A Smart Quantitative Assessment Method and System for Community Fire Resilience

Abstract

This invention discloses an intelligent quantitative assessment method and system for community fire resilience, comprising: collecting community fire accident cases, establishing a community fire accident text database based on key stages of community fire resilience, and identifying relevant entities and their relationships; defining the knowledge graph pattern layer of community fire accidents, establishing the knowledge graph, and summarizing and extracting influencing factors of community fire resilience; improving the TOPSIS method based on the characteristics of community fire resilience, screening key influencing factors, establishing a community fire resilience assessment index system, and constructing a resilience network; and establishing a Bayesian network assessment model by combining parameter learning and the Delphi method, conducting causal reasoning and sensitivity analysis, quantitatively assessing the level of community fire resilience, identifying weak links and key influencing indicators of community fire resilience, and proposing targeted improvement strategies. This method and system provide a practical tool for decision-making departments to strengthen fire safety governance and promote the construction of resilient and sustainable communities.

Description

Technical Field

[0001] This invention relates to the field of emergency management technology, specifically to a method and system for intelligent quantitative assessment of community fire resilience. Background Technology

[0002] With the rapid pace of urbanization, the number of urban communities in my country has increased dramatically. Compared to traditional rural settlements, urban communities have a high concentration of people, property, and resources, leading to more fire hazards, more frequent accidents, and increasingly prominent fire safety issues. As an important component of community resilience, community fire resilience reflects a community's ability to prevent, respond to, recover from, and adapt to fire impacts, and is a key indicator for assessing the level of community safety governance. Enhancing community fire resilience has become a crucial issue for sustainable community development and disaster management. Community fire resilience assessment, as an important tool for understanding and improving resilience levels, helps community managers identify weaknesses and key influencing factors in fire management, providing data support and decision-making basis for developing targeted resilience enhancement measures.

[0003] Several scholars have proposed methods for resilience assessment. Su Xin et al. proposed an efficient flood resilience assessment method and system based on natural language processing and data-driven models (Authorization Announcement No. CN 118643997 B). This method collects multi-source data, constructs a flood resilience knowledge graph, selects indicators using correlation and causality analysis, establishes a resilience indicator system, and uses machine learning algorithms to build a resilience assessment model. Wang Jian et al. proposed a community fire resilience assessment method and system based on association rules and complex networks (Application Publication No.: CN 119515157 A). This method analyzes the causes of historical accident cases, establishes a community fire resilience assessment indicator system, and uses association rule algorithms and complex networks to identify key indicators. Huang Ying et al. proposed a community disaster prevention resilience assessment method based on the improved entropy weight-CRITIC method (Authorization Announcement No. CN 114565249 B). This method establishes a community disaster prevention resilience assessment indicator system, calculates indicator weights using field data and the improved entropy weight-CRITIC method, and uses the TOPSIS method to calculate the resilience level of the evaluated object. The patents above reveal that existing assessment methods largely rely on existing data and expert experience, lacking in-depth analysis of historical accident cases. Furthermore, the indicator selection process primarily analyzes the correlation and causality between indicators, neglecting the degree of their impact on disasters. In addition, most assessment methods focus on comprehensive scoring or grading, failing to effectively link assessment results with specific improvement measures and lacking a systematic identification of key weaknesses, thus limiting their value in practical management and decision-making.

[0004] Influencing factor analysis and screening are core steps in constructing an evaluation index system. Regarding influencing factor analysis, Yang Xin et al. proposed a knowledge graph-based prediction method for rail transit accidents (Application Publication No.: CN 116484056A). Based on rail transit accident data, they identified relevant knowledge entities and their interrelationships, established a rail transit hazard source association analysis model based on knowledge graph theory, and clarified its analysis indicators. Zhang Yibin et al. proposed a knowledge graph-based safety risk management method and system (Application Publication No.: CN 117787725 A). By analyzing data related to power grid infrastructure construction safety in infrastructure accident cases and extracting relevant entities and relationships, they constructed a multimodal knowledge graph for power grid infrastructure construction safety management and established a risk identification index system. Regarding the screening of key influencing factors, Shi Zheqi et al. conducted a study on environmental risk screening for petrochemical enterprises based on the TOPSIS-AHP method. They used the TOPSIS method to calculate the weighted sum of squared distances between positive and negative ideal points, ranked the risk indicators according to their superiority or inferiority, and screened key risk indicators. Tong Junhao et al. proposed a method for assessing the fire resistance toughness of highway tunnels (Application Publication No.: CN 118195133 A). Based on the highway tunnel fire resistance toughness assessment index system, they adopted an improved TOPSIS method to calculate the importance, discriminative power, and impact of each index, thereby selecting key indicators. As can be seen from the above patents, current patent applications or authorizations mainly focus on extracting accident causes and modeling the causal relationships of disaster paths, paying less attention to the system resilience characteristics reflected by accidents. They lack in-depth mining of resilience characteristics in historical accidents, making it difficult to support the construction and identification of a resilience-oriented index system. Furthermore, as a multi-objective decision-making method, TOPSIS in existing research does not pay sufficient attention to the characteristics of specific research objects, especially lacking analysis of the dynamic temporal changes in resilience. It has not yet formed a targeted and systematic index selection method, limiting the applicability and effectiveness of this method in the field of resilience assessment.

