Worker unsafe behavior trend prediction method and system based on dual index clustering

Abstract

The invention relates to the technical field of trend prediction, in particular to a worker unsafe behavior trend prediction method and system based on dual index clustering. The method comprises the following steps: carrying out stratified sampling on all coal miners; constructing a basic risk index and an unsafe behavior risk index; based on the basic risk index and the unsafe behavior risk index, kmeans clustering is carried out on individuals, teams or districts in the whole amount of coal miners; constructing a Bayesian discrimination model, and carrying out class cluster division on individuals, teams or districts in the whole amount of coal miners based on the Bayesian discrimination model; and on the basis of a class cluster division result, unsafe behaviors of individuals, teams or district teams in all coal mine workers are predicted, and corresponding prevention safety education plans and prevention management and control measures are formulated. Therefore, the problem that an existing prediction method cannot be suitable for the basic risk index and the unsafe behavior risk index at the same time is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of trend prediction, and in particular, to a method and system for predicting the trend of workers' unsafe behaviors based on dual-index clustering. Background Art

[0002] Underground coal mine operations have always been of relatively high risk. According to statistics, more than half of the underground coal mine time is due to the personal basic risks and unsafe behaviors of workers. The safety education for coal miners and preventive measures for their behaviors have always been the key tasks in underground operations. Therefore, the prediction of the trend of workers' unsafe behaviors is particularly important.

[0003] In the prior art, most of the predictions of coal miners' behaviors are based on the method of clustering coal miners. By analyzing the multi-dimensional attributes of workers to construct user portraits, each worker is regarded as a node, and clustering is carried out using the similarity values of node vectors. However, this method cannot ensure the applicability of the clustering results to subsequent behaviors. In addition, the prior art for predicting the trend of unsafe behaviors is limited to predicting the number of unsafe behaviors through other indicators, and does not include profiling different levels of objects (such as individuals, teams or sections) and predicting unsafe behaviors.

[0004] In view of the above problems, the present invention proposes a method and system for predicting the trend of workers' unsafe behaviors based on dual-index clustering. Summary of the Invention

[0005] To solve the problem that the existing prediction methods cannot be applicable to both the basic risk indicators and the unsafe behavior risk indicators at the same time, the present invention provides a method and system for predicting the trend of workers' unsafe behaviors based on dual-index clustering.

[0006] In a first aspect, the present invention provides a method for predicting the trend of workers' unsafe behaviors based on dual-index clustering, including:

[0007] Performing stratified sampling on all coal miners;

[0008] Constructing basic risk indicators based on the personal basic indicators, safety education indicators, work situation indicators and health physiological indicators of the sampled workers;

[0009] Constructing unsafe behavior risk indicators based on the personal unsafe behavior events of the sampled workers, and the personal unsafe behavior events are obtained based on the accident dimension, awareness dimension and overall dimension;

[0010] Performing kmeans clustering on individuals, teams or sections among all coal miners based on the basic risk indicators and the unsafe behavior risk indicators;

[0011] Construct a Bayesian discriminant model, and divide all coal miners into individual, team or district teams based on the Bayesian discriminant model;

[0012] Based on the results of the cluster division, the unsafe behaviors of individuals, teams or teams among all coal miners are predicted, and corresponding preventive safety education plans and preventive control measures are formulated.

[0013] In some embodiments, the personal basic indicators specifically include length of service, education level and professional skill level;

[0014] The safety education indicators include safety education participation and safety education assessment scores;

[0015] The work situation indicators include attendance, social exchange relationship I and social exchange relationship O;

[0016] The health physiological indicators include cardiovascular and cerebrovascular health, basal metabolic health, fatigue level, heart rate abnormality and body temperature abnormality.

[0017] In some embodiments, the accident-causing dimensions include: operating errors / ignoring safety / ignoring warnings, causing safety device failure, using unsafe equipment, operating with hands instead of tools, improper storage of objects, venturing into dangerous places, climbing in unsafe positions, working / staying under the crane arm, refueling / repairing / checking the machine while it is running, distracting behavior, not using personal protective equipment / equipment correctly, unsafe clothing, and incorrect handling of flammable / explosive and other dangerous goods;

[0018] The awareness dimension includes conscious unsafe behaviors and unconscious unsafe behaviors;

[0019] The overall dimension is the number of unsafe behaviors.

[0020] In some embodiments, the kmeans clustering of the basic risk indicators and unsafe behavior risk indicators of individuals, teams or teams among all coal miners includes:

[0021] Quantify the basic risk indicators and unsafe behavior risk indicators of individuals, teams or teams among all coal miners;

[0022] Initialize the k value to 2 and cluster the quantified basic risk indicators and unsafe behavior risk indicators;

[0023] After completing the clustering of the initialized k value, the value of k is increased from the initial value, and the quantified basic risk indicators and unsafe behavior risk indicators are clustered again;

[0024] After each clustering, the sum of squares of the data errors within each cluster is calculated;

[0025] Draw a line chart of k based on the sum of squared errors of data for each clustering.

