A warehouse environment risk monitoring method and system

Abstract

The present application relates to warehouse environment risk monitoring technical field, specifically to a kind of warehouse environment risk monitoring method and system, comprising the following steps: obtaining the scheduling table of target warehouse inside forklift driver or inspector and current operation time, then collecting the task load factor of current task type and the temperature and noise level of operation area, and the scenario fatigue risk value is calculated.The present application is by analyzing the scheduling data of forklift driver and inspector and real-time task load, environmental stress indicators, quantitatively assesses the comprehensive influence of personnel operation state and environmental condition on fatigue risk, makes risk evaluation basis more comprehensive and specific, avoids the singularity of risk judgment;Real-time acquisition warehouse layout and cargo distribution information, and combine the readings of warehouse fire smoke, toxic gas sensor, associate multiple source risk elements to specific warehouse grid unit, realize the more fine spatial granularity, the fusion of more comprehensive risk information.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of warehouse environment risk monitoring, and in particular to a warehouse environment risk monitoring method and system. BACKGROUND

[0002] The technical field of warehouse environment risk monitoring mainly involves technical means for comprehensive monitoring, analysis and evaluation of various environmental risk sources in the warehouse scene.

[0003] The prior art only focuses on a single type of environmental risk indicator, such as fire or gas leakage, when monitoring the warehouse environment risk, which can easily result in incomplete risk assessment coverage and blind spots in the monitoring results. In addition, the prior art ignores the dynamic changes in personnel and equipment positions and state information within the warehouse, resulting in insufficient timeliness of risk response decisions, difficulty in matching emergency resource scheduling based on real-time conditions, and easy response delays for inspection personnel or equipment, low resource utilization efficiency, and potential impact on disposal effectiveness, even causing personnel safety accidents or economic losses. Therefore, improvements are needed. SUMMARY

[0004] The purpose of the present application is to solve the problems existing in the prior art and to provide a warehouse environment risk monitoring method and system.

[0005] In order to achieve the above-mentioned purpose, the present application adopts the following technical solution: a warehouse environment risk monitoring method, comprising the following steps:

[0006] Obtaining the scheduling table and current work time of the forklift driver or the inspector in the target warehouse, collecting the task load factor of the current task type and the temperature and noise level of the work area, and calculating the scenario fatigue risk value;

[0007] Calling the warehouse layout map and cargo distribution information of the warehouse, integrating the real-time readings of the fire smoke sensor and the toxic gas sensor, associating and assigning the scenario fatigue risk value and the sensor readings on the corresponding grid elements of the warehouse layout map, generating a partition risk element set, labeling a high-risk area based on the partition risk element set, and establishing a fusion risk situation map;

[0008] Based on the fusion risk situation map, the positions and working states of the inspection personnel, forklifts and fire-fighting robots positioned by the Internet of Things are obtained in real time, the positions of the fixed fire hydrants, fire extinguishers and emergency exits are retrieved, a resource availability state list is established, and an emergency resource response capability matrix is calculated based on the resource availability state list;

[0009] The high-risk areas in the fusion risk situation map are prioritized, and resources are selected from the emergency resource response capability matrix for matching, to generate an emergency resource scheduling instruction set.

[0010] Preferably, the obtaining step of the situational fatigue risk value is:

[0011] Analyzing the shift schedule of the forklift driver or the inspector and the current operation time period, extracting the start and end time of the continuous operation period of the personnel, and calculating the total length of time without alternating operation in the continuous period to obtain the continuous operation time value of the personnel;

[0012] According to the current task number, the corresponding task type is matched and the task load factor uniquely associated with the task type is extracted, the temperature sensor and the noise sensor are called synchronously to obtain the temperature value and the noise level value of the current operation area, each environment value is standardized by comparing with the corresponding comfort threshold and the danger threshold, and each environment factor is weighted by the risk enhancement coefficient corresponding to the environment factor to form a comprehensive environmental stress parameter;

[0013] According to the continuous operation time value of the personnel, the reference operation time value, the task load factor and the comprehensive environmental stress parameter, the situational fatigue risk value is calculated.

[0014] Preferably, the obtaining step of the partition risk element set is:

[0015] The warehouse layout map and the cargo distribution information of the warehouse are called, the layout map is divided into grid units of fixed size, the cargo type, the cargo storage quantity and the storage location in the cargo distribution information are analyzed and mapped to the corresponding grid unit one by one, and a warehouse grid cargo distribution mapping table is generated;

[0016] Based on the warehouse grid cargo distribution mapping table, the real-time readings of the fire smoke sensor and the toxic gas sensor are called, the installation position coordinates of each sensor are analyzed and associated with the corresponding grid unit, and the smoke concentration value and the toxic gas concentration value are assigned to each grid unit to generate a warehouse grid environment sensor reading distribution table;

[0017] Based on the warehouse grid environment sensor reading distribution table and the situational fatigue risk value, the smoke concentration value, the toxic gas concentration value and the situational fatigue risk value are assigned to each grid unit according to the warehouse grid unit coordinates, and a partition risk element set is generated.

[0018] Preferably, the obtaining step of the fusion risk situation map is:

[0019] Based on the set of partition risk elements, safety thresholds of smoke concentration value, toxic gas concentration value and situational fatigue risk value are set, and the smoke concentration value, toxic gas concentration value and situational fatigue risk value in each grid cell are compared with the corresponding safety thresholds one by one. If any value exceeds the corresponding safety threshold, the grid cell is marked as a high-risk area. According to the risk marking result of the grid cell, the risk area is marked on the warehouse layout map by color classification, and the fusion risk situation map is established.