[0005] Therefore, based on community fire accident cases and combined with key characteristics of community fire resilience, this application develops an intelligent quantitative assessment method and system for community fire resilience using knowledge graphs and an improved TOPSIS method. This addresses the shortcomings of existing assessment methods in terms of utilizing historical experience, identifying phased characteristics, and analyzing the correlation between indicators and disasters. It has significant practical and social value for improving the comprehensive response capabilities of community fires and promoting refined community fire safety management. Summary of the Invention

[0006] This invention aims to provide a method and system for intelligent quantitative assessment of community fire resilience. Based on the key stages of community fire resilience, it deeply analyzes accident investigation reports, establishes a knowledge graph of community fire accidents, and extracts influencing factors of community fire resilience. Using an improved TOPSIS method, it analyzes the impact of these factors on community fires, screens key factors, establishes a community fire resilience assessment index system and resilience network, and based on this, constructs a Bayesian assessment model for community fire resilience, conducts causal reasoning and sensitivity analysis, achieves quantitative assessment of community fire resilience, and identifies weak links and key indicators in community fire safety management.

[0007] On the one hand, the intelligent quantitative assessment method for community fire resilience proposed in this invention includes the following steps:

[0008] Step S1: Collect community fire accident cases, combine them with the key stages of community fire resilience, establish a community fire accident text database, and identify relevant entities and their interrelationships related to community fire resilience.

[0009] Furthermore, the aforementioned key stages of community fire resilience, based on resilience theory and combined with the actual work of community fire management, are divided into the fire prevention stage, the emergency response stage, the post-disaster recovery stage, and the improvement and enhancement stage.

[0010] Furthermore, the community fire accident cases are stored in the community fire accident text database after format preprocessing. The community fire accident text database includes at least columns for basic accident information, accident causes, emergency response measures, post-disaster handling measures, and prevention measures.

[0011] Optionally, the entities and relationships related to community fire resilience can be described using natural language processing tools; no specific restrictions are imposed here.

[0012] Step S2: Based on the community fire safety resilience entities and their relationships, clarify the pattern layer structure of the community fire accident knowledge graph, establish the community fire accident knowledge graph, and summarize and extract the influencing factors of community fire safety resilience; the pattern layer of the community fire accident knowledge graph includes a basic information layer, a resilience stage layer, an accident cause layer, and an influencing factor layer, which are used to describe the constituent elements and logical relationships of the community fire accident cases.

[0013] Furthermore, the factors influencing community fire resilience include those in the fire prevention phase, emergency response phase, recovery phase, and enhancement phase. Each category of influencing factors is further subdivided into personnel factors, building factors, environmental factors, and management factors, specifically including:

[0014] The main influencing factors in the fire prevention stage include: human factors: insufficient fire risk prevention capabilities among residents; building factors: old buildings, tall buildings, low fire resistance ratings, missing or ineffective fire compartments, abnormal operation of internal equipment, and use of flammable decoration and finishing materials; environmental factors: the accumulation of combustible materials and the illegal addition of facilities; and management factors: inadequate supervision of six groups of people, insufficient efforts in hazard investigation and rectification, insufficient funding for fire safety work, an imperfect fire safety governance system, inadequate fire safety publicity, education and training, and inadequate maintenance of fire protection facilities.

[0015] The main influencing factors in the emergency response phase include: personnel factors: insufficient emergency response and escape capabilities; building factors: insufficient unobstructed evacuation routes, incomplete or ineffective fire-fighting facilities; environmental factors: missing or ineffective outdoor fire-fighting facilities, and obstruction of fire truck access or elevated areas; and management factors: inadequate construction of basic rescue forces, and lack of or incomplete emergency plans.

[0016] The main influencing factors during the recovery phase include management factors: insurance claims, public opinion guidance, compensation for victims, medical assistance, resettlement of residents after the disaster, and repair and replenishment of equipment after the disaster.

[0017] The factors influencing the improvement phase mainly include management factors: improving the fire protection governance system and local supervision mechanism, deepening residents' fire safety education and awareness, strengthening the construction of diverse fire protection forces and emergency mobilization capabilities, the application and use of new fire protection technologies, and the system for learning from historical accident cases.

