk characteristics based on the value-taking conditions to determine the risk characteristic values ​​corresponding to the risk characteristics.

[0079] In some embodiments, the processor may determine a risk signature tree based on the security responsibility list. For example, a user may annotate the security risk control type, functional domain, and corresponding risk signature data in the security responsibility list based on needs and experience, and the processor may determine the risk signature tree based on the annotated content.

[0080] In some embodiments of this specification, by determining a three-layer risk characteristic tree based on a safety responsibility list, comprehensive safety risk control types, functional domains, and risk characteristic data can be summarized based on the safety performance obligations of each unit and position, and displayed intuitively in a tree structure, which facilitates subsequent accident attribution based on risk characteristic data and the determination of accurate risk factors.

[0081] Step 332: Acquire historical accident data of multiple historical accidents.

[0082] Historical accidents refer to engineering safety incidents that occurred in the past. Examples include scaffolding collapse, construction worker falls from height, and building collapses. Historical accident data refers to data related to historical accidents, such as the time, location, cause, and casualties of the accident.

[0083] In some embodiments, the processor can obtain historical accident data based on various channels such as government platforms, third-party websites, internal corporate information, and web crawling.

[0084] Step 333: Determine multiple accident vectors corresponding to multiple historical accidents based on the historical accident data and the risk characteristic data.

[0085] An accident vector refers to a feature vector associated with a historical accident. In some embodiments, an accident vector may include at least engineering characteristic elements, risk characteristic elements, and accident characteristic elements. Engineering characteristic elements refer to elements corresponding to engineering characteristic data, while risk characteristic elements refer to elements corresponding to risk characteristic data. For more information on engineering characteristic data and risk characteristic data, please refer to the aforementioned description. Accident characteristic elements refer to elements corresponding to accident characteristic data. Accident characteristic data refers to data related to the accident, such as the accident type (e.g., collapse, explosion, impact, electric shock), and the severity of the accident (which can be expressed according to national statutory levels and corresponding numerical values, such as general accident, major accident, serious accident, and particularly serious accident). In some embodiments, accident characteristic data may include at least an accident characteristic value and the corresponding value conditions for the accident characteristic value. An accident characteristic value refers to the quantified numerical value of the accident. For example, if the accident is a foundation pit collapse, the accident characteristic value can be 0 or 1. For another example, if the accident is an explosion, the risk characteristic value can be 1, 2, 3, or 4. Value conditions refer to the rules for assigning accident characteristic values. For example, when the accident is a foundation pit collapse, the value condition can be: if no foundation pit collapse accident occurs, the accident characteristic value is 0; if a foundation pit collapse accident occurs, the accident characteristic value is 1. For another example, when the accident is an explosion, the value condition can be: if the explosion accident is a general accident, the accident characteristic value is 1; if the explosion accident is a major accident, the accident characteristic value is 2; if the explosion accident is a major accident, the accident characteristic value is 3; if the explosion accident is an extremely major accident, the accident characteristic value is 4. In some embodiments, the value condition can be preset based on experience or demand, and the processor can automatically collect relevant data of the accident, and assign values ​​to risk characteristics based on the value condition to determine the accident characteristic value corresponding to the accident. An exemplary accident vector can be {[project category C, project scale D], [(risk characteristic 1: risk characteristic value 1), (risk characteristic 2: risk characteristic value 2),…, (risk characteristic n: risk characteristic value n)], [(accident 1: accident characteristic value 1), (accident 2: accident characteristic value 2),…, (accident m: accident characteristic value m)]}, etc.

[0086] In some embodiments, the processor may further determine accident-related engineering characteristic elements, risk characteristic elements, and accident characteristic elements based on the acquired historical accident data and risk characteristic data, and construct multiple accident vectors corresponding to multiple historical accidents.

[0087] In some embodiments, the processor may standardize the number of elements in the accident vectors to ensure that all accident vectors have a uniform standard format. For example, the processor may take the risk signature element and the accident signature element in the standard format of the accident vector and calculate the union of the risk signature elements and the accident signature elements in all accident vectors, and fill in the missing elements with a predefined character. For example, when the blank parts are filled with 0, when the accident vector 1 is {[project category, project scale], [(risk feature 1: risk feature value 1), (risk feature 2: risk feature value 2)], [(accident 1: accident feature value 1)]}, and the accident vector 2 is {[project category, project scale], [(risk feature 4: risk feature value 4), (risk feature 8: risk feature value 8)], [(accident 2: accident feature value 2), (accident m: accident feature value m)]}, the processor can determine that the standard format of the accident vector is {[project category, project scale], [(risk feature 1: risk feature value 1), (risk feature 2: risk feature value 2), (risk feature 4: risk feature value 4), (risk feature 8: risk feature value 8)], [(accident 1: accident feature value 1)]}. At this time, the accident vector 1 converted to the standard format is {[project category, project scale], [(risk characteristic 1: risk characteristic value 1), (risk characteristic 2: risk characteristic value 2), (risk characteristic 4: 0), (risk characteristic 8: 0)], [(accident 1: accident characteristic value 1), (accident 2: 0), (accident m: 0)]}, and the accident vector 2 converted to the standard format is {[project category, project scale], [(risk characteristic 1: 0), (risk characteristic 2: 0), (risk characteristic 4: risk characteristic value 4), (risk characteristic 8: risk characteristic value 8)], [(accident 1: 0), (accident 2: accident characteristic value 2), (accident m: accident characteristic value m)]}.