[0026] Determine the optimal k value using the elbow method based on the line chart.

[0027] Select the clustering result corresponding to the optimal k value as the clustering result of kmeans clustering.

[0028] Where k is the number of clusters.

[0029] In some embodiments, calculating the sum of squared errors within each cluster SSE, and its formula is:

[0030] ;

[0031] Where SSE is the sum of squared errors of data, m is the total number of samples, is the i-th sample, is the center point of the j-th cluster, is the indicator function, is the sample and the squared Euclidean distance between the cluster center and when belongs to then has a value of 1, otherwise it is 0.

[0032] In some embodiments, constructing a Bayesian discriminant model and performing cluster division on individuals, teams or sections among all coal miners based on the Bayesian discriminant model, including:

[0033] Based on each cluster of the clustering results, calculate the prior probability of each cluster;

[0034] Based on each cluster of the clustering results, estimate the conditional probability density function of each cluster;

[0035] According to Bayes' law, calculate the posterior probability that an individual, team or section among all coal miners belongs to each cluster in the clustering results, and its formula is:

[0036] ;

[0037] Where, is the posterior probability, is the prior probability, is the conditional probability density function, is the marginal probability of the sample x;

[0038] Assign an individual, team or section among all coal miners to the cluster with the maximum posterior probability in the clustering results.

[0039] In some embodiments, the mean value of the unsafe behavior risk indicators in the class clusters to which individuals, teams or work areas in the total number of coal miners belong is used as the first unsafe behavior portrait;

[0040] Based on the first unsafe behavior portrait, an safety education plan and control measures are formulated for the corresponding individuals, teams or work areas.

[0041] In some embodiments, based on the result of the class cluster division, the unsafe behaviors of individuals, teams or work areas among the total number of coal miners are predicted, and corresponding preventive safety education plans and preventive control measures are formulated, including:

[0042] The basic risk indicators and unsafe behavior risk indicators of individuals, teams or work areas at the T-1 stage among the total number of coal miners are input into the Bayesian discriminant model to obtain the first class cluster of individuals, teams or work areas at the T-1 stage among the total number of coal miners;

[0043] The basic risk indicators of individuals, teams or work areas at the T stage and the unsafe behavior risk indicators at the T-1 stage among the total number of coal miners are input into the Bayesian discriminant model to obtain the second class cluster of individuals, teams or work areas at the T stage among the total number of coal miners;

[0044] In response to the first class cluster and the second class cluster not being the same class cluster, the mean value of the unsafe behavior risk indicators of individuals, teams or work areas in the second class cluster at the T stage is used as the second unsafe behavior portrait;

[0045] Based on the second unsafe behavior portrait, corresponding preventive safety education plans and preventive control measures are formulated.

[0046] In some embodiments, in response to the first class cluster and the second class cluster being the same class cluster, the safety education plan and control measures are continued to be used for the individuals, teams or work areas among the total number of coal miners.

[0047] In a second aspect, the present invention proposes a system for predicting the trend of workers' unsafe behaviors based on dual-index clustering, including the following modules:

[0048] Sampling module: used to implement stratified sampling of the total number of coal miners;

[0049] Quantification module: used to construct basic risk indicators based on the personal basic indicators, safety education indicators, work situation indicators and health physiological indicators of the sampled workers, and construct unsafe behavior risk indicators based on the personal unsafe behavior events of the sampled workers, and the personal unsafe behavior events are obtained based on the accident dimension, awareness dimension and overall dimension;

[0050] Clustering module: used to implement kmeans clustering of individuals, teams or work areas among the total number of coal miners based on the basic risk indicators and unsafe behavior risk indicators;

[0051] Discrimination module: It is used to implement the construction of a Bayesian discrimination model, and based on the Bayesian discrimination model, classify individuals, teams or sections among all coal miners into clusters, predict the unsafe behaviors of individuals, teams or sections among all coal miners based on the results of the cluster classification, and formulate corresponding preventive safety education plans and preventive control measures.

[0052] To solve the problem that the existing prediction methods cannot be applied to both basic risk indicators and unsafe behavior risk indicators at the same time, the present invention has the following advantages:

[0053] Through the technical solution of the present invention, the trend of unsafe behaviors of individuals, teams or sections can be predicted based on the basic risk indicators and unsafe behavior risk indicators of coal miners. And appropriate safety education plans and control measures are selected according to the results of the trend prediction, so as to reduce the underground operation risks brought by the factors of coal miners themselves. The targeted education plans and control measures can also greatly save labor and material costs. Brief Description of the Drawings

[0054] Figure 1 Shows a flowchart of a method for predicting the trend of workers' unsafe behaviors based on dual-index clustering;

[0055] Figure 2 Shows a structural block diagram of a system for predicting the trend of workers' unsafe behaviors based on dual-index clustering. Detailed Embodiments

[0056] Now, the content of the present disclosure will be discussed with reference to several exemplary embodiments. It should be understood that these embodiments are discussed only to enable those of ordinary skill in the art to better understand and thus implement the content of the present disclosure, rather than implying any limitation to the scope of the present disclosure.