[0018] Optionally, the community fire accident knowledge graph can be built using the Neo4j graph database; no specific restrictions are imposed here.

[0019] Step S3: Based on the characteristics of community fire resilience, improve the TOPSIS method, conduct quantitative analysis of the influencing factors of community fire resilience, screen key influencing factors, establish a community fire resilience assessment index system, and construct a community fire resilience network.

[0020] Furthermore, the TOPSIS method is improved based on the characteristics of community fire resilience. The specific steps are as follows:

[0021] The characteristics of community fire resilience stages were analyzed, and the attributes of the influencing factors were determined. Specifically, the influencing factors for the fire prevention stage are the probability of occurrence, the severity of consequences, and the controllability; the influencing factors for the emergency response stage are the severity of consequences, the controllability, and the timeliness; the influencing factors for the post-disaster recovery stage are the recovery speed, resource input, and the degree of recovery; and the influencing factors for the improvement and enhancement stage are the experience conversion rate, the degree of capability enhancement, and the system redundancy.

[0022] Knowledge from K experts was collected through questionnaires, and the attributes of the factors influencing community fire resilience mentioned in stage t were quantitatively scored to establish an initial decision matrix X:

[0023]

[0024] In the formula, t represents the community fire resilience stage. These are the initial decision matrices for the fire prevention phase, the emergency response phase, the post-disaster recovery phase, and the improvement and enhancement phase, respectively. For the Kth expert's analysis of the factors influencing community fire resilience in stage t. The There are several attribute values, where m is the total number of community fire resilience influencing factors in stage t, i=1, 2, 3…m, j=1, 2, 3;

[0025] The initial decision matrix is ​​normalized to establish a normalized matrix. :

[0026]

[0027] In the formula, The normalized value of the j-th attribute of the community fire resilience influencing factor i in stage t;

[0028] Knowledge from K experts was collected using a questionnaire survey to evaluate the influence of attribute q on attribute j at stage t, with a scoring range of [-1, 1]. An attribute correlation matrix was then established. :

[0029]

[0030] In the formula, This represents the evaluation of the degree of influence of attribute q on attribute j by the Kth expert in stage t, where q = 1, 2, 3; This represents the average value of the evaluation of the degree of influence. >0 indicates an enhancing effect. <0 indicates an inhibitory effect;

[0031] To enhance the expressive power of the community fire resilience in stage t, the influence of adjacent attributes is introduced into the attributes of the factors influencing community fire resilience, and an enhanced attribute value matrix is ​​established. :

[0032]

[0033]

[0034] In the formula, This represents the attribute value after considering the interaction between attributes. This represents the normalized value of the q-th attribute of the community fire resilience influencing factor i in stage t.

[0035] For the enhanced attribute value matrix Perform normalization processing to establish a normalized enhanced attribute value matrix. :

[0036]

[0037] In the formula, This refers to the normalized result of the attribute values ​​after considering the mutual influence between attributes;

[0038] The attribute weights in stage t for:

[0039]

[0040] In the formula, Furthermore, the sum of the attribute weights in stage t is 1; The attribute information quantity represents the reciprocal of the sum of the squared distances from all influencing factor values ​​of attribute j in stage t to boundary points 0 and 1.

[0041] To calculate the optimal weight vector of the attributes of the community fire resilience influencing factors in stage t. When distance is used as a constraint, the sum of the weighted squared distances of all the influencing factor values ​​in stage t should be satisfied. Minimum;

[0042] The objective function is:

[0043]

[0044] Factors affecting community fire resilience in stage t Calculate the influence index :

[0045]

[0046] For the community fire resilience influencing factor i in stage t, calculate the identifiability. :

[0047]

[0048] In the formula, Let be the entropy weight coefficient of the community fire resilience influencing factor i in stage t. The entropy value of the community fire resilience influencing factor i in stage t is calculated using the following formula:

[0049]

[0050]

[0051] In the formula, The attribute influence ratio of the community fire resilience influencing factor i in stage t is calculated using the following formula:

[0052]

[0053] In the formula, This is the global influence ratio coefficient of the community fire resilience influencing factor i in stage t;

[0054] Calculate the contribution of influencing factors :

[0055]

[0056] For the community fire resilience influencing factors at stage t, a threshold is set, and the corresponding values ​​are calculated. , and ;when At that time, the factors affecting community fire resilience were determined to have low influence, i.e., unimportant; when When the factors affecting community fire resilience are determined to have no significant impact on decision-making; when At that time, the factors affecting the community's fire resilience were determined to have a low contribution.