[0020] Preferably, the resource availability state list acquisition step is:

[0021] The coordinates of all high-risk area grid cells marked in the fusion risk situation map are parsed, the positioning coordinates and operation state identifiers of the inspection personnel, forklifts and fire-fighting robots are obtained, and the positioning information of the fixed fire hydrants, fire extinguishers and emergency exits in the warehouse is called to form a resource availability state list classified by resource type.

[0022] Preferably, the emergency resource response capability matrix acquisition step is:

[0023] Based on the resource availability state list, a comprehensive response capability score value is calculated.

[0024] Based on the comprehensive response capability score value, a comprehensive response capability score value table is constructed according to the two-dimensional combination structure of resource items and high-risk areas, the emergency resource items are represented by rows, and the high-risk areas are represented by lists, and an emergency resource response capability matrix is generated.

[0025] Preferably, the emergency resource scheduling instruction set acquisition step is:

[0026] All grid cells marked as high-risk areas in the fusion risk situation map are called, and the situational fatigue risk value, smoke concentration value and toxic gas concentration value in each grid cell are parsed one by one. The situational fatigue risk value, smoke concentration value and toxic gas concentration value are compared with the corresponding safety risk thresholds one by one, the number of risk values exceeding the safety risk threshold in each grid cell is calculated, and the high-risk area priority ranking table is formed according to the descending order of the number of risk values exceeding the limit from more to less.

[0027] Based on the sorting results of each grid cell in the high-risk area priority ranking table, the comprehensive response capability score values of all emergency resource items corresponding to each grid cell in the emergency resource response capability matrix are called, the numerical values of all comprehensive response capability score values in each grid cell are compared one by one, and the emergency resource item with the largest numerical value is selected as the response resource of the corresponding grid cell to form the high-risk area emergency resource matching relationship table.

[0028] Preferably, the obtaining step of the emergency resource scheduling instruction set further comprises: based on the high-risk area emergency resource matching relationship table, extracting the corresponding response resource type, the current position coordinates of the response resource and the grid cell coordinates of the target high-risk area for each grid cell one by one, generating resource scheduling instructions one by one according to the information structure of emergency resource type, departure location and target location, and forming an emergency resource scheduling instruction set.

[0029] The application also provides a warehouse environment risk monitoring system, comprising:

[0030] A scenario risk assessment module obtains the work schedule and current work time of a forklift driver or a patrol in a target warehouse, and collects the task load factor of the current task type and the temperature and noise level of the work area, and calculates a scenario fatigue risk value.

[0031] A fusion risk modeling module calls the warehouse layout map and cargo distribution information of the warehouse, integrates the real-time readings of the fire smoke sensor and the toxic gas sensor, associates and assigns the scenario fatigue risk value and the sensor readings on the corresponding grid cells of the warehouse layout map, generates a partition risk element set, labels a high-risk area according to the partition risk element set, and establishes a fusion risk situation map.

[0032] An emergency resource assessment module obtains the position and working state of the Internet of Things positioning patrol personnel, forklifts and fire-fighting robots in real time based on the fusion risk situation map, calls the fixed fire hydrant, fire extinguisher and emergency exit position information, establishes a resource availability state list, and calculates an emergency resource response capability matrix based on the resource availability state list.

[0033] A resource scheduling instruction module prioritizes the high-risk areas in the fusion risk situation map, selects resources from the emergency resource response capability matrix for matching, and generates an emergency resource scheduling instruction set.

[0034] Compared with the prior art, the application has the following advantages and positive effects:

[0035] This invention quantitatively assesses the combined impact of personnel working conditions and environmental conditions on fatigue risk by analyzing the shift scheduling data of forklift drivers and inspectors, as well as real-time task load and environmental pressure indicators. This makes risk assessment more comprehensive and specific, avoiding the simplistic nature of risk judgment. It acquires real-time warehouse layout maps and cargo distribution information, and combines this with readings from fire smoke and toxic gas sensors within the warehouse to link multi-source risk factors to specific warehouse grid units. This achieves finer spatial granularity and more comprehensive risk information integration, facilitating precise risk area definition. Furthermore, based on risk area division and real-time positioning, it acquires the location and status information of emergency resources and calculates the comprehensive response capabilities of different resources for specific risk areas. By integrating and evaluating multiple factors such as the status factors of emergency resources, risk handling effectiveness, and path response time, it improves the matching degree between resources and risks, ensuring the accuracy and timeliness of emergency response resource allocation, and enhancing the overall efficiency and safety of warehouse environmental risk monitoring and emergency response. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0038] Please see Figure 1 This invention provides a technical solution, a method for monitoring warehouse environmental risks, comprising the following steps:

[0039] Obtain the shift schedule and current working time of forklift drivers or inspectors in the target warehouse, then collect the task load factor of the current task type and the temperature and noise level of the work area, and calculate the scenario fatigue risk value.

[0040] The warehouse layout map and cargo distribution information are called, and the real-time readings of fire smoke sensors and toxic gas sensors are integrated. The scenario fatigue risk value and sensor readings are associated and assigned on the corresponding grid cells of the warehouse layout map to generate a set of risk elements for each zone. Based on the set of risk elements for each zone, high-risk areas are marked and a fusion risk situation map is established.

[0041] Based on the fused risk situation map, the location and working status of inspection personnel, forklifts, and fire-fighting robots located by the Internet of Things are obtained in real time, and the location information of fixed fire hydrants, fire extinguishers, and emergency exits is retrieved to establish a resource availability status list. Based on the resource availability status list, the emergency resource response capability matrix is ​​calculated and obtained.