[0088] Step 334: construct an accident vector library based on the multiple accident vectors.

[0089] The accident vector library refers to a data set consisting of multiple accident vectors. In some embodiments, the processor can integrate multiple accident vectors to construct the accident vector library.

[0090] Step 335 : Based on the engineering feature elements, the accident vector library is divided into multiple sub-vector libraries.

[0091] A sub-vector library refers to a data set that is further subdivided from the accident vector library. In some embodiments, the processor may divide the accident vector library into multiple sub-vector libraries based on project characteristic elements. For example, the processor may divide accident vectors in the accident vector library with the same project category into one sub-vector library. For another example, the processor may divide accident vectors in the accident vector library with the same project scale into one sub-vector library.

[0092] In some embodiments, for each sub-vector library, the processor may execute steps 3351 to 3353 to construct a security risk model.

[0093] Step 3351: Determine the association relationship between multiple risk feature elements and multiple accident feature elements in the sub-vector library.

[0094] An association is a specific relationship that indicates the degree of connection between things. Examples include strong association, weak association, and no association. Associations can be expressed numerically, such as 100% or 60%. Larger numbers indicate stronger associations.

[0095] In some embodiments, the processor may determine, based on a first preset rule, an association relationship between multiple risk feature elements and multiple accident feature elements in the sub-vector library. The first preset rule may be preset based on experience or demand. An exemplary first preset rule may be determining that a strong association exists between an accident feature element having an accident feature value not equal to zero and a risk feature element having a risk feature value not equal to zero in each accident vector in the sub-vector library.

[0096] In some embodiments, for each risk feature element and each accident feature element in the sub-vector library, the processor can determine the risk feature value and the accident feature value, thereby determining a first classification result and a second classification result; based on the first classification result and the second classification result, determine the association relationship between the risk feature element and the accident feature element. For more information, please refer to Figure 6 and its related descriptions.

[0097] In some embodiments, the processor may filter a preset number of risk feature elements from a plurality of risk feature elements in the sub-vector library to form a risk feature element combination, determine the corresponding risk feature value combination and accident feature value, thereby determining the third classification result and the fourth classification result; based on the third classification result and the fourth classification result, determine the correlation relationship between the risk feature element combination and the accident feature element. For more information, please refer to Figure 8 and its related descriptions.

[0098] Step 3352: Based on the association relationship, determine the risk levels corresponding to the multiple risk feature elements and the multiple accident feature elements in the sub-vector library.

[0099] Risk levels are prioritized based on a comprehensive assessment of the probability and severity of potential risk events within a project. Examples include high risk, medium risk, and low risk. Risk levels can be expressed numerically, such as 80% or 40%. Higher numbers indicate a higher risk level.

[0100] In some embodiments, the processor may determine the risk levels corresponding to multiple risk feature elements and multiple accident feature elements in the subvector library based on a second preset rule. The second preset rule may be based on experience or demand. Exemplary second preset rules include determining the hazard level of an accident based on the accident feature value; and determining the risk levels corresponding to the risk feature elements and accident feature elements based on the hazard level and the correlation relationship. Specific grading rules may be based on experience or demand. For example, the processor may determine an accident with an accident feature value of 0 as no hazard, an accident with an accident feature value of 1 as low hazard, an accident with an accident feature value of 2 as medium hazard, and an accident with accident feature values ​​of 3 or 4 as high hazard. The processor may also determine the risk levels corresponding to risk feature elements with a high hazard level and a strong correlation with accident feature elements as high risk, the risk levels corresponding to risk feature elements with a medium hazard level and a strong correlation with accident feature elements as medium risk, and the risk levels corresponding to risk feature elements with other hazard levels and other correlation relationships with accident feature elements as low risk. Hazard levels may be represented numerically, such as 100% or 60%. Larger numbers indicate higher hazard levels. When the hazard level and the relationship are expressed in numbers, the risk level can be determined based on the geometric calculation results between the aforementioned numbers. For example, you can set risk level = a Hazard level +b Association relationship, a, b are constants, etc.

[0101] Step 3353: construct a safety risk model based on the multiple risk feature elements and multiple accident feature elements, association relationships, and risk levels in the sub-vector library.

[0102] In some embodiments, the processor may obtain each risk characteristic element, each accident characteristic element and all corresponding associations and risk levels respectively based on the aforementioned method and integrate them to construct a safety risk model.

[0103] Based on the safety responsibility list, risk characteristic data for functional domains under multiple safety risk control types is determined, where the risk characteristic data corresponds to risk hazards; historical accident data for multiple historical accidents is obtained; multiple accident vectors corresponding to the multiple historical accidents are determined based on the historical accident data and risk characteristic data, where the accident vectors include at least engineering characteristic elements, risk characteristic elements, and accident characteristic elements; an accident vector library is constructed based on the multiple accident vectors; the accident vector library is divided into multiple sub-vector libraries based on the engineering characteristic elements; for each sub-vector library, the correlation between multiple risk characteristic elements and multiple accident characteristic elements in the sub-vector library is determined; based on the correlation, the risk levels corresponding to the multiple risk characteristic elements and multiple accident characteristic elements in the sub-vector library are determined; and a safety risk model is constructed based on the multiple risk characteristic elements and multiple accident characteristic elements in the sub-vector library, the correlation, and the risk levels. This method comprehensively considers the causal relationship between various risk factors and accidents, and determines a safety risk model with the highest possible accuracy, thereby making the generation process of risk warning results faster and more accurate, facilitating the subsequent effective handling of risk factors by relevant personnel, reducing the probability of safety accidents, achieving automated comprehensive analysis of risk factors, and reducing subjective errors caused by manual determination.