[0057] Step S4: Based on the community fire resilience network, a Bayesian network evaluation model for community fire resilience is established by combining the parameter learning method and the Delphi method to calculate node probabilities. The parameter learning method constructs a parameter learning training set based on the community fire accident cases and uses the expectation-maximization algorithm to calculate the prior and conditional probabilities of nodes. The Delphi method collects expert knowledge through questionnaires and uses defuzzification and the Noisy-MAX model to supplement the node probabilities that are difficult to quantify in parameter learning.

[0058] Step S5: Use the community fire resilience Bayesian network assessment model to perform causal reasoning and sensitivity analysis, quantitatively assess the community fire resilience level, and identify weak links and key indicators affecting community fire resilience.

[0059] Furthermore, the quantitative assessment result of the community fire resilience level is the edge probability of the top node in the community fire resilience Bayesian network assessment model.

[0060] Optionally, the sensitivity analysis may use the mutual information method, and no specific restrictions are imposed here.

[0061] Step S6: Propose targeted strategies to enhance community fire resilience, addressing the weak links and key indicators affecting community fire resilience.

[0062] On the other hand, the present invention provides a community fire resilience intelligent quantitative assessment system, the system comprising: a historical fire accident analysis module, an indicator system construction module, a community fire resilience assessment module, and a community fire resilience improvement module, the modules being connected in sequence.

[0063] Furthermore, the historical fire accident analysis module is used to process collected community fire accident cases, extract relevant entities and relationships of community fire resilience, and identify influencing factors of community fire resilience; the indicator system construction module is used to screen key influencing factors that have a high impact on community fire resilience, strong distinguishing characteristics, and high contribution, and establish a community fire resilience assessment indicator system; the community fire resilience assessment module is used to conduct quantitative assessment of community fire resilience and screen weak links and key indicators of community fire resilience; the community fire resilience improvement module proposes targeted improvement strategies based on weak links and key indicators.

[0064] Compared with the prior art, the beneficial effects of this invention are:

[0065] This invention systematically analyzes historical fire accidents in communities from the perspective of community fire resilience, establishes a knowledge graph of community fire accidents, and deeply mines the resilience elements reflected in the accidents and their inherent relationships, achieving structured integration and semantic expression of unstructured accident information. Based on the key stages of resilience, it uses an improved TOPSIS method to introduce influencing factor attributes strongly correlated with accidents, screens key influencing factors, and constructs a community fire resilience assessment index system, improving the pertinence and scientific nature of index selection and overcoming the problems of vague stage division and incomplete index coverage in traditional assessment systems. The resilience assessment model based on Bayesian networks supports causal reasoning and sensitivity analysis, realizing dynamic quantitative assessment of fire resilience levels in complex community systems, effectively identifying weak links and key influencing factors in fire management. Ultimately, it optimizes the allocation and governance strategies of community fire resources, improving the comprehensive fire response capability and the refinement level of fire safety governance in communities. Attached Figure Description

[0066] Figure 1 This is a flowchart of a community fire resilience intelligent quantitative assessment method according to an embodiment of the present invention;

[0067] Figure 2 This describes the entity and relation extraction process in an embodiment of the present invention.

[0068] Figure 3 This is a schematic diagram of the knowledge graph pattern layer according to an embodiment of the present invention;

[0069] Figure 4 This is a schematic diagram of a knowledge graph according to an embodiment of the present invention;

[0070] Figure 5 These are the factors influencing community fire resilience in embodiments of the present invention;

[0071] Figure 6 The results of the improved TOPSIS method in this embodiment of the invention are shown.

[0072] Figure 7 This is a Bayesian assessment model for community fire resilience in an embodiment of the present invention;

[0073] Figure 8 This is a schematic diagram of the structural composition of a community fire resilience intelligent quantitative assessment system according to an embodiment of the present invention. Detailed Implementation

[0074] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0075] Example 1

[0076] like Figure 1 As shown, the intelligent quantitative assessment method for community fire resilience in this embodiment includes the following steps:

[0077] Step S1: Collect community fire accident cases, combine them with the key stages of community fire resilience, establish a community fire accident text database, and identify relevant entities and their interrelationships related to community fire resilience.

[0078] The key stages of community fire resilience, based on resilience theory and combined with the actual situation of community fire management, are divided into the fire prevention stage, the emergency response stage, the post-disaster recovery stage, and the improvement and enhancement stage.

[0079] In this embodiment, the community fire accident cases are sourced from unstructured text data such as government public databases, news reports, accident investigation reports, and industry annual reports. A total of 175 community fire accident cases were collected. After format preprocessing, the accident cases are stored in the community fire accident text database. The community fire accident text database includes columns for accident time, accident location, accident consequences, accident causes, emergency response measures, post-accident handling measures, and prevention measures.