[0042] The high-risk areas in the fusion risk situation map are prioritized, and resources are selected from the emergency resource response capability matrix for matching, to generate a set of emergency resource scheduling instructions.

[0043] The scenario fatigue risk value is obtained by:

[0044] The scheduling table of the forklift driver or the inspector is analyzed, and the start and end times of the continuous work period of the personnel are extracted. The total length of the continuous period without alternating operation is calculated, and the personnel continuous operation time value is obtained.

[0045] According to the current task number, the corresponding task type is matched and the task load factor uniquely associated with the task type is extracted. The temperature sensor and noise sensor are called synchronously to obtain the temperature value and noise level value of the current operation area. Each environmental value is standardized by comparing it with the corresponding comfort threshold and danger threshold. Then, each environmental factor is weighted by the risk enhancement coefficient corresponding to it to form a comprehensive environmental stress parameter.

[0046] According to the personnel continuous operation time value, the reference operation time value, the task load factor and the comprehensive environmental stress parameter, the scenario fatigue risk value is calculated, and the calculation formula is:

[0047] ;

[0048] Among them, represents the scenario fatigue risk value of the personnel, represents the continuous operation time value of the personnel before the current task, represents the preset reference operation time value, represents the basic fatigue aggravation index, represents the task load adjustment coefficient, represents the task load factor corresponding to the task type of the personnel, represents the environmental factor collection value of the first environmental factor, represents the comfort threshold of the first environmental factor, represents the danger threshold of the first environmental factor, represents the risk enhancement coefficient of the first environmental factor, represents the total number of environmental factors currently participating in the calculation (such as 2 for temperature and noise).

[0049] ​Specifically, according to the shift schedule of the forklift driver or the inspector and the current operation time period, the system first accesses and analyzes the electronic shift schedule data, which is stored in a standard data format, and each record contains an employee number, a date, a planned start time, a planned end time, a task code, and an operation area identifier. The system obtains the current system timestamp accurate to the second through the network time protocol to locate the current working state of the personnel. To determine the starting point of continuous operation, the system first queries the operation item covering the current time point in the shift schedule according to the employee number and the current timestamp, and extracts the task code. For example, it is found that the forklift driver with the employee number W07 is currently performing the "high-bay picking" operation with the task code FL-P01 at 10:45. The system then establishes a "non-alternating operation" judgment standard, which defines whether the task type has changed and whether the rest time has interrupted the continuity. The change of the task type is determined by comparing whether the task codes of the two operation items before and after belong to the same preset task category. The task category table classifies specific task codes such as FL-P01 and FL-M03 into the "forklift operation" category, and classifies IN-S01 into the "manual inspection" category. The interruption threshold of the rest time is set to 30 minutes, which is based on the regulations on break time and the internal management rules of the enterprise. An interval below this threshold is considered a brief pause that does not interrupt the calculation of continuous operation. The system starts from the current operation item and queries the shift records of the employee on the same day in reverse order. For the immediately preceding operation item, the system calculates the time difference between its end time and the start time of the current operation item, and compares the task categories to which the task codes of the two items belong. For example, the previous task is FL-M03 "flat warehouse transfer" from 08:00 to 10:00, the task category is also "forklift operation", and the time interval between the end time 10:00 and the start time 10:15 of the current task is 15 minutes, which is less than the 30-minute interruption threshold. Therefore, the system determines that the operation is continuous, and traces the starting point of the continuous operation back to 08:00. The system repeats this reverse tracing process until it encounters a record with a different task category or a time interval exceeding 30 minutes. For example, the record before 08:00 is the off-duty record of the previous day, and the tracing stops. Finally, the system calculates the difference between the traced continuous operation start time (08:00) and the current system time (10:45) to obtain the personnel continuous operation time value in hours, which is 2.75 hours.

[0050] According to the current task number, the system first performs an accurate match in the "task load factor knowledge base", which stores the unique task load factor corresponding to each task type. The determination of the task load factor is based on a systematic analysis of all types of operations in the warehouse. A quantitative task complexity evaluation model is introduced to calculate the task load factor. The specific calculation method is as follows:

[0051]

[0052] wherein, , , respectively represent the scores of mental load, physical load and time pressure load given by the assessment team consisting of three senior safety engineers and two team leaders for each task, the score range is 1 to 10, , , are the corresponding weight coefficients, which are respectively set to 0.4, 0.4, 0.2, the weight is based on the statistical analysis of the warehouse historical safety events, indicating that physical and mental load are the main factors leading to operation errors, for example, for the "high rack picking" task, the three scores are 7, 8 and 6 respectively, and the task load factor is:

[0053] The calculation result is divided by the maximum value of the score 10 to be normalized, and 0.72 is obtained, in the extraction of the task load factor, the system sends a data request to the Internet of Things environment sensor array deployed in the current operation area according to the current operation area identifier, and obtains the temperature value and noise level value in real time, for example, the current area temperature is 29℃ and the noise is 86 decibels, next, the system standardizes the two environment values, the processing formula is: wherein is the environment factor collection value, is the comfort threshold, is the danger threshold, the comfort threshold and the danger threshold of temperature are set to 22℃ and 35℃ according to the international standardization organization ISO7730 standard, and the comfort threshold and the danger threshold of noise are set to 70 decibels and 95 decibels according to the occupational safety and health administration OSHA standard, therefore, the standardized values of temperature and noise are and respectively, finally, the weighted sum of the standardized environment values with their corresponding risk enhancement coefficients is obtained to obtain the comprehensive environmental stress parameter, the setting of each risk enhancement coefficient ( ) is derived from the regression analysis of nearly one hundred minor accident reports in the warehouse in the past two years, the analysis result shows that the contribution of noise to fatigue is slightly higher than that of temperature, therefore, the risk enhancement coefficient of noise is set to 0.4, and the risk enhancement coefficient of temperature is set to 0.3, finally, the comprehensive environmental stress parameter is calculated as:

[0054] .