[0104] Figure 6 This is a flowchart of exemplary determination of association relationships according to some embodiments of this specification. Figure 6 As shown, the process 600 includes the following steps: In some embodiments, the processor may execute the process 600 for each risk feature element and each accident feature element in the sub-vector library.

[0105] Step 610 : Determine the risk characteristic value corresponding to the risk characteristic element and the accident characteristic value corresponding to the accident characteristic element. In some embodiments, step 610 may be performed by the safety risk model building module 230 .

[0106] In some embodiments, the processor may determine the value conditions corresponding to the risk characteristic elements based on the risk characteristic tree; and determine the risk characteristic values ​​corresponding to the risk characteristic elements in the sub-vector library based on the historical accident data and the value conditions. For example, the processor may determine, based on the risk characteristic tree, that the value conditions corresponding to the risk characteristic element "Special Plan" in the sub-vector library for the project type "Foundation Pit Excavation" are: if a special plan is in place before construction, the risk characteristic value is 1; if a special plan is supplemented during construction, the risk characteristic value is 0; if a special plan is not in place after construction, the risk characteristic value is 0. Based on the historical accident data of foundation pit collapse and the lack of a special plan, the risk characteristic value corresponding to this risk characteristic element is determined to be 0.

[0107] In some embodiments, the processor may determine the value conditions corresponding to the accident characteristic element based on user input or other means; and determine the accident characteristic value corresponding to the accident characteristic element in the sub-vector library based on historical accident data and the value conditions. For example, the processor may determine based on user input that the value condition corresponding to the accident characteristic element is a general accident (according to national standards, such as accidents resulting in fewer than three deaths, fewer than ten serious injuries, or direct economic losses of less than 10 million yuan) and assign a value of 1. Based on historical accident data, a foundation pit collapse resulting in one death is considered a general accident, and the accident characteristic value corresponding to this accident characteristic element is 1.

[0108] For more information about risk characteristic elements, risk characteristic values, accident characteristic elements, accident characteristic values, risk characteristic trees, value conditions, historical accident data, etc., please refer to Figures 3A-3B and its related descriptions.

[0109] Step 620 : Based on the risk feature value, classify the accident vectors in the sub-vector library to determine a first classification result. In some embodiments, step 620 may be performed by the safety risk model building module 230 .

[0110] The first classification result refers to the accident vector classification result determined based on a risk characteristic element. Figure 7A is a schematic diagram of exemplary accident vectors according to some embodiments of this specification. Figure 7A As shown in Figure 1, the accident vector can be divided into the engineering part (i.e., the engineering characteristic element part, not shown in the figure), the characteristic part (i.e., the risk characteristic element part), and the accident part (i.e., the accident characteristic element part). For more information about the accident vector, please refer to Figure 3A-3B And related descriptions. In some embodiments, the processor can classify all accident vectors based on the number of risk characteristic value types of a certain risk characteristic element and determine a first classification result. For example, the processor can classify all accident vectors into three categories based on the fact that there are three risk characteristic values ​​of risk characteristic element 1 (for example, 0, 0.5, 1), and determine the first classification result 1 corresponding to risk characteristic element 1, that is, the accident vector with a risk characteristic value of 0 for risk characteristic element 1 is classified as category 1, the accident vector with a risk characteristic value of 0.5 for risk characteristic element 1 is classified as category 2, and the accident vector with a risk characteristic value of 1 for risk characteristic element 1 is classified as category 3. Figure 7B is a schematic diagram of an exemplary classification accident vector according to some embodiments of this specification, such as Figure 7BAs shown in the figure, if there are 9 accident vectors G1-G9 in the sub-vector library, for the risk characteristic element "special plan", all accident vectors can be divided into 3 categories. The first category is "construction without plan" (that is, the risk characteristic value is 0) [G1, G2, G3], the second category is "construction first and then supplementary plan" (that is, the risk characteristic value is 0.5) [G4, G5, G6], and the third category is "construction with plan" (that is, the risk characteristic value is 1) [G7, G8, G9].

[0111] Step 630 : Based on the accident feature value, classify the accident vectors in the sub-vector library to determine a second classification result. In some embodiments, step 630 may be performed by the safety risk model building module 230 .

[0112] The second classification result refers to the classification result of the accident vector determined based on an accident characteristic element. In some embodiments, the processor can classify all accident vectors based on the number of accident characteristic value types of a certain accident characteristic element to determine the second classification result. For example, the processor can classify all accident vectors into two categories based on the fact that there are two risk characteristic values ​​of accident characteristic element 1 (for example, 0, 1), and determine the second classification result 1 corresponding to accident characteristic element 1, that is, the accident vector with the accident characteristic value of accident characteristic element 1 being 0 is classified as category 1, and the accident vector with the risk characteristic value of accident characteristic element 1 being 1 is classified as category 2. Figure 7B As shown in the figure, if there are 9 accident vectors G1-G9 in the sub-vector library, all accident vectors can be divided into 3 categories based on the accident characteristic value. The first category is [G1, G2, G3, G4, G5] where "an accident occurred" (i.e., the accident characteristic value is 1), and the second category is [G6, G7, G8, G9] where "no accident occurred" (i.e., the accident characteristic value is 0).