[0080] like Figure 2As shown, this embodiment utilizes the Natural Language Processing Platform (LTP) and employs modules such as word segmentation, part-of-speech tagging, dependency parsing, and semantic role labeling to collaboratively process and extract the community fire resilience-related entities and their relationships.

[0081] Step S2: Based on the community fire safety resilience entities and their relationships, clarify the pattern layer structure of the community fire accident knowledge graph, establish the community fire accident knowledge graph, and summarize and extract the influencing factors of community fire safety resilience; the pattern layer of the community fire accident knowledge graph includes a basic information layer, a resilience stage layer, an accident cause layer, and an influencing factor layer, which are used to describe the constituent elements and logical relationships of the community fire accident cases.

[0082] The community fire accident knowledge graph model layer in this embodiment is as follows: Figure 3 As shown.

[0083] like Figure 4 As shown, this embodiment uses the Neo4j graph database to store and visualize community fire incidents.

[0084] like Figure 5 As shown, this embodiment clarifies the influencing factors of community fire resilience based on the relevant entities of the community fire resilience and the fire supervision and inspection records.

[0085] Step S3: Based on the characteristics of community fire resilience, improve the TOPSIS method, conduct quantitative analysis of the influencing factors of community fire resilience, screen key influencing factors, establish a community fire resilience assessment index system, and construct a community fire resilience network.

[0086] The improved TOPSIS method described herein comprises the following steps:

[0087] The characteristics of community fire resilience stages were analyzed, and the attributes of the influencing factors were determined. Specifically, the influencing factors for the fire prevention stage are the probability of occurrence, the severity of consequences, and the controllability; the influencing factors for the emergency response stage are the severity of consequences, the controllability, and the timeliness; the influencing factors for the post-disaster recovery stage are the recovery speed, resource input, and the degree of recovery; and the influencing factors for the improvement and enhancement stage are the experience conversion rate, the degree of capability enhancement, and the system redundancy.

[0088] Knowledge from K experts was collected through questionnaires, and the attributes of the factors influencing community fire resilience mentioned in stage t were quantitatively scored to establish an initial decision matrix X:

[0089]

[0090] In the formula, t represents the community fire resilience stage. These are the initial decision matrices for the fire prevention phase, the emergency response phase, the post-disaster recovery phase, and the improvement and enhancement phase, respectively. For the Kth expert's analysis of the factors influencing community fire resilience in stage t. The There are several attribute values, where m is the total number of community fire resilience influencing factors in stage t, i=1, 2, 3…m, j=1, 2, 3;

[0091] The initial decision matrix is ​​normalized to establish a normalized matrix. :

[0092]

[0093] In the formula, The normalized value of the j-th attribute of the community fire resilience influencing factor i in stage t;

[0094] Knowledge from K experts was collected using a questionnaire survey to evaluate the influence of attribute q on attribute j at stage t, with a scoring range of [-1, 1]. An attribute correlation matrix was then established. :

[0095]

[0096] In the formula, This represents the evaluation of the degree of influence of attribute q on attribute j by the Kth expert in stage t, where q = 1, 2, 3; This represents the average value of the evaluation of the degree of influence. >0 indicates an enhancing effect. <0 indicates an inhibitory effect;

[0097] To enhance the expressive power of the community fire resilience in stage t, the influence of adjacent attributes is introduced into the attributes of the factors influencing community fire resilience, and an enhanced attribute value matrix is ​​established. :

[0098]

[0099]

[0100] In the formula, This represents the attribute value after considering the interaction between attributes. This represents the normalized value of the q-th attribute of the community fire resilience influencing factor i in stage t.

[0101] For the enhanced attribute value matrix Perform normalization processing to establish a normalized enhanced attribute value matrix. :

[0102]

[0103] In the formula, This refers to the normalized result of the attribute values ​​after considering the mutual influence between attributes;

[0104] The attribute weights in stage t for:

[0105]

[0106] In the formula, Furthermore, the sum of the attribute weights in stage t is 1; The attribute information quantity represents the reciprocal of the sum of the squared distances from all influencing factor values ​​of attribute j in stage t to boundary points 0 and 1.