[0057] As used herein, the term "including" and its variants are to be construed as open-ended terms meaning "including but not limited to". The term "based on" is to be construed as "at least partially based on". The terms "one embodiment" and "an embodiment" are to be construed as "at least one embodiment". The term "another embodiment" is to be construed as "at least one other embodiment". The orientation or positional relationships indicated by the terms "upper", "lower", "left", "right", "front", "rear", "top", "bottom", "inner", "outer", "vertical", "horizontal", "lateral", "longitudinal", etc. are based on the orientation or positional relationships shown in the drawings. These terms are mainly used to better describe the present application and its embodiments, and are not used to limit that the indicated devices, elements or components must have a specific orientation, or be constructed and operated in a specific orientation. Moreover, in addition to being used to represent orientation or positional relationships, some of the above terms may also be used to represent other meanings. For example, the term "upper" may also be used to represent a certain attachment relationship or connection relationship in some cases. For those of ordinary skill in the art, the specific meanings of these terms in the present application can be understood according to specific circumstances. In addition, the terms "installed", "set", "provided with", "connected", "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral structure; it may be a mechanical connection or an electrical connection; it may be directly connected, or indirectly connected through an intermediate medium, or there may be internal communication between two devices, elements or components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances. In addition, the terms "first", "second", etc. are mainly used to distinguish different devices, elements or components (the specific types and structures may be the same or different), and are not used to indicate or imply the relative importance and quantity of the indicated devices, elements or components. Unless otherwise specified, the meaning of "a plurality" is two or more.

[0058] In a first aspect, the present embodiment discloses a method for predicting the trend of workers' unsafe behaviors based on dual-index clustering, as Figure 1 shown, including:

[0059] Stratified sampling is performed on all coal mine workers;

[0060] Based on the personal basic indicators, safety education indicators, work situation indicators and health physiological indicators of the sampled workers, basic risk indicators are constructed;

[0061] Based on the personal unsafe behavior events of the sampled workers, unsafe behavior risk indicators are constructed, and the personal unsafe behavior events are obtained based on the accident dimension, awareness dimension and overall dimension;

[0062] Perform kmeans clustering on individuals, teams or sections among all coal miners based on the above-mentioned basic risk indicators and unsafe behavior risk indicators;

[0063] Construct a Bayesian discriminant model, and perform cluster division on individuals, teams or sections among all coal miners based on the Bayesian discriminant model;

[0064] Predict the unsafe behaviors of individuals, teams or sections among all coal miners based on the results of the above-mentioned cluster division, and formulate corresponding preventive safety education plans and preventive control measures.

[0065] In this embodiment, a method for predicting the trend of workers' unsafe behaviors based on dual-index clustering is proposed. First, perform stratified sampling on all coal miners, and construct basic risk indicators based on the personal basic indicators, safety education indicators, work situation indicators and health physiological indicators of the sampled workers, and construct unsafe behavior risk indicators based on the personal unsafe behavior events of the sampled workers in terms of accident dimensions, awareness dimensions and overall dimensions. Secondly, perform kmeans clustering on individuals, teams or sections among all coal miners based on the basic risk indicators and unsafe behavior indicators. Thirdly, construct a Bayesian discriminant model, and divide individuals, teams or sections among all coal miners into the clustering results of kmeans clustering through the Bayesian discriminant model. Finally, predict the unsafe behaviors of individuals, teams or sections among all coal miners according to the division results of the Bayesian discriminant model, and formulate corresponding preventive safety education plans and preventive control measures.

[0066] Specifically, the stratified sampling is specifically divided by age groups, including four layers: under 30 years old, 30 - 40 years old (excluding 40), 40 - 50 years old (excluding 50), and 50 years old and above. Perform simple random sampling on the stratified results, and sample 20% of the workers from each layer as the research samples.

[0067] Specifically, the basic risk indicators are the basic information of individual coal miners, including information such as working years, education level, attendance, social relations and personal health conditions. The unsafe behavior risk indicators refer to the types and quantities of accidents caused, the number of conscious unsafe behaviors and the overall number of unsafe behaviors, etc.

[0068] Specifically, K-Means clustering is a widely used unsupervised learning algorithm for dividing a data set into K clusters, so that the data points within each cluster are as similar as possible to each other, while the data points in different clusters are as different as possible.

[0069] Specifically, during clustering, all sub - categories included in the basic risk indicators and unsafe behavior risk indicators are clustered simultaneously, and the final clustering results are divided into individual, team, and section - level ones. This enables the clustering results to be applicable to both basic risk indicators and unsafe behavior risk indicators. Meanwhile, for individuals or collectives at different levels, unsafe behavior portraits of individuals or collectives can be formed based on the clustering results.