[0113] Step 640 : Determine the association relationship between the risk characteristic element and the accident characteristic element based on the first classification result and the second classification result. In some embodiments, step 640 may be performed by the safety risk model building module 230 .

[0114] In some embodiments, the processor can compare the first classification result and the second classification result. If a certain category in the first classification result is included in a certain category in the second classification result, it means that the risk characteristic element corresponding to the risk characteristic value of this category and the accident characteristic element corresponding to the accident characteristic value of this category are strongly correlated; if a certain category in the first classification result is not completely included in a certain category in the second classification result, it means that the risk characteristic element corresponding to the risk characteristic value of this category and the accident characteristic element corresponding to the accident characteristic value of this category are weakly correlated. Figure 7BAs shown, [G1, G2, G3] of "construction without a plan" are all included in [G1, G2, G3, G4, G5] of "an accident occurred", so it can be determined that the risk characteristic element "construction without a plan" and the accident characteristic element "an accident occurred" are strongly correlated; [G7, G8, G9] of "construction with a plan" are all included in [G6, G7, G8, G9] of "no accident occurred", so it can be determined that the risk characteristic element "construction with a plan" and the accident characteristic element "no accident occurred" are strongly correlated; [G4, G5, G6] of "construction first and then supplementary plan" are partially included in [G1, G2, G3, G4, G5] of "an accident occurred", and partially included in [G6, G7, G8, G9] of "no accident occurred", so the risk characteristic element "construction first and then supplementary plan" and the accident characteristic element "an accident occurred" and the accident characteristic element "no accident occurred" are all weakly correlated.

[0115] In some embodiments, when the association relationship is represented by a number, the processor may determine the ratio of the number of a certain category in the first classification result that is included in a certain category in the second classification result to the total number as the association relationship between the risk characteristic element corresponding to the risk characteristic value of that category and the accident characteristic element corresponding to the accident characteristic value of that category. For example, [G1, G2, G3] of "no plan construction" are all in [G1, G2, G3, G4, G5] of "accident occurred", so it can be determined that the association relationship between the risk characteristic element "no plan construction" and the accident characteristic element "accident occurred" is 100%, and only two vectors of [G4, G5, G6] of "construction first and supplementary plan later" are included in [G1, G2, G3, G4, G5] of "accident occurred", so it can be determined that the association relationship between the risk characteristic element "construction first and supplementary plan later" and the accident characteristic element "accident occurred" is 66.67%.

[0116] In some embodiments, all risk feature elements and all accident feature elements in the sub-vector library are traversed according to the above method to obtain the association relationship between all risk feature elements and all accident feature elements.

[0117] In some embodiments of the present specification, for each risk characteristic element and each accident characteristic element in the sub-vector library, the risk characteristic value corresponding to the risk characteristic element and the accident characteristic value corresponding to the accident characteristic element are determined; based on the risk characteristic value, the accident vectors in the sub-vector library are classified to determine a first classification result; based on the accident characteristic value, the accident vectors in the sub-vector library are classified to determine a second classification result; based on the first classification result and the second classification result, the correlation relationship between the risk characteristic element and the accident characteristic element is determined. This can clearly and intuitively derive the correlation relationship between different risk characteristic elements and different accident characteristic elements, can conveniently and quickly find strong correlation factors for risk accidents, and facilitate the processing of the corresponding strong correlation factors.

[0118] Figure 8 This is a flowchart of exemplary determination of association relationships according to some embodiments of this specification. Figure 8 As shown, the process 800 includes the following steps. In some embodiments, the process 800 can be executed by a processor.

[0119] Step 810 : Filter a preset number of risk feature elements from the plurality of risk feature elements in the sub-vector library to form a risk feature element combination, and determine a risk feature value combination corresponding to the risk feature element combination. In some embodiments, step 810 may be performed by the security risk model construction module 230 .

[0120] In some embodiments, the processor may screen a preset number of risk feature elements (which may be set based on experience or demand) from multiple risk feature elements in the sub-vector library, combine these risk feature elements together to determine a risk feature element combination, then determine the risk feature values ​​corresponding to these risk feature elements respectively, and then combine these risk feature values ​​together to determine a risk feature value combination corresponding to the aforementioned risk feature element combination. For a specific method of determining the risk feature values ​​corresponding to the risk feature elements, please refer to Figure 3A-3B 、 Figure 6 and its related description.

[0121] Step 820, for each accident feature element in the sub-vector library, determine the accident feature value corresponding to the accident feature element. In some embodiments, step 820 can be performed by the safety risk model building module 230. For the specific method of determining the accident feature value corresponding to the accident feature element, please refer to Figure 3A-3B 、 Figure 6 and its related description.

[0122] Step 830 : Based on the risk feature value combination, the accident vectors in the sub-vector library are classified to determine a third classification result. In some embodiments, step 820 may be performed by the safety risk model building module 230 .

[0123] The third classification result refers to the accident vector classification result determined based on multiple risk characteristic elements. Figure 7ASchematic diagram of exemplary accident vectors according to some embodiments of this specification. In some embodiments, the processor can classify all accident vectors based on the number of risk feature value combinations and determine a third classification result. For example, when the preset number is 2, the processor can determine that the number of risk feature value combinations corresponding to the risk feature element combination consisting of risk feature element 2 and risk feature element 3 is 6 (i.e., (0, 0), (0, 0.5), (0, 1), (1, 0), (1, 0.5), (1, 1)) based on the fact that there are 2 risk feature values ​​for risk feature element 2 and 3, and 3 risk feature values ​​for risk feature element 3. Based on these 6 risk feature value combinations, all accident vectors are classified into 6 categories and the third classification result corresponding to the risk feature element combination is determined.