[0055] formula: The beneficial part of the formula is that the power function part on the left side of the formula The non-linear growth law of fatigue with working time is described, and its growth rate will be dynamically adjusted according to the task load ) to achieve accurate characterization of the difference in fatigue accumulation speed of different intensity operations. The exponential function part on the right side of the formula , as an environmental stress multiplier, quantifies the amplification effect of adverse environments (such as high temperature, high noise) on basic fatigue, so that the risk assessment results can reflect the specific environmental situation of the operating personnel in real time. This design of coupling individual circadian rhythm, task characteristics and environmental factors in a unified framework makes risk assessment no longer a static or single-dimensional judgment, but can capture the complex risk evolution process under the interaction of multiple factors, thereby providing more accurate and timely decision-making basis for preventive intervention and resource scheduling.

[0056] represents the current continuous operation time value of the personnel before the task, in hours, which is calculated by analyzing the personnel schedule, determining the task continuity and interruption conditions in the previous step, and reflects the time accumulation basis of fatigue. According to the calculation, the continuous operation time value of the personnel is 2.75 hours, so .

[0057] represents the preset reference operation time value, in hours, which is a fatigue critical working time benchmark set based on industry safety specifications and enterprise long-term production data statistics. By analyzing the correlation between the forklift driver error rate and continuous working time in the warehouse over the past year, it was found that after continuous work for more than 4 hours, the error rate showed a significant exponential growth. Combined with the national regulation that special workers should be arranged for a mandatory rest every 4 hours, the reference value is set to 4 hours, so .

[0058] represents the basic fatigue aggravation index, which is a dimensionless constant that defines the basic rate of natural growth of fatigue with time even under the lightest load task. The determination of this value is by selecting the most relaxed "inventory" task in the warehouse as the benchmark, recording the brain wave theta band energy changes of multiple employees during the operation process in a controlled environment, and performing nonlinear regression fitting with continuous operation time to obtain the benchmark slope of the fatigue growth curve. Finally, its value is calibrated to 1.15, so .

[0059] represents the task load adjustment coefficient, which is a dimensionless constant used to adjust the task load factor Influence on fatigue aggravation degree. In order to calibrate the coefficient, two tasks of "picking in high shelf area" (high load) and "transferring in warehouse" (medium load) are selected, and the heart rate variability (HRV) reduction rate of the operator in the simulation environment is measured when performing the task, the correlation between the HRV change under different loads and the subjective score of fatigue is compared, and the coefficient value that best matches the actual physiological index change is determined as 0.6 through least square optimization, so .

[0060] The task load factor corresponding to the current task type of the personnel is a dimensionless value between 0 and 1, which is calculated by the task complexity evaluation model quantified in the foregoing steps. For the "picking in high shelf area" task, the task load factor is 0.72, so .

[0061] It is a comprehensive environmental stress parameter, which reflects the additional physiological and psychological stress caused by the current environment to the personnel as a whole. Its calculation process and result have been described in detail, and the calculation result is 0.4174, wherein, represents the collection value of the environmental factor, represents its comfort threshold, represents its danger threshold, represents its risk enhancement coefficient, represents the total number of environmental factors participating in the calculation. In this example, , the environmental factors are temperature and noise. The collection value of temperature is ℃, the comfort threshold is ℃, the danger threshold is ℃, and the risk enhancement coefficient is ; the collection value of noise is decibels, the comfort threshold is decibels, the danger threshold is decibels, and the risk enhancement coefficient is .

[0062] Calculation process:

[0063] According to the specific values of the above parameters, the formula is calculated as follows:

[0064] Calculate the exponential part of the comprehensive environmental stress parameter: ;

[0065] Calculate the exponential part of the time ratio term:

[0066]

[0067] ;

[0068] The power operation part of the time ratio term is calculated:

[0069] ;

[0070] Calculate the final scenario fatigue risk value : ;

[0071] The result shows that the current forklift driver's scenario fatigue risk value is 0.8020. This risk value is a normalized risk index, and the system internally sets three risk thresholds: low risk ( ), medium risk ( ), and high risk ( ). The current calculation result 0.8020 is in the low risk interval, indicating that the driver's fatigue state at the current time is within an acceptable range, but is close to the boundary of medium risk.

[0072] The acquisition step of the partitioned risk element set is:

[0073] Call the warehouse layout map and cargo distribution information of the warehouse, divide the layout map into fixed-size grid units, parse the cargo type, cargo storage quantity and storage location in the cargo distribution information, and map them to the corresponding grid units one by one to generate a warehouse grid cargo distribution mapping table;

[0074] Based on the warehouse grid cargo distribution mapping table, call the real-time readings of the fire smoke sensor and toxic gas sensor, parse the installation position coordinates of each sensor and associate the corresponding grid units, assign smoke concentration values and toxic gas concentration values to each grid unit, and generate a warehouse grid environment sensor reading distribution table;

[0075] Based on the warehouse grid environment sensor reading distribution table and the scenario fatigue risk value, correspond to each warehouse grid unit coordinate, merge and assign smoke concentration values, toxic gas concentration values and scenario fatigue risk values, and generate a partitioned risk element set.