[0070] Specifically, the Bayesian discriminant model (Bayesian Classifier) is a statistical classification method based on Bayes' theorem, aiming to predict the class to which a data point belongs according to known features. It calculates the posterior probability of each class and selects the class with the highest posterior probability as the prediction result.

[0071] Specifically, after obtaining the classification of workers through stratified sampling, the constructed Bayesian discriminant model is used to divide all coal miners into different clusters in the kmeans clustering results (if the clustering objects are teams or sections, then the teams or sections are divided into the corresponding clusters of the clustering results). Then, based on the changes in the current cluster and the cluster in the previous period, the future unsafe behaviors of workers, teams, or sections are predicted, and corresponding preventive safety education plans and preventive control measures are formulated according to the prediction results.

[0072] Specifically, through the above - mentioned method, based on the clustering of dual indicators of basic risk indicators and unsafe behavior risk indicators, the clustering results can be applicable to both indicators simultaneously. And through the Bayesian discriminant model, all individuals, teams, or sections are divided into clusters of the clustering results to predict the trend of unsafe behaviors. It can effectively capture the behavior changes of individuals, teams, or sections, thereby predicting their subsequent behaviors and making targeted safety education and preventive measures. This can greatly reduce the accident rate caused by workers during underground operations, and the targeted safety education and preventive measures can also greatly reduce the human and material costs.

[0073] In some embodiments, the specific individual basic indicators include years of work experience, education level, and professional skill level;

[0074] The safety education indicators include the situation of participating in safety education and the safety education assessment scores;

[0075] The work - related situation indicators include attendance, social exchange relationship I, and social exchange relationship O;

[0076] The health - related physical indicators include cardiovascular and cerebrovascular health, basic metabolic health, fatigue level, heart rate abnormality, and body temperature abnormality.

[0077] Table 1 Basic Risk Indicators

[0078]

[0079] In this embodiment, as shown in Table 1, the detailed classification of personal basic indicators, safety education indicators, work situation indicators, and health physiological indicators in the basic risk indicators is introduced.

[0080] Specifically, the personal basic indicators include working years, education level, and vocational skill level. By normalizing the working years of workers after sampling, a dimensionless index value is obtained. The normalization formula is:

[0081] ;

[0082] where is the value of the normalized working years index of worker i, is the original working years of worker i, takes 0, takes 37 (assuming coal miners retire at 55 years old and the maximum working years is 37 years). The simplified formula is:

[0083] ;

[0084] The quantification method of education level is:

[0085] ;

[0086] The quantification method of vocational skill level is:

[0087] .

[0088] Specifically, the safety education indicators include the participation in safety education and the safety education assessment score. The quantification method of the participation in safety education is:

[0089] ;

[0090] The quantification method of the safety education assessment score is:

[0091] ;

[0092] where is the quantification result of the safety education assessment score indicator, M is the number of assessment participations, is the score of worker i in the mth assessment participation, is the total score of the mth assessment.

[0093] Specifically, the work situation indicators include attendance, social exchange relationship I, and social exchange relationship O. The attendance situation indicator does not consider the factor of taking leave, and only counts the supposed attendance and actual attendance. The quantification method of the attendance situation indicator is:

[0094] ;

[0095] The data of social exchange relationships comes from the mutual evaluation form of the work conditions of workers in the team / section. The mutual evaluation form is for each worker in the underground team / section of the coal mine to evaluate the work conditions of other workers, and the evaluation results have three options: excellent, medium, and poor. For example, is the evaluation of worker j by worker i. More specifically, the social exchange relationship I represents the evaluation results of all other workers on a certain worker i, highlighting the work conditions of a certain worker in the eyes of other workers. To highlight the risk tendency, excellent and medium are regarded as one category here, and the quantification method is the proportion of poor evaluations. For example, the social exchange relationship I of worker i is:

[0096] ;

[0097] The social exchange relationship O represents the evaluation results of a certain worker i on all other workers. Here, excellent and medium are also regarded as one category, and the quantification method is the proportion of poor evaluations. For example, the social exchange relationship O of worker i is:

[0098] .

[0099] Specifically, the healthy physiological indicators include cardiovascular and cerebrovascular health, basal metabolic health, fatigue level, heart rate abnormality, and body temperature abnormality. More specifically, the cardiovascular and cerebrovascular health focuses on whether the worker has cardiovascular and cerebrovascular diseases, including hypertension, hyperlipidemia, atherosclerosis, etc., and its quantification method is:

[0100] ;

[0101] The basal metabolic health focuses on whether the worker has basal metabolic diseases, including diabetes, thyroid function abnormality, hyperuricemia, liver function disorder, and its quantification method is:

[0102] ;

[0103] The data of the fatigue level indicator comes from the fatigue warning events of the intelligent helmets worn by coal miners. Take the number x of fatigue warning events in the recent month, and the quantification method of the fatigue level is:

[0104] ;

[0105] The data of the heart rate abnormality indicator comes from the heart rate abnormality warning events of the intelligent bracelets worn by coal miners. Take the number x of heart rate abnormality warning events in the recent month, and the quantification method of the heart rate abnormality is:

[0106] ;

[0107] The body temperature abnormality index data comes from the body temperature abnormality warning events of the smart bracelets worn by coal miners. Take the number x of body temperature abnormality warning events in the recent month, and the quantization method of body temperature abnormality is as follows:

[0108] .