[0124] Step 840 : Based on the accident feature value, classify the accident vectors in the sub-vector library to determine a fourth classification result. In some embodiments, step 820 may be performed by the safety risk model building module 230 .

[0125] The fourth classification result refers to the accident vector classification result determined based on an accident characteristic element. The method for determining the fourth classification result can refer to the method for determining the second classification result described above.

[0126] Step 850 : Based on the third classification result and the fourth classification result, determine the association relationship between the risk characteristic element combination and the accident characteristic element. In some embodiments, step 850 may be performed by the safety risk model building module 230 .

[0127] In some embodiments, the processor can compare the third classification results and the fourth classification results. If a certain category in the third classification results is included in a certain category in the fourth classification results, it means that the risk characteristic element combination corresponding to this type of risk characteristic value combination and the accident characteristic element corresponding to this type of accident characteristic value are strongly correlated; if a certain category in the third classification results is not completely included in a certain category in the fourth classification results, it means that the risk characteristic element combination corresponding to this type of risk characteristic value combination and the accident characteristic element corresponding to this type of accident characteristic value are weakly correlated.

[0128] In some embodiments, when the association relationship is expressed digitally, the processor can determine the ratio of the number of a certain category in the third classification result included in a certain category in the third classification result to the total number as the association relationship between the risk feature element combination corresponding to this type of risk feature value combination and the accident feature element corresponding to this type of accident feature value.

[0129] In some embodiments, the above method is used to traverse all preset numbers of risk feature elements in the sub-vector library, all risk feature element combinations consisting of the preset number of risk feature elements, and all accident feature elements to obtain the association relationship between all risk feature elements and all accident feature elements.

[0130] In some embodiments of the present specification, a preset number of risk characteristic elements are screened from multiple risk characteristic elements in a sub-vector library to form a risk characteristic element combination, and a risk characteristic value combination corresponding to the risk characteristic element combination is determined; for each accident characteristic element in the sub-vector library, the accident characteristic value corresponding to the accident characteristic element is determined; based on the risk characteristic value combination, the accident vectors in the sub-vector library are classified to determine a third classification result; based on the accident characteristic value, the accident vectors in the sub-vector library are classified to determine a fourth classification result; based on the third classification result and the fourth classification result, the correlation relationship between the risk characteristic element combination and the accident characteristic element is determined. On the basis of determining the correlation relationship between a single risk characteristic element and the accident characteristic element, the correlation relationship between the combination of multiple risk characteristic elements and the accident characteristic element can be further determined. It can be understood that the occurrence of an accident is often not affected by only a single factor. Only by comprehensively considering multiple factors and the common influence between them can the determination result of the important correlation factors corresponding to the accident be more accurate and more in line with the actual situation.

[0131] Figure 9 This is a flowchart of exemplary determination of key risk feature elements according to some embodiments of this specification. Figure 9 As shown, the process 900 includes the following steps. In some embodiments, the process 900 can be executed by a processor.

[0132] Step 910 : For each risk characteristic element in the accident vector corresponding to each historical accident, determine the contribution parameter of each risk characteristic element to the historical accident. In some embodiments, step 910 may be performed by the safety risk model building module 230 .

[0133] The contribution parameter is a parameter that indicates the degree of impact of a risk characteristic element on an accident. The larger the contribution parameter, the greater the impact of the risk characteristic element on the accident.

[0134] In some embodiments, the processor may determine a contribution parameter of each risk characteristic element to the historical accident based on a fifth preset rule. The fifth preset rule may be preset based on experience or demand. An exemplary fifth preset rule may be directly determining the association between the risk characteristic element and the corresponding accident characteristic element as the contribution parameter of the risk characteristic element to the historical accident.

[0135] In some embodiments, for each risk signature element, the processor may determine a first association parameter of each risk signature element with the accident.

[0136] The first correlation parameter refers to a parameter that represents the correlation between a certain risk characteristic element and an accident characteristic element corresponding to a historical accident. In some embodiments, the processor can determine the first correlation parameter by determining the correlation in the aforementioned step 640. Figure 6 and its related descriptions.

[0137] Afterwards, the processor may determine, for each risk feature element combination in the accident vector corresponding to the historical accident, a second association parameter between the risk feature element combination and the corresponding historical accident.