[0076] Specifically, the system calls the warehouse layout map in digital computer-aided design (CAD) format and the real-time extracted goods distribution information from the warehouse management system (WMS) database. The system first calibrates the two-dimensional warehouse layout map in the coordinate system, taking the southwest corner of the warehouse as the coordinate origin (0, 0) to establish a Cartesian coordinate system covering the entire warehouse area. Then, the entire layout map area is cut according to a fixed size of 2 meters by 2 meters to generate a series of grid cells with unique coordinate identifiers (for example, GID-X01Y01 composed of row number and column number). This size is set by considering the standard pallet size (1.2 meters x 1 meter) and the minimum aisle width required for forklift operation (about 3.5 meters), ensuring that each shelf unit and main aisle can be accurately divided into independent grids. Next, the system starts the analysis program for goods distribution information. The program reads each data record exported from the WMS one by one, and each record contains material code, storage quantity, storage unit, and storage location code. The system queries a pre-configured "storage location-coordinate mapping table" according to the storage location code. This table is established by manual surveying and mapping during the initialization of the warehouse information system, and accurately defines the two-dimensional coordinates of the center point of each physical storage location (such as A-01-03-B). The system converts the parsed storage location coordinates (for example, x = 25.4, y = 66.8) into the grid cell coordinates (GID-X13Y34) where they are located. At the same time, through the association of the material code with the "material master data table", the detailed category information of the goods is extracted, such as "acetone (flammable liquid)" or "cotton textiles (combustible solid)". Finally, the system aggregates the parsed goods category and storage quantity information into a list form in the corresponding grid cell data structure. If a grid cell contains multiple storage locations, the data structure of the cell will contain the list of all goods. Finally, the warehouse grid goods distribution mapping table covering the entire warehouse is generated.

[0077] Based on the warehouse grid goods distribution mapping table generated in the previous stage, the system subscribes to the data streams of all deployed fire smoke sensors and toxic gas sensors through the Internet of Things (IoT) Message Queue Telemetry Transport (MQTT) protocol. These sensors report real-time readings at a frequency of every 5 seconds, with data packet contents including a unique sensor device ID, timestamp, and measurement values such as smoke concentration (unit: PPM) and carbon monoxide concentration (unit: PPM). After receiving the data, the system first queries the "Sensor Deployment Location List" based on the sensor device ID. This list is a static configuration file generated during the sensor network installation and debugging phase, recording the precise three-dimensional physical coordinates (x, y, z) of each sensor. The system extracts the two-dimensional plane coordinates (x, y) and associates them to a unique grid cell using the same grid division logic as in the previous step. Next, the system iterates through all grid cells and assigns smoke concentration values and toxic gas concentration values. For grid cells that directly contain one or more sensors, the environmental reading values are the average of all sensor readings of the same type within the cell. For grid cells that do not directly contain any sensors, the system uses the inverse distance weighted interpolation method (IDW) to estimate the values. The specific operation is as follows: taking the center point of the grid cell to be calculated as the reference, searching for all sensors of the same type within a radius of 10 meters, selecting the four closest sensors as the interpolation source, and then calculating the weighted average of these sensor readings according to the inverse square of the distance from the reference point as the weight, thus obtaining the estimated environmental readings of the grid cell. This 10-meter search radius is set based on empirical data of the average speed of air convection in the warehouse and the effective coverage range of the sensor. By performing the above assignment or interpolation calculation on all grid cells, a real-time updated warehouse grid environmental sensor reading distribution table is finally generated.

[0078] Based on the real-time updated warehouse grid environment sensor reading distribution table and the previously calculated scenario fatigue risk value for a specific person, the system first obtains the real-time coordinates of all workers in the warehouse through the worker positioning system (e.g., positioning tags based on ultra-wideband UWB technology), and converts these coordinates to corresponding grid cell IDs. Subsequently, the system performs a data merging process based on the warehouse grid environment sensor reading distribution table as the basic data structure, and processes each grid cell one by one. For any grid cell, the system first checks whether there is a worker located within it. If there is no worker, the scenario fatigue risk value of the grid cell is assigned as 0. If there is one or more workers located within the cell, the system extracts the scenario fatigue risk values of all these workers and selects the maximum value as the final scenario fatigue risk value of the grid cell. For example, in grid GID-X13Y34, there are two inspectors with scenario fatigue risk values of 0.65 and 0.81 respectively. The scenario fatigue risk value of the grid cell is determined as 0.81 based on the conservative principle of risk management, focusing on the highest risk level in the area. After assigning the scenario fatigue risk value, the system combines it with the existing smoke concentration value and toxic gas concentration value of the grid cell to form a data tuple containing three risk dimensions. The system repeats this operation for all grid cells in the warehouse, accurately corresponds and integrates the three key risk indicators (smoke concentration, toxic gas concentration, and scenario fatigue risk) according to the unified grid coordinates, and finally generates a complete and dynamic partition risk factor set.

[0079] The acquisition steps of the fusion risk situation map are as follows:

[0080] Based on the partition risk factor set, safety thresholds for smoke concentration value, toxic gas concentration value, and scenario fatigue risk value are set. The smoke concentration value, toxic gas concentration value, and scenario fatigue risk value in each grid cell are compared with the corresponding safety thresholds one by one. If any value exceeds the corresponding safety threshold, the grid cell is marked as a high-risk area. According to the risk marking results of the grid cells, the area risk is marked on the warehouse layout map by color classification, and the fusion risk situation map is established.