[0109] In some embodiments, the accident-causing dimensions include: operation errors / ignoring safety / ignoring warnings, causing safety device failures, using unsafe equipment, using hands instead of tools, improper storage of objects, venturing into dangerous places, sitting on unsafe positions, working / staying under the lifting arm, refueling / repairing / checking when the machine is running, having distracting behaviors, not correctly using personal protective equipment / appliances, unsafe dressing, and mishandling dangerous goods such as flammable / explosive substances;

[0110] The awareness dimension includes conscious unsafe behaviors and unconscious unsafe behaviors;

[0111] The overall dimension is the number of unsafe behaviors.

[0112] In this embodiment, as shown in Table 2, the detailed classification of personal basic indicators, safety education indicators, work situation indicators, and health physiological indicators in the basic risk indicators is introduced.

[0113] Specifically, as shown in Table 2, the quantization methods of the R01-R15 indicators are the same, that is, the ratio of the number of unsafe behaviors corresponding to a certain indicator of an employee in the recent month to the total number of his / her unsafe behaviors:

[0114] ;

[0115] Among them, is the quantization value of indicator m, is the number of unsafe behaviors of worker i in indicator m in the recent month, is the total number of unsafe behaviors of worker i in the recent month.

[0116] Specifically, the quantization method of the R16 indicator is to take the number x of unsafe behaviors of a certain worker in the recent month and calculate:

[0117] .

[0118] Table Unsafe Behavior Risk Indicators

[0119]

[0120] In some embodiments, the kmeans clustering of the basic risk indicators and unsafe behavior risk indicators of individuals, teams, or teams in the full set of coal miners includes:

[0121] Quantify the basic risk indicators and unsafe behavior risk indicators of individuals, teams or sections among all coal miners;

[0122] Initialize the value of k to 2, and cluster the quantified basic risk indicators and unsafe behavior risk indicators;

[0123] After completing the clustering with the initialized value of k, increment the value of k from the initial value, and cluster the quantified basic risk indicators and unsafe behavior risk indicators again;

[0124] After each clustering, calculate the sum of squared errors of the data within each cluster;

[0125] Draw a line chart of k based on the sum of squared errors of the data for each clustering;

[0126] Determine the optimal value of k using the elbow method according to the line chart;

[0127] Select the clustering result corresponding to the optimal value of k as the clustering result of kmeans clustering;

[0128] Where k is the number of clusters.

[0129] In some embodiments, the calculation of the sum of squared errors of the data within each cluster has the formula:

[0130] ;

[0131] Where SEE is the sum of squared errors, m is the total number of samples, is the i-th sample, is the center point of the j-th cluster, is the indicator function, is the sample and the cluster center The square of the Euclidean distance between them, and when belongs to then has a value of 1, otherwise it is 0.

[0132] In this embodiment, the specific process of kmeans clustering is introduced. First, quantify the basic risk indicators and unsafe behavior risk indicators of individuals, teams or sections among all coal miners. Then initialize the value of k to 2 and cluster the quantified indicators simultaneously. Finally, increment the value of k and perform clustering separately, and use the elbow method to select the optimal value of k, and obtain the clustering result corresponding to the optimal value of k as the final clustering result.

[0133] Specifically, initialize the value of k to 2, and randomly select k centroids in the sample space. Assign each sample (coal miner) to the nearest cluster center (centroid). At this time, the distance is calculated using the Euclidean distance. For the sample and the cluster center , the distance calculation formula is:

[0134] ;

[0135] wherein, is the m-th feature (index) value of the cluster center the m-th feature value of

[0136] Recalculate the position of each cluster center. The new cluster center is the mean of all data points assigned to that cluster. For cluster j, the new cluster center is calculated by the formula:

[0137] ;

[0138] wherein, is the set of all data points assigned to cluster j, is the number of data points in. Repeat this step until the clusters of the sample points no longer change, and complete the calculation of clustering.

[0139] Specifically, when using the elbow method to select the optimal k value, starting from the initial k value, after each clustering, calculate the sum of squared errors within each cluster SSE, and the formula is:

[0140] ;

[0141] wherein, m is the total number of samples, k is the number of clusters, is the i-th sample, is the center point of the j-th cluster, is the indicator function, is the sample and the squared Euclidean distance between the cluster center , and when belongs to , the value of is 1, otherwise it is 0.