[0107] To calculate the optimal weight vector of the attributes of the community fire resilience influencing factors in stage t. When distance is used as a constraint, the sum of the weighted squared distances of all the influencing factor values ​​in stage t should be satisfied. Minimum;

[0108] The objective function is:

[0109]

[0110] Factors affecting community fire resilience in stage t Calculate the influence index :

[0111]

[0112] For the community fire resilience influencing factor i in stage t, calculate the identifiability. :

[0113]

[0114] In the formula, Let be the entropy weight coefficient of the community fire resilience influencing factor i in stage t. The entropy value of the community fire resilience influencing factor i in stage t is calculated using the following formula:

[0115]

[0116]

[0117] In the formula, The attribute influence ratio of the community fire resilience influencing factor i in stage t is calculated using the following formula:

[0118]

[0119] In the formula, This is the global influence ratio coefficient of the community fire resilience influencing factor i in stage t;

[0120] Calculate the contribution of influencing factors :

[0121]

[0122] For the community fire resilience influencing factors at stage t, a threshold is set, and the corresponding values ​​are calculated. , and ;when At that time, the factors affecting community fire resilience were determined to have low influence, i.e., unimportant; when When the factors affecting community fire resilience are determined to have no significant impact on decision-making; when At that time, the factors affecting the community's fire resilience were determined to have a low contribution.

[0123] This embodiment invited 33 experts in the field of fire resilience to use a five-point Likert scale to quantify and score the attributes of the influencing factors and the degree of influence between their respective attributes, and to use... =3, =3, =3 is the threshold, such as Figure 6 The figure shows the screening results of the community fire resilience influencing factors using the improved TOPSIS method in this embodiment. A total of 23 key influencing factors were extracted to establish the community fire resilience assessment index and the community fire resilience network.

[0124] Step S4: Based on the community fire resilience network, a Bayesian network evaluation model for community fire resilience is established by combining the parameter learning method and the Delphi method to calculate node probabilities. The parameter learning method constructs a parameter learning training set based on the community fire accident cases and uses the expectation-maximization algorithm to calculate the prior and conditional probabilities of nodes. The Delphi method collects expert knowledge through questionnaires and uses defuzzification and the Noisy-MAX model to supplement the node probabilities that are difficult to quantify in parameter learning.

[0125] like Figure 7 As shown, this embodiment uses the "GeNIe 4.0 Academic" software to establish the Bayesian network evaluation model based on the community fire resilience network.

[0126] Step S5: Use the Bayesian network evaluation model to perform causal reasoning and sensitivity analysis to quantitatively assess the community's fire resilience level and identify weak links and key indicators affecting the community's fire resilience.

[0127] The quantitative assessment result of the community's fire resilience level is the edge probability of the top node in the BN model.

[0128] This embodiment uses the Bayesian network assessment model for community fire resilience to conduct causal inference and finds that: the probability of Chinese communities being in a "High Resilience" state is 62%. The highest probability of failure is found in the lack of a sound fire governance system and local supervision mechanisms, as well as insufficient emergency response and escape capabilities, both at 6%, representing key weaknesses in improving community fire resilience at present. Furthermore, electrical equipment failures, inadequate hazard identification and management, poor emergency response and escape capabilities among residents, and obstruction of fire rescue routes are the main reasons for reduced community fire resilience. Sensitivity analyses were conducted using mutual information methods on the fire prevention stage, the emergency response stage, the post-disaster recovery stage, and the improvement and enhancement stage. The results showed that: the accumulation of combustible and flammable materials was the most sensitive factor to fire prevention capabilities, with an entropy reduction of 5.68%, followed by abnormal internal equipment operation, with an entropy reduction of 4.47%; insufficient emergency response and escape capabilities were key factors in improving community fire emergency response capabilities, with an entropy reduction of 5.65%; among recovery capabilities, medical assistance and support were the most sensitive, with an entropy reduction of 4.38%; and among improvement and enhancement capabilities, establishing a sound fire management system and local supervision mechanism, as well as a system for learning from historical accident cases, were highly sensitive, with entropy reductions of 3.51% and 3.81%, respectively.

[0129] Step S6: Propose targeted strategies to enhance community fire resilience, addressing the weak links and key indicators affecting community fire resilience.

[0130] Based on the key weaknesses and key indicators described in this embodiment, taking the fire prevention stage as an example, the following targeted community fire safety resilience enhancement strategies are proposed: Regarding fire prevention capabilities, communities should broaden their publicity and education channels and, in conjunction with practical considerations, establish a tiered and categorized fire safety publicity and education mechanism targeting key groups such as the elderly, conducting differentiated education to improve residents' awareness of fire risks and prevention. Simultaneously, a hazard investigation and rectification system should be formulated and implemented, based on a grid management system, ensuring 100% coverage of key inspection areas such as stairwells and fire truck access routes, forming a closed-loop hazard management system.

[0131] Example 2

[0132] This embodiment provides a community fire resilience intelligent quantitative assessment system based on Embodiment 1, such as... Figure 8As shown in the schematic diagram of the system disclosed in this embodiment, the system includes: a historical fire accident analysis module, an indicator system construction module, a community fire resilience assessment module, and a community fire resilience enhancement module, with each module connected in sequence.