[0081] Specifically, based on the partition risk factor set, the system first loads the preset risk judgment safety threshold. The safety threshold of smoke concentration is set to 50 PPM according to the requirements of warehouse environment early warning. This value is usually the critical point of early warning of the detector. The safety threshold of toxic gas (taking carbon monoxide as an example) is the short-term exposure limit (STEL) value 25 PPM. The safety threshold of situational fatigue risk value is based on the risk classification standard established in the previous step. The high-risk value is set to 2.0. The system then starts a traversal program to read the data tuples of each grid cell in the partition risk factor set one by one. In each grid cell, the system compares the measured smoke concentration value, toxic gas concentration value and situational fatigue risk value with the above-mentioned safety thresholds of 50 PPM, 25 PPM and 2.0 respectively in terms of numerical size. The judgment logic is: as long as any one of the three values is greater than or equal to the corresponding safety threshold, the risk state of the grid cell is marked as "high risk". If all values are below the threshold, it is marked as "low risk". After completing the risk marking of all grid cells, the system visualizes the rendering on the loaded warehouse layout map according to the marking results, and uses a clear color classification standard to mark the area risk. Specifically, the grid cells marked as "high risk" are filled with red, and the grid cells marked as "low risk" are filled with green. At the same time, in order to provide more detailed early warning, for the grid cells with values between 75% and 100% of the corresponding safety threshold, the system fills them with yellow, indicating "medium risk" or "alert" state. In this way, abstract risk data is converted into intuitive visual information to establish a fusion risk situation map.

[0082] The resource availability state list acquisition step is:

[0083] The coordinates of the grid cells marked in the fusion risk situation map are analyzed to obtain the positioning coordinates and operation state identifiers of the inspection personnel, forklifts and fire-fighting robots. The positioning information of the fixed fire hydrants, fire extinguishers and emergency exits in the warehouse is called to form a resource availability state list classified by resource type.

[0084] Specifically, first, the fusion risk situation map generated in the previous stage is analyzed, and the coordinates of all grid cells marked as "high risk" are automatically filtered and extracted to form a high-risk area coordinate list, such as [GID-X13Y34, GID-X14Y34, GID-X25Y50]. At the same time, the system sends a real-time data request to the Internet of Things positioning network in the warehouse (a system composed of ultra-wideband UWB base stations and positioning tags) to obtain the current accurate position coordinates and operation status identifiers of all mobile emergency resources, including patrol personnel, forklifts, and fire robots. The operation status identifier is a digital code, and its corresponding relationship is clearly defined in the system configuration: 01 represents "idle standby", 02 represents "executing routine tasks", 03 represents "charging or maintenance", and 04 represents "fault or offline". These states are automatically reported by the device's own control system or personnel's handheld terminal. Then, the system retrieves the location information of all fixed emergency resources from a static facility information database, which has been measured and entered during the initial digital construction of the warehouse. This database contains the precise plane coordinates of each fire hydrant, handheld fire extinguisher, and emergency exit. Finally, the system integrates the dynamically obtained mobile resource information with the static fixed resource information, constructs a structured data list according to the pre-set classification standard (mobile personnel, mobile equipment, fixed facilities), and each item in the list contains the unique number of the resource, the type of the resource, the real-time or fixed position coordinates, and a clear availability status, thereby forming a resource availability status list classified by resource type.

[0085] The steps for obtaining the emergency resource response capability matrix are as follows:

[0086] Based on the resource availability status list, the comprehensive response capability score value is calculated, and the calculation formula is:

[0087] ;

[0088] Among them, represents the comprehensive response capability score value of the th emergency resource item to the th high-risk area, represents the availability status factor of the th emergency resource item, which is obtained by converting the operation status identifier, represents the processing efficiency coefficient of the th emergency resource item to the th high-risk area risk source, represents the length difference value of the th path in the horizontal axis direction, represents the length difference value of the th path in the vertical axis direction, represents the The average travel speed of resource items on a segment path, Indicates the first The resource item to the first The shortest travel path segments for each high-risk area Indicates the first Characteristic response time thresholds for resource types;

[0089] Based on the comprehensive response capability score, a comprehensive response capability score table is constructed according to the two-dimensional combination structure of resource items and high-risk areas. The emergency resource items are represented by rows and the high-risk areas are represented by columns, thus generating an emergency resource response capability matrix.

[0090] Specifically, the formula: The advantage of the formula lies in the introduction of an availability status factor. The formula can eliminate or reduce the weight of resources that are currently executing tasks or are in a faulty state in real time, ensuring scheduling feasibility. Processing efficiency coefficient. This achieves precise matching of resources and risk types, avoiding ineffective dispatching, such as sending forklifts without firefighting capabilities to handle fires. Most importantly, it utilizes the exponential decay term. It non-linearly transforms the key variable of response time into a penalty factor, accurately simulating the concept of "golden rescue time," which means that as response time increases, the effectiveness of resources decreases exponentially.

[0091] Indicates the first The availability status factor of each emergency resource item is derived from the resource's operational status identifier and is used to quantify the resource's immediate availability. This conversion is based on a preset "status-factor" mapping table, which is set according to the ease with which the resource is interrupted or accessed. For example, the operational status identifiers of inspection personnel, forklifts, and firefighting robots are mapped to... The transition rule is: when the state is "idle and ready" (identifier 01), When the status is 1.0, and the status is "performing a routine task" (identifier 02), it can be interrupted urgently, but there will be a certain delay and cost. The value is 0.4; when the status is "charging or maintenance" (identifier 03) or "fault or offline" (identifier 04), the resource is completely unavailable. The default status is 0. For fixed facilities such as fire hydrants and fire extinguishers, the status is set to available by default. The value is always 1.0.