[0142] Then, according to the SEE results of each clustering, draw a line chart of k, and select the k value corresponding to the elbow position as the optimal k value, and select the clustering result corresponding to the optimal k value as the final result.

[0143] Specifically, the elbow method is a heuristic method for determining the optimal number of clusters k in K-means clustering. It helps to select the optimal number of clusters by analyzing the change trend of the sum of squares within clusters under different k values.

[0144] In some embodiments, constructing the Bayesian discriminant model and performing cluster division on individuals, teams or sections among all coal miners based on the Bayesian discriminant model includes:

[0145] Calculating the prior probability of each cluster based on the clustering results;

[0146] Estimating the conditional probability density function of each cluster based on the clustering results;

[0147] According to Bayes' law, calculating the posterior probability that an individual, team or section among all coal miners belongs to each cluster in the clustering results, and its formula is:

[0148] ;

[0149] where is the posterior probability, is the prior probability, is the conditional probability density function, is the marginal probability of the sample x;

[0150] Assigning the individuals, teams or sections among all coal miners to the cluster with the maximum posterior probability in the clustering results.

[0151] In this embodiment, it is introduced how to construct the Bayesian discriminant model and use the model to assign individuals, teams or sections among all coal miners to the corresponding clusters. First, for each cluster in the clustering results, calculate the prior probability, that is, the proportion belonging to this cluster among all samples. Second, for each cluster, estimate its conditional probability density, assuming that each cluster follows a normal distribution at this time. Finally, calculate the posterior probability that an individual, team or section among all coal miners belongs to each cluster according to Bayes' theorem, and divide it into the cluster with the maximum posterior probability.

[0152] Specifically, the Bayesian discriminant model (Bayesian Classifier) is a statistical classification method based on Bayes' theorem. It determines which category a new data point is most likely to belong to by calculating the posterior probability of each category. This method not only considers the distribution of the data but also utilizes prior knowledge, so it can provide relatively accurate classification results in many cases.

[0153] Specifically, by constructing the Bayesian discriminant model, the corresponding individuals, teams or sections are divided into the clusters in the clustering results to which they belong, which is convenient for obtaining the unsafe behavior portraits of all coal miners subsequently.

[0154] In some embodiments, taking the mean value of the unsafe behavior risk indicators in the cluster to which an individual, team or section among all coal miners belongs as the first unsafe behavior portrait;

[0155] Based on the first unsafe behavior profile, formulate safety education plans and control measures for the corresponding individuals, teams or sections.

[0156] In this embodiment, after classifying individuals, teams or sections among all coal miners into their respective clusters through the Bayesian discriminant model, the mean value of the unsafe behavior risk indicators in the accumulation of individuals, teams or sections is used as the first unsafe behavior profile. Through the first unsafe behavior profile, safety education plans and control measures for the corresponding individuals, teams or sections can be formulated for the occurred unsafe behaviors.

[0157] In some embodiments, predicting the unsafe behaviors of individuals, teams or sections among all coal miners based on the result of the cluster division, and formulating corresponding preventive safety education plans and preventive control measures, including:

[0158] Input the basic risk indicators and unsafe behavior risk indicators of individuals, teams or sections at the T-1 period among all coal miners into the Bayesian discriminant model to obtain the first cluster of individuals, teams or sections at the T-1 period among all coal miners;

[0159] Input the basic risk indicators of individuals, teams or sections at the T period and the unsafe behavior risk indicators at the T-1 period among all coal miners into the Bayesian discriminant model to obtain the second cluster of individuals, teams or sections at the T period among all coal miners;

[0160] In response to the first cluster and the second cluster not being the same cluster, use the mean value of the unsafe behavior risk indicators of individuals, teams or sections at the T period in the second cluster as the second unsafe behavior profile;

[0161] Formulate corresponding preventive safety education plans and preventive control measures based on the second unsafe behavior profile.

[0162] In this embodiment, it is introduced how to predict the trend of unsafe behaviors of coal miners through the Bayesian discriminant model. First, bring the basic risk indicators and unsafe behavior risk indicators of individuals, teams or sections at the T-1 period (i.e., the previous period) into the Bayesian discriminant model to obtain the first cluster they belong to at the T-1 period. Then, bring the basic risk indicators of individuals, teams or sections at the T period (i.e., the current period) and the unsafe behavior risk indicators at the T-1 period into the Bayesian discriminant model to obtain the second cluster they belong to at the T period. Finally, compare the first cluster and the second cluster. When the first cluster and the second cluster are not the same cluster, use the mean value of the unsafe behavior risk indicators at the T period in the second cluster as the second unsafe behavior profile of individuals, teams or sections, and formulate corresponding preventive safety education plans and preventive control measures according to the second unsafe behavior profile.