[0133] The historical fire accident analysis module processes collected community fire accident cases, extracts relevant entities and relationships related to community fire resilience, and identifies influencing factors. The indicator system construction module screens key influencing factors that have a high impact on community fire resilience, strong distinguishing characteristics, and high contribution, establishing a community fire resilience assessment indicator system. The community fire resilience assessment module conducts quantitative assessments of community fire resilience, identifying weak links and key indicators. The community fire resilience improvement module proposes targeted improvement strategies based on these weak links and key indicators.

[0134] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent quantitative assessment of community fire resilience, characterized in that, The method includes the following steps: Step S1: Collect community fire accident cases, combine them with the key stages of community fire resilience, establish a community fire accident text database, and identify relevant entities and their interrelationships related to community fire resilience. Step S2: Based on the community fire safety resilience entities and their relationships, clarify the pattern layer structure of the community fire accident knowledge graph, establish the community fire accident knowledge graph, and summarize and extract the influencing factors of community fire safety resilience; the pattern layer of the community fire accident knowledge graph includes a basic information layer, a resilience stage layer, an accident cause layer, and an influencing factor layer, which are used to describe the constituent elements and logical relationships of the community fire accident cases. Step S3: Based on the characteristics of community fire resilience, improve the TOPSIS method, quantitatively analyze the influencing factors of community fire resilience, screen key influencing factors, establish a community fire resilience assessment index system, and construct a community fire resilience network. Step S4: Based on the community fire resilience network, a community fire resilience Bayesian network evaluation model is established by combining the parameter learning method and the Delphi method to calculate node probabilities. The parameter learning method constructs a parameter learning training set based on the community fire accident cases and uses the expectation-maximization algorithm to calculate the prior and conditional probabilities of nodes. The Delphi method uses a questionnaire survey to collect expert knowledge and employs defuzzification and the Noisy-MAX model to supplement the node probabilities that are difficult to quantify in parameter learning. Step S5: Use the community fire resilience Bayesian network assessment model to perform causal reasoning and sensitivity analysis, quantitatively assess the community fire resilience level, and identify weak links and key indicators affecting community fire resilience. Step S6: Propose targeted strategies to enhance community fire resilience, addressing the weak links and key indicators affecting community fire resilience.

2. The method according to claim 1, characterized in that, The key stages of community fire resilience described in step S1 are divided into fire prevention, emergency response, post-disaster recovery, and improvement stages, based on resilience theory and combined with the actual work of community fire management.

3. The method according to claim 1, characterized in that, The community fire accident cases mentioned in step S1 are stored in the community fire accident text database after format preprocessing. The community fire accident text database includes at least the following columns: basic accident information, accident cause, emergency response measures, aftermath handling measures, and prevention measures.

4. The method according to claim 1, characterized in that, The factors influencing community fire resilience mentioned in step S2 include factors influencing the fire prevention phase, the emergency response phase, the recovery phase, and the enhancement phase. Each category of factors is further subdivided into personnel factors, building factors, environmental factors, and management factors, specifically including: The main influencing factors in the fire prevention stage include: human factors: insufficient fire risk prevention capabilities among residents; building factors: old buildings, tall buildings, low fire resistance ratings, missing or ineffective fire compartments, abnormal operation of internal equipment, and use of flammable decoration and finishing materials; environmental factors: the accumulation of combustible materials and the illegal addition of facilities; and management factors: inadequate supervision of six groups of people, insufficient efforts in hazard investigation and rectification, insufficient funding for fire safety work, an imperfect fire safety governance system, inadequate fire safety publicity, education and training, and inadequate maintenance of fire protection facilities. The main influencing factors in the emergency response phase include: personnel factors: insufficient emergency response and escape capabilities; building factors: insufficient unobstructed evacuation routes, incomplete or ineffective fire-fighting facilities; environmental factors: missing or ineffective outdoor fire-fighting facilities, and obstruction of fire truck access or elevated areas; and management factors: inadequate construction of basic rescue forces, and lack of or incomplete emergency plans. The main influencing factors during the recovery phase include management factors: insurance claims, public opinion guidance, compensation for victims, medical assistance, resettlement of residents after the disaster, and repair and replenishment of equipment after the disaster. The factors influencing the improvement phase mainly include management factors: improving the fire protection governance system and local supervision mechanism, deepening residents' fire safety education and awareness, strengthening the construction of diverse fire protection forces and emergency mobilization capabilities, the application and use of new fire protection technologies, and the system for learning from historical accident cases.