[0092] Indicates the first One emergency resource item to respond to the first The processing efficiency coefficient of the main risk source in a high-risk area. This coefficient is stored in a "resource-risk" efficiency matrix, which is developed by a warehouse safety management expert team based on equipment performance specifications, personnel skill certifications, and emergency response plans. The matrix evaluates the effectiveness of each resource against different risks (such as fire, toxic gas, personnel fatigue). For example, the main risk source in a high-risk area is smoke concentration exceeding standards, then the settings for this area are as follows: fire-fighting robots equipped with high-pressure water cannons, with the highest efficiency, 0.9; handheld fire extinguishers, effective but limited in range and capacity, 0.7; inspection personnel, trained in fire fighting and can use fire extinguishers, but their main responsibility is to confirm the fire and guide, 0.3; forklifts, no fire-fighting capabilities, 0.

[0093] represents the shortest travel time (in seconds) of a resource to a high-risk area . This time is calculated by a path planning algorithm. The system uses the A* (A-star) algorithm on a pre-set warehouse electronic map (including shelves, walls, and other obstacles) to calculate the shortest feasible path from the resource's current location to the center point of the high-risk area. The algorithm divides the path into straight line segments. For the th segment of the path, and are its projection lengths on the coordinate axes, which is the length of this segment of the path (in meters). is the average travel speed of the resource on this segment of the path (in meters / second), which is obtained from a pre-set speed table according to the resource type and the type of the area where the path is located (such as main aisle, narrow lane, personnel passage). For example, a fire-fighting robot in a 3.5-meter-wide main aisle has a speed of 2.5 meters / second, and in a 2.5-meter-wide narrow lane between shelves it has a speed of 1.5 meters / second.

[0094] represents the characteristic response time threshold (in seconds) of the th type of resource, which defines the "golden time window" for effective response of this type of resource. The setting of this threshold is based on relevant safety standards and emergency drill data analysis. For example, for fire-fighting resources (fire-fighting robots, fire extinguishers), the critical time window for initial fire control is 60-120 seconds, so the threshold is set to 90 seconds. For inspection personnel who need to be replaced due to fatigue, the urgency of their response is slightly lower, ​It can be set to 300 seconds.

[0095] Calculation process:

[0096] Take the response ability of the fire-fighting robot (resource ) to a high-risk area (area ) as an example.

[0097] Known:

[0098] High-risk area The center coordinates are (60, 70), and the main risk is that the smoke concentration exceeds the standard.

[0099] Fire-fighting robot The current coordinates are (10, 10), and the working state is "idle standby".

[0100] Get parameter values:

[0101] : The robot state is "idle standby", and the table lookup gets .

[0102] : The robot should respond to the fire risk, and the efficiency matrix gets .

[0103] : The characteristic response time threshold of the fire-fighting robot is seconds.

[0104] Calculate the shortest travel time:

[0105] The A* algorithm plans the path, which includes two segments:

[0106] First segment (main channel): from (10, 10) to (60, 10). , The distance is 50 meters. This channel is a wide main channel, and the robot speed meters / second. The travel time is seconds.

[0107] Second segment (main channel): from (60, 10) to (60, 70). , The distance is 60 meters. This channel is also a main channel, and the speed meters / second. The travel time is seconds.

[0108] The shortest total travel time is seconds.

[0109] Substitute the formula to calculate :

[0110]

[0111] The result shows that the fire-fighting robot has a comprehensive response capability score of 0.552 for this high-risk area. This score reflects its current availability, professional matching, and ability to arrive at the scene in a "golden time" much shorter than its characteristic response time, so it is a very prioritized and efficient response option. The score value will be used later to build an emergency resource response capability matrix, providing a direct quantitative basis for dispatch decisions.

[0112] Based on the series of comprehensive response capability score values calculated in the previous stage, the system starts the matrix construction program. First, the program dynamically creates a two-dimensional table data structure, with the rows and columns dynamically defined according to the current emergency situation. The rows are designated to represent all emergency resources in an available or semi-available state, and the table header of each row is the unique number of the resource from the resource availability state list, such as "fire-fighting robot-01", "patrol officer-W07", "fire extinguisher-C2-04". The columns are used to represent all high-risk areas marked in the fusion risk situation map, and the table header of each column is the coordinate value of the corresponding high-risk grid cell, such as "GID-X13Y34", "GID-X25Y50". Subsequently, the system iterates through all "resource-high risk area" combination pairs through a double loop. For each combination, the system retrieves the unique comprehensive response capability score value calculated in the previous step from memory , and fills this value into the corresponding cell at the intersection of the row and column in the table, for example, the score value of 0.552 for fire-fighting robot-01 corresponding to high-risk area GID-X13Y34 will be filled into the designated cell. This filling process covers all identified high-risk areas and all available emergency resources. For any combination resulting in a score of zero due to the resource being completely unavailable (e.g., in a fault state, ), the corresponding cell will be filled with 0. After all cells are assigned values, the two-dimensional table filled with score values constitutes the current emergency resource response capability matrix.

[0113] The acquisition step of the emergency resource dispatch instruction set is:

[0114] Call all grid cells marked as high-risk areas in the fusion risk situation map, and analyze the scenario fatigue risk value, smoke concentration value, and toxic gas concentration value in each grid cell one by one. Compare the scenario fatigue risk value, smoke concentration value, and toxic gas concentration value with the corresponding safety risk threshold value one by one, calculate the number of risk values exceeding the safety risk threshold value in each grid cell, and arrange them in descending order according to the number of risk values exceeding the threshold from most to least to form a high-risk area priority ranking table.