[0163] Specifically, when the first type of cluster and the second type of cluster are not of the same type, it indicates that for this individual, team, or work area, since their basic risk indicators have changed, their unsafe behavior risk indicators are also about to change. Therefore, the average value of the unsafe behavior indicators in the T period is taken as the second unsafe behavior portrait, which is used as the unsafe behavior portrait for future moments, and preventive safety education plans and preventive control measures are formulated accordingly.

[0164] Specifically, by using the above method to complete the prediction of the unsafe behavior trend of individuals, teams, or work areas in a coal mine, targeted education plans and control measures can be formulated based on the prediction results, thereby reducing the accident rate during underground coal mine operations. At the same time, the targeted education plans and control measures can also save a large amount of human and material resources.

[0165] In some embodiments, in response to the first type of cluster and the second type of cluster being of the same type, the individuals, teams, or work areas among the full amount of coal miners continue to use the safety education plan and control measures.

[0166] In this embodiment, if the first type of cluster and the second type of cluster are of the same type, it means that the unsafe behavior portrait of the individual, team, or work area will not change significantly at the T moment, so the existing safety education plan and control measures can be continued to be used.

[0167] Specifically, by continuing to use the existing safety education plan and control measures, a large amount of human and material resources can be saved while ensuring a relatively low accident rate.

[0168] In a second aspect, the present invention proposes a prediction system for the unsafe behavior trend of workers based on dual-index clustering, including the following modules:

[0169] Sampling module: used to implement stratified sampling of the full amount of coal miners;

[0170] Quantification module: used to implement the construction of basic risk indicators based on the personal basic indicators, safety education indicators, work situation indicators, and health and physiological indicators of the sampled workers, and the construction of unsafe behavior risk indicators based on the personal unsafe behavior events of the sampled workers. The personal unsafe behavior events are obtained based on the accident dimension, awareness dimension, and overall dimension;

[0171] Clustering module: used to implement kmeans clustering of individuals, teams, or work areas among the full amount of coal miners based on the basic risk indicators and unsafe behavior risk indicators;

[0172] Discrimination module: It is used to implement the construction of a Bayesian discrimination model, and based on the Bayesian discrimination model, classify individuals, teams or sections among all coal miners into clusters, predict the unsafe behaviors of individuals, teams or sections among all coal miners based on the results of the cluster classification, and formulate corresponding preventive safety education plans and preventive control measures.

[0173] In summary, through the technical solution of the present invention, the trend of unsafe behaviors of individuals, teams or sections can be predicted based on the basic risk indicators and unsafe behavior risk indicators of coal miners. And according to the results of the trend prediction, appropriate safety education plans and control measures are selected, so as to reduce the underground operation risks brought by the factors of coal miners themselves. The targeted education plans and control measures can also greatly save labor and material costs.

[0174] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to include all changes falling within the meaning and scope of the equivalent elements of the claims in the present invention.

[0175] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other implementation manners that can be understood by those skilled in the art.

Claims

1. A method for predicting the trend of workers' unsafe behaviors based on double-index clustering, characterized in that Including: Stratified sampling is carried out on all coal miners; Based on the personal basic indicators, safety education indicators, work situation indicators and health physiological indicators of the sampled workers, basic risk indicators are constructed; Based on the personal unsafe behavior events of the sampled workers, unsafe behavior risk indicators are constructed, and the personal unsafe behavior events are obtained based on the accident dimension, awareness dimension and overall dimension; Based on the basic risk indicators and unsafe behavior risk indicators, kmeans clustering is carried out on individuals, teams or sections in all coal miners; A Bayesian discriminant model is constructed, and based on the Bayesian discriminant model, class cluster division is carried out on individuals, teams or sections in all coal miners; Based on the results of the class cluster division, the unsafe behaviors of individuals, teams or sections in all coal miners are predicted, and corresponding preventive safety education plans and preventive control measures are formulated.

2. The method for predicting the trend of workers' unsafe behaviors based on dual-index clustering according to claim 1, wherein, The personal basic indicators specifically include working years, education level and professional skill level; The safety education indicators include the participation in safety education and the safety education assessment scores; The work situation indicators include attendance, social exchange relationship I and social exchange relationship O; The health physiological indicators include cardiovascular and cerebrovascular health, basic metabolic health, fatigue degree, heart rate abnormality and body temperature abnormality.

3. The method for predicting the trend of workers' unsafe behaviors based on dual-index clustering according to claim 1, characterized in that, The accident dimension includes: operation error / ignoring safety / ignoring warning, causing safety device failure, using unsafe equipment, using hands instead of tools, improper object storage, taking risks to enter dangerous places, climbing and sitting in unsafe positions, working / staying under the lifting arm, refueling / repairing / checking when the machine is running, having distracting behaviors, not correctly using personal protective equipment / utensils, unsafe dressing and handling flammable / explosive and other dangerous goods incorrectly; The awareness dimension includes conscious unsafe behaviors and unconscious unsafe behaviors; The overall dimension is the number of unsafe behaviors.