[0138] The second association parameter refers to a parameter representing the association relationship between a certain risk characteristic element and an accident characteristic element corresponding to a historical accident when the risk characteristic element is included in a certain risk characteristic element combination. In some embodiments, the processor may determine the first association relationship of the risk characteristic element combination when the risk characteristic element combination includes the risk characteristic element by determining the association relationship in the aforementioned step 850, and then determine the second association relationship of the risk characteristic element or combination when the risk characteristic element combination does not include the risk characteristic element by determining the association relationship in the aforementioned step 640 or the aforementioned step 850 (it is understood that if the combination includes only two risk characteristic elements, then only one risk characteristic element remains after removing the risk characteristic element, and the method in step 640 can be used to calculate the association relationship; if the combination includes three or more risk characteristic elements, then after removing the risk characteristic element, the risk characteristic element combination remains, and the method in step 850 must still be used to calculate the association relationship), and the difference between the first association relationship and the second association relationship is used as the second association parameter. For example, if risk characteristic element combination 1 (risk characteristic element 1, risk characteristic element 2) has a 70% correlation with historical accident 1, and risk characteristic element 2 has a 20% correlation with historical accident 1, then when risk characteristic element 1 is included in risk characteristic element combination 1, the second correlation parameter for historical accident 1 is 50%. For another example, if risk characteristic element combination 2 (risk characteristic element 1, risk characteristic element 5, risk characteristic element 8) has a 60% correlation with historical accident 1, and risk characteristic element combination 3 (risk characteristic element 5, risk characteristic element 8) has a 40% correlation with historical accident 1, then when risk characteristic element 1 is included in risk characteristic element combination 2, the second correlation parameter for historical accident 1 is 20%. It will be appreciated that the same risk characteristic element may appear in multiple risk characteristic element combinations, so the calculated second correlation parameter may also be multiple.

[0139] Finally, the processor can determine the contribution parameter of each risk characteristic element to the historical accident based on the first associated parameter and the second associated parameter. In some embodiments, the processor can determine the target weight; based on the first associated parameter, the second associated parameter and the target weight, determine the contribution parameter of each risk characteristic element to the historical accident. For example, the processor can add the sum of the product of each second associated parameter of the risk characteristic element and its corresponding target weight to the first associated parameter to determine the contribution parameter of the risk characteristic element to the historical accident, that is, contribution parameter = first associated parameter + second associated parameter 1 Target weight of risk characteristic element combination 1 + ... + second association parameter n The target weight n of the risk characteristic element combination n. The target weight refers to the contribution weight of the risk characteristic element relative to the risk characteristic element combination. In some embodiments, the target weight may be related to at least one of the number of elements in the accident vector, the number of elements in the risk characteristic element combination, the determination time of the risk characteristic element, and the weight of the risk characteristic element itself. For example, the target weight may be related to the number of elements in the accident vector and the number of elements in the risk characteristic element combination. Exemplarily, the target weight may be calculated based on a preset formula. For example, the preset formula may be , where the number of elements refers to the number of risk characteristic elements. It is understandable that the target weight is intended to represent the value increase of the entire risk characteristic element combination after the addition of a certain risk characteristic element. Therefore, when setting the preset formula for the target weight, it is necessary to first consider the possible combinations of other risk characteristic elements in the risk characteristic element combination except for the risk characteristic element, and then add the risk characteristic element. For example, the numerator in the aforementioned preset formula is (the number of permutations and combinations of other risk characteristic elements in the risk characteristic element combination except for the risk characteristic element) (The number of permutations and combinations that include this risk characteristic element as the last element is omitted from the formula because there is only one) (The number of permutations and combinations of elements in the accident vector that are not in the risk feature element combination when the order is considered), the denominator is (the number of permutations and combinations of all elements in the accident vector when the order is considered). For another example, the target weight may be related to the determination time of the risk feature element. The closer the determination time of the risk feature element is to the current time, the larger the target weight is set. For another example, the target weight may be related to the risk feature element's own weight. The greater the risk feature element's own weight, the greater the target weight is set. The risk feature element's own weight may be set based on experience or demand. For example, the risk feature element's own weight may be related to the correlation between the risk feature element and the accident feature element. The greater the correlation, the higher the risk feature element's own weight. The processor may traverse each risk feature element based on the above method to determine the contribution parameter of each risk feature element to the historical accident.

[0140] It can be understood that there are many possible permutations and combinations of risk characteristic element combinations, and the gain effects generated by risk characteristic elements in each combination are also different. If the second associated parameters are simply and roughly evenly divided and set, the contribution parameters obtained will undoubtedly have large deviations. In some embodiments of the present specification, the contribution parameters are determined based on the first associated parameters, the second associated parameters and the target weight, and the target weight is set to be related to at least one of the number of elements in the accident vector, the number of elements in the risk characteristic element combination, the determination time of the risk characteristic element and the weight of the risk characteristic element itself. The possibility of all risk characteristic element combinations can be comprehensively considered, and the weight distribution can be determined as accurately as possible based on various parameters such as the scale of each combination, the contribution of the element in each combination, and the importance of the element itself, thereby helping to improve the accuracy and reliability of the contribution parameters.

[0141] In some embodiments of the present specification, for each risk characteristic element, a first association parameter between each risk characteristic element and a historical accident is determined; for each risk characteristic element combination in an accident vector corresponding to a historical accident, a second association parameter between the risk characteristic element combination and the corresponding historical accident is determined; based on the first association parameter and the second association parameter, a contribution parameter of each risk characteristic element to the historical accident is determined. On the basis of determining the association relationship between the risk characteristic element and the accident, its impact on the accident when in the combination can be further determined, and the total impact of the element can be comprehensively determined, so that the calculation of the risk characteristic impact parameter is more reasonable.

[0142] Step 920, based on each risk characteristic element and the corresponding contribution parameter, determine the key risk characteristic element of the historical accident. In some embodiments, step 920 can be performed by the safety risk model building module 230. For more information about key risk characteristic elements, please refer to Figure 3A-Figure 8 and its related descriptions.

[0143] In some embodiments, the processor may compare the contribution parameters of each risk characteristic element and determine the risk characteristic element with the maximum contribution parameter as the key risk characteristic element of the historical accident.