5. The method according to claim 2, characterized in that, Step S3 describes an improved TOPSIS method based on the characteristics of community fire resilience. The specific steps are as follows: The characteristics of community fire resilience stages were analyzed, and the attributes of the influencing factors were determined. Specifically, the influencing factors for the fire prevention stage are the probability of occurrence, the severity of consequences, and the controllability; the influencing factors for the emergency response stage are the severity of consequences, the controllability, and the timeliness; the influencing factors for the post-disaster recovery stage are the recovery speed, resource input, and the degree of recovery; and the influencing factors for the improvement and enhancement stage are the experience conversion rate, the degree of capability enhancement, and the system redundancy. Knowledge from K experts was collected through questionnaires, and the attributes of the factors influencing community fire resilience mentioned in stage t were quantitatively scored to establish an initial decision matrix X: In the formula, t represents the community fire resilience stage. These are the initial decision matrices for the fire prevention phase, the emergency response phase, the post-disaster recovery phase, and the improvement and enhancement phase, respectively. For the Kth expert's analysis of the factors influencing community fire resilience in stage t. The There are several attribute values, where m is the total number of community fire resilience influencing factors in stage t, i=1, 2, 3…m, j=1, 2, 3; The initial decision matrix is ​​normalized to establish a normalized matrix. : In the formula, The normalized value of the j-th attribute of the community fire resilience influencing factor i in stage t; Knowledge from K experts was collected using a questionnaire survey to evaluate the influence of attribute q on attribute j at stage t, with a scoring range of [-1, 1]. An attribute correlation matrix was then established. : In the formula, This represents the evaluation of the degree of influence of attribute q on attribute j by the Kth expert in stage t, where q = 1, 2, 3; This represents the average value of the evaluation of the degree of influence. >0 indicates an enhancing effect. <0 indicates an inhibitory effect; To enhance the expressive power of the community fire resilience in stage t, the influence of adjacent attributes is introduced into the attributes of the factors influencing community fire resilience, and an enhanced attribute value matrix is ​​established. : In the formula, This represents the attribute value after considering the interaction between attributes. This represents the normalized value of the q-th attribute of the community fire resilience influencing factor i in stage t. For the enhanced attribute value matrix Perform normalization processing to establish a normalized enhanced attribute value matrix. : In the formula, This refers to the normalized result of the attribute values ​​after considering the mutual influence between attributes; Attribute weights in stage t for: In the formula, Furthermore, the sum of the attribute weights in stage t is 1; The attribute information quantity represents the reciprocal of the sum of the squared distances from all influencing factor values ​​of attribute j in stage t to boundary points 0 and 1. To calculate the optimal weight vector of the attributes of the community fire resilience influencing factors in stage t. When distance is used as a constraint, the sum of the weighted squared distances of all the influencing factor values ​​in stage t should be satisfied. Minimum; The objective function is: Factors affecting community fire resilience in stage t Calculate the influence index : For the community fire resilience influencing factor i in stage t, calculate the identifiability. : In the formula, Let be the entropy weight coefficient of the community fire resilience influencing factor i in stage t. The entropy value of the community fire resilience influencing factor i in stage t is calculated using the following formula: In the formula, The attribute influence ratio of the community fire resilience influencing factor i in stage t is calculated using the following formula: In the formula, This is the global influence ratio coefficient of the community fire resilience influencing factor i in stage t; Calculate the contribution of influencing factors : For the community fire resilience influencing factors at stage t, a threshold is set, and the corresponding values ​​are calculated. , and ;when At that time, the factors affecting community fire resilience were determined to have low influence, i.e., unimportant; when When the factors affecting community fire resilience are determined to have no significant impact on decision-making; when At that time, the factors affecting the community's fire resilience were determined to have a low contribution.

6. The method according to claim 1, characterized in that, The quantitative assessment result of the community fire resilience level in step S5 is the edge probability of the top node in the community fire resilience Bayesian network assessment model.

7. A community fire resilience intelligent quantitative assessment system, used to perform the method according to any one of claims 1-6, characterized in that, The system includes: a historical fire accident analysis module, an indicator system construction module, a community fire resilience assessment module, and a community fire resilience improvement module, with each module connected in sequence. The historical fire accident analysis module is used to process collected community fire accident cases, extract relevant entities and relationships of community fire resilience, and identify the influencing factors of community fire resilience; The indicator system construction module is used to screen key influencing factors that have a high impact on community fire resilience, strong distinguishing characteristics, and high contribution, and to establish a community fire resilience assessment indicator system. The community fire resilience assessment module is used to conduct quantitative assessments of community fire resilience and to identify weak links and key indicators in community fire resilience. The community fire safety resilience enhancement module proposes targeted improvement strategies based on weak links and key indicators.

Images