4. The method for predicting the trend of workers' unsafe behaviors based on double-index clustering according to claim 1, wherein The kmeans clustering of the basic risk indicators and unsafe behavior risk indicators for individuals, teams or sections in all coal miners includes: Quantify the basic risk indicators and unsafe behavior risk indicators for individuals, teams or sections in all coal miners; Initialize the value of k to 2, and cluster the quantified basic risk indicators and unsafe behavior risk indicators; After completing the clustering with the initial value of k, increase the value of k from the initial value, and cluster the quantified basic risk indicators and unsafe behavior risk indicators again; After each clustering ends, calculate the sum of squared errors of the data within each cluster; Draw a line chart of k based on the sum of squared errors of the data for each clustering; Use the elbow method to determine the optimal value of k according to the line chart; Select the clustering result corresponding to the optimal value of k as the clustering result of kmeans clustering; Where k is the number of clusters.

5. The method for predicting the trend of workers' unsafe behaviors based on double-index clustering according to claim 4, wherein The formula for calculating the sum of squared errors of the data within each cluster is: ； where SSE is the sum of squared errors of data, m is the total number of samples, is the i-th sample, is the center point of the j-th cluster, is the indicator function, is the sample and the Euclidean distance squared between the sample and the cluster center and when belongs to the value of is 1, otherwise it is 0.

6. The method for predicting the trend of workers' unsafe behaviors based on double-index clustering according to claim 4, wherein The construction of the Bayesian discriminant model and the class cluster division of individuals, teams or sections in all coal miners based on the Bayesian discriminant model include: Based on each class cluster of the clustering results, calculate the prior probability of each class cluster; Based on each class cluster of the clustering results, estimate the conditional probability density function of each class cluster; According to Bayes' law, calculate the posterior probability that an individual, a team, or a section in the entire coal mine workers belongs to each cluster in the clustering result. The formula is as follows: ； wherein, is the posterior probability, is the prior probability, is the conditional probability density function, is the marginal probability of the sample x; Assign the individuals, teams, or sections in the entire coal mine workers to the cluster with the highest posterior probability in the clustering result.

7. The method for predicting the trend of workers' unsafe behaviors based on double-index clustering according to claim 6, wherein Take the mean of the unsafe behavior risk indicators in the cluster to which the individuals, teams, or sections in the entire coal mine workers belong as the first unsafe behavior portrait. Based on the first unsafe behavior portrait, formulate safety education plans and control measures for the corresponding individuals, teams, or sections.

8. The method for predicting the trend of workers' unsafe behaviors based on double-index clustering according to claim 6, wherein, Predict the unsafe behaviors of individuals, teams, or sections among the entire coal mine workers based on the result of the cluster division, and formulate corresponding preventive safety education plans and preventive control measures, including: Input the basic risk indicators and unsafe behavior risk indicators of individuals, teams, or sections at the T-1 period among the entire coal mine workers into the Bayesian discriminant model to obtain the first cluster of individuals, teams, or sections at the T-1 period among the entire coal mine workers. Input the basic risk indicators at the T period and the unsafe behavior risk indicators at the T-1 period of individuals, teams, or sections among the entire coal mine workers into the Bayesian discriminant model to obtain the second cluster of individuals, teams, or sections at the T period among the entire coal mine workers. In response to the first cluster and the second cluster not being the same cluster, take the mean of the unsafe behavior risk indicators of individuals, teams, or sections at the T period in the second cluster as the second unsafe behavior portrait. Formulate corresponding preventive safety education plans and preventive control measures based on the second unsafe behavior portrait.

9. The method for predicting the trend of workers' unsafe behaviors based on dual-index clustering according to claim 8, wherein, In response to the first cluster and the second cluster being the same cluster, the individuals, teams, or sections among the entire coal mine workers continue to use the safety education plan and control measures.

10. A prediction system for the trend of workers' unsafe behaviors based on dual-index clustering, characterized in that, It includes the following modules: Sampling module: used to implement stratified sampling of the entire coal mine workers. Quantification module: used to implement the construction of basic risk indicators based on the personal basic indicators, safety education indicators, work situation indicators, and health physiological indicators of the sampled workers, and the construction of unsafe behavior risk indicators based on the personal unsafe behavior events of the sampled workers. The personal unsafe behavior events are obtained based on the accident dimension, awareness dimension, and overall dimension. Clustering module: used to implement kmeans clustering of individuals, teams, or sections among the entire coal mine workers based on the basic risk indicators and unsafe behavior risk indicators. Discrimination module: used to implement the construction of a Bayesian discriminant model, and based on the Bayesian discriminant model, conduct cluster division on individuals, teams, or sections among the entire coal mine workers, predict the unsafe behaviors of individuals, teams, or sections among the entire coal mine workers based on the result of the cluster division, and formulate corresponding preventive safety education plans and preventive control measures.

Images