[0144] In some embodiments, the processor may determine whether a contribution parameter of each risk characteristic element to historical accidents satisfies a first preset condition.

[0145] The first preset condition refers to a condition for screening key risk characteristic elements based on the contribution parameter. The first preset condition can be set based on experience or demand. For example, the first preset condition can be screening risk characteristic elements whose contribution parameters exceed a contribution parameter threshold. The contribution parameter threshold can be set based on experience or demand. The first preset condition can also be sorting risk characteristic elements from largest to smallest based on the contribution parameter, screening risk characteristic elements that rank in the top x% of the contribution parameter, etc.

[0146] In response to the contribution parameter satisfying the first preset condition, the processor may determine that the risk signature element is a key risk signature element.

[0147] In some embodiments of the present specification, by determining whether the contribution parameter of each risk characteristic element to historical accidents meets the first preset condition; in response to the contribution parameter meeting the first preset condition, determining that the risk characteristic element is a key risk characteristic element, the key risk characteristic elements with larger contribution parameters can be accurately screened out, facilitating timely rectification and avoiding safety risk accidents.

[0148] In some embodiments of the present specification, for each risk characteristic element in the accident vector corresponding to each historical accident, the contribution parameter of each risk characteristic element to the historical accident is determined; based on each risk characteristic element and the corresponding contribution parameter, the key risk characteristic elements of the historical accident are determined. The risk characteristic elements that have a greater impact on safety accidents can be determined and used as key characteristic elements. Compared with simply judging by the correlation relationship, the key characteristic elements obtained in this way are more reasonable.

[0149] While the basic concepts have been described above, it will be apparent to those skilled in the art that the detailed disclosure is merely illustrative and does not limit this specification. Although not explicitly stated herein, various modifications, improvements, and revisions to this specification may be made by those skilled in the art. Such modifications, improvements, and revisions are suggested in this specification and remain within the spirit and scope of the exemplary embodiments of this specification.

[0150] This specification also uses specific terms to describe the embodiments of this specification. For example, "one embodiment," "an embodiment," and / or "some embodiments" refer to a feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "one embodiment," "an embodiment," or "an alternative embodiment" two or more times in different locations in this specification do not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of this specification may be appropriately combined.

[0151] In addition, unless expressly stated in the claims, the order of the processing elements and sequences, the use of alphanumeric characters, or the use of other names described in this specification are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some of the invention embodiments currently considered useful through various examples, it should be understood that such details are for illustrative purposes only, and the appended 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 spirit and scope of the embodiments of this specification. 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.

[0152] Similarly, it should be noted that, in order to simplify the presentation of this specification and thus facilitate understanding of one or more embodiments of the invention, the foregoing descriptions of the embodiments of this specification sometimes combine multiple features into a single embodiment, figure, or description thereof. However, this disclosure method does not imply that the subject matter of this specification requires more features than those recited in the claims. In fact, an embodiment may have fewer features than all of the features of a single disclosed embodiment.

[0153] 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 stated, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the description and claims are approximate values, which may vary according to the required features of the 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 their range in some embodiments of this specification are approximate values, in specific embodiments, the settings of such numerical values ​​are as accurate as possible within the feasible range.

[0154] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, and documents, cited in this specification is hereby incorporated by reference in its entirety. This excludes any application history documents that are inconsistent with or conflicting with the content of this specification, as well as any documents (currently or subsequently appended to this specification) that limit the broadest scope of the claims of this specification. It should be noted that if the descriptions, definitions, and / or terminology used in the accompanying materials are inconsistent or conflicting with the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0155] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. A method for engineering safety risk management, characterized in that: include: Determining a target vector based on the field engineering collected data, wherein the target vector includes at least a field engineering characteristic element and a field risk characteristic element; as well as, Based on the target vector and the safety risk model, early warning information related to the engineering safety risk is determined and sent to the user, wherein the determining of the early warning information related to the engineering safety risk includes determining the early warning information based on key risk characteristic elements and early warning rules, and constructing the safety risk model includes: Based on the security responsibility list, determine risk characteristic data of functional domains under multiple security risk control types, where the risk characteristic data corresponds to potential risks; Obtain historical accident data for multiple historical accidents; Determining, based on the historical accident data and the risk characteristic data, a plurality of accident vectors corresponding to the plurality of historical accidents, the accident vectors including at least an engineering characteristic element, the risk characteristic element, and the accident characteristic element; For each risk characteristic element in the accident vector corresponding to each historical accident, determining a contribution parameter of each risk characteristic element to the historical accident includes: For each of the risk characteristic elements, determining a first association parameter between each of the risk characteristic elements and the historical accident, wherein the first association parameter is a parameter representing an association relationship between each of the risk characteristic elements and the accident characteristic element of the accident vector corresponding to the historical accident; For each of the risk characteristic elements, determining a second association parameter between each of the risk characteristic elements and the corresponding historical accident when the risk characteristic element is in the risk characteristic element combination, wherein the second association parameter is a parameter representing an association relationship between each of the risk characteristic elements and the accident characteristic element of the accident vector corresponding to the historical accident when the risk characteristic element is in the risk characteristic element combination; Determining the contribution parameter of each risk characteristic element to the historical accident based on the first correlation parameter and the second correlation parameter; Determining the key risk characteristic element of the historical accident based on each risk characteristic element and the corresponding contribution parameter; constructing an accident vector library based on the multiple accident vectors; Based on the engineering characteristic elements, the accident vector library is divided into a plurality of sub-vector libraries; For each of the sub-vector libraries, determining association relationships between a plurality of the risk characteristic elements and a plurality of the accident characteristic elements in the sub-vector library; Based on the association relationship, determining the risk levels corresponding to the multiple risk characteristic elements and the multiple accident characteristic elements in the sub-vector library; and The safety risk model is constructed based on the multiple risk characteristic elements and the multiple accident characteristic elements in the sub-vector library, the association relationship and the risk level.

2. The method according to claim 1, characterized in that Determining risk characteristic data of functional domains under multiple security risk control types includes: Based on the security responsibility list, a risk characteristic tree is determined, and the risk characteristic tree includes at least three layers, wherein the first layer includes the multiple security risk control types, the second layer includes the functional domains contained in each of the security risk control types, and the third layer includes the risk characteristic data corresponding to each of the functional domains, and the risk characteristic data at least includes risk characteristic values ​​and value conditions corresponding to the risk characteristic values.

3. The method according to claim 1, characterized in that Determining the association relationship between the plurality of risk characteristic elements and the plurality of accident characteristic elements in the sub-vector library includes: For each risk feature element and each accident feature element in the sub-vector library, Determining a risk characteristic value corresponding to the risk characteristic element and an accident characteristic value corresponding to the accident characteristic element; classifying the accident vectors in the sub-vector library based on the risk characteristic value to determine a first classification result; Based on the accident feature value, classify the accident vectors in the sub-vector library to determine a second classification result; and Based on the first classification result and the second classification result, an association relationship between the risk characteristic element and the accident characteristic element is determined.

4. The method according to claim 1, wherein Determining the association relationship between the plurality of risk characteristic elements and the plurality of accident characteristic elements in the sub-vector library includes: screening a preset number of risk characteristic elements from the plurality of risk characteristic elements in the sub-vector library to form a risk characteristic element combination, and determining a risk characteristic value combination corresponding to the risk characteristic element combination; For each of the accident characteristic elements in the sub-vector library, determining an accident characteristic value corresponding to the accident characteristic element; classifying the accident vectors in the sub-vector library based on the risk feature value combination to determine a third classification result; Based on the accident feature value, classify the accident vectors in the sub-vector library to determine a fourth classification result; and Based on the third classification result and the fourth classification result, an association relationship between the risk characteristic element combination and the accident characteristic element is determined.

5. The method according to claim 1, characterized in that The step of determining, for each risk characteristic element, a second association parameter between each risk characteristic element and the corresponding historical accident when the risk characteristic element is in the risk characteristic element combination, includes: When it is determined that the risk characteristic element combination includes the risk characteristic element, a first association relationship between the risk characteristic element combination and the accident characteristic element of the accident vector corresponding to the historical accident; When it is determined that the risk characteristic element combination does not include the risk characteristic element, a second association relationship between the remaining risk characteristic elements or the remaining risk characteristic element combination and the accident characteristic element of the accident vector corresponding to the historical accident; A difference between the first association relationship and the second association relationship is determined as the second association parameter.

6. The method according to claim 1, characterized in that The determining, based on the first correlation parameter and the second correlation parameter, the contribution parameter of each risk characteristic element to the historical accident includes: determining a target weight, where the target weight represents a contribution weight of the risk characteristic element to a risk characteristic element combination including the risk characteristic element, the target weight being related to at least one of the number of elements in the accident vector, the number of elements in the risk characteristic element combination, a determination time of the risk characteristic element, and a weight of the risk characteristic element itself; Based on the first association parameter, the second association parameter and the target weight, the contribution parameter of each risk characteristic element to the historical accident is determined. The determination of the contribution parameter includes adding the sum of the products of each second association parameter of the risk characteristic element and its corresponding target weight to the first association parameter, and determining the contribution parameter of the risk characteristic element to the historical accident.

7. The method according to claim 1, characterized in that The target weight is determined based on a preset formula, which includes: Target Weight = 。 8. The method according to claim 1, characterized in that The key risk characteristic elements for determining the historical accident include: determining whether the contribution parameter of each risk characteristic element to the historical accident satisfies a first preset condition; and In response to the contribution parameter satisfying the first preset condition, the risk characteristic element is determined to be the key risk characteristic element.

9. The method according to claim 1, characterized in that The determining, based on the target vector and the safety risk model, early warning information related to the engineering safety risk and sending the information to the user includes: Determining the safety risk model based on the on-site engineering characteristic elements in the target vector; Based on the on-site risk characteristic element in the target vector, query the safety risk model to determine a target accident characteristic element whose association with the on-site risk characteristic element satisfies a second preset condition and a risk level corresponding to the target accident characteristic element; and Based on the target accident characteristic elements, the risk level and the warning rules, the warning information is determined and sent to the user.

10. The method according to claim 1, characterized in that An accident occurs at the engineering site, the target vector site further includes a site accident feature element, and determining, based on the target vector and the safety risk model, early warning information related to the engineering safety risk and sending it to the user comprises: Determining the safety risk model based on the on-site engineering characteristic elements in the target vector; Based on the target vector, query the safety risk model to determine the key risk characteristic elements of the historical accident; and Based on the key risk characteristic elements and warning rules, the warning information is determined and sent to the user.

11. An engineering safety risk management system, characterized in that: The system comprises a processor configured to execute the method according to any one of claims 1 to 10.

Images