[0115] Based on the ranking results of each grid cell in the high-risk area priority ranking table, the comprehensive response capability score values of all emergency resource items corresponding to each grid cell in the emergency resource response capability matrix are called one by one, the numerical values of all comprehensive response capability score values in each grid cell are compared, the emergency resource item with the largest numerical value is selected as the response resource of the corresponding grid cell, and a high-risk area emergency resource matching relationship table is formed.

[0116] Based on the high-risk area emergency resource matching relationship table, the response resource type, response resource current position coordinates and grid cell coordinates of the target high-risk area corresponding to each grid cell are extracted one by one, and resource scheduling instructions are generated one by one according to the information structure of emergency resource type, departure location and target location, forming an emergency resource scheduling instruction set.

[0117] Specifically, the system calls the list of all grid cells marked as high-risk areas in the fusion risk situation map, and for each grid cell in the list, the system parses the scenario fatigue risk value, smoke concentration value and toxic gas concentration value from the associated partition risk element set one by one. Subsequently, the system compares these three values with the safety risk threshold set in the previous step, specifically, the scenario fatigue risk value is compared with 2.0, the smoke concentration value is compared with 50PPM, and the toxic gas concentration value is compared with 25PPM. The system initializes a counter for each grid cell. Whenever a risk value is greater than or equal to its corresponding safety risk threshold, the counter of that grid cell is incremented by one. For example, for grid cell GID-X13Y34, its risk data is {scenario fatigue risk value: 2.2; smoke concentration: 65PPM; toxic gas concentration: 10PPM}, so the scenario fatigue risk value (2.2≥2.0) and smoke concentration (65PPM≥50PPM) exceed the limit, and the value of its counter is 2. For grid cell GID-X25Y50, its risk data is {scenario fatigue risk value: 1.8; smoke concentration: 55PPM; toxic gas concentration: 15PPM}, only the smoke concentration exceeds the limit, and the value of its counter is 1. After completing the statistical analysis of all high-risk area grid cells, the system sorts all high-risk area grid cells in descending order according to the values of their counters. If there are cases where the counter values are the same, further sorting is performed according to the sum of the exceeding limit amplitude percentages of the exceeding limit risk items in each grid cell. The exceeding limit amplitude percentage calculation formula is: ((actual value-threshold value) / threshold value). The larger the sum of the percentages of all exceeding limit items is, the higher the ranking is. In this way, a high-risk area priority ranking table is generated, which clearly indicates the danger level and complexity of each high-risk area.

[0118] Based on the high-risk area priority ranking table generated in the previous stage, the system processes each high-risk grid cell in order of priority from high to low. For the currently processed grid cell, the system calls the emergency resource response capability matrix generated in the previous step and extracts the corresponding column of data from it. This column of data contains the comprehensive response capability score values of all available emergency resource items for the specific high-risk grid cell. The system then compares the score values in this column and identifies the resource item with the highest comprehensive response capability score value by performing a maximum value search operation. This resource item is determined to be the optimal choice for processing the current high-risk grid cell. For example, for the highest priority grid cell GID-X13Y34, the system finds the corresponding column in the emergency resource response capability matrix and discovers that the fire robot-01 has a score of 0.552, the inspector-W07 has a score of 0.213, and the fire extinguisher-C2-04 has a score of 0.450. After comparing these values, the system determines that the fire robot-01 (0.552) is the best response resource and immediately establishes a unique matching relationship between "fire robot-01" and "GID-X13Y34". To avoid resource duplication, once a resource (such as fire robot-01) is assigned, it will be temporarily marked as "assigned" in the subsequent matching process for other low-priority high-risk areas and will no longer participate in the calculation unless a higher-priority task requires it. The system will reallocate according to the pre-set preemption rules. The system sequentially completes the matching of the optimal resource for each high-risk area in the ranking table and records each matching result (including the high-risk area coordinates and the selected response resource number) to form a complete high-risk area emergency resource matching relationship table.

[0119] Based on the high-risk area emergency resource matching relationship table, the system starts the instruction generation program, which reads the matching records in the table one by one. For each record, the system first extracts three key information: the grid cell coordinates of the target high-risk area, such as GID-X13Y34, the unique number of the matched response resource, such as fire robot-01, and the type of the response resource, i.e. "fire robot". The system then queries the resource availability status list to obtain the current real-time position coordinates of the response resource "fire robot-01", such as (10, 10). At the same time, the system converts the grid cell coordinates GID-X13Y34 of the target high-risk area into the physical coordinates of the center point, such as (60, 70). By now, the core elements required for a dispatch instruction ("who", "from where", "to where") have been prepared. The system then formats these information according to a predefined instruction template, which is structured as "{instruction ID: XXXX, resource type: [resource type], resource number: [resource number], departure location: [departure coordinates], target location: [target coordinates], instruction status: pending}". The extracted information is filled into the template to generate a specific dispatch instruction, such as "{instruction ID: CMD001, resource type: fire robot, resource number: fire robot-01, departure location: (10, 10), target location: (60, 70), instruction status: pending}". For personnel-type resources, the instruction content will also be attached with a brief task description, such as "go to the target area to confirm the fire and use fire extinguishing equipment". The system repeats this process for each record in the matching relationship table to generate structured resource dispatch instructions one by one, and collects all generated instructions to form the final emergency resource dispatch instruction set.

[0120] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments without departing from the technical solution content of the present application still falls within the protection scope of the present application.

Claims

1. A warehouse environment risk monitoring method, characterized by, The method comprises the following steps: obtaining the scheduling table and current working time of the forklift driver or the inspector in the target warehouse, collecting the task load factor of the current task type and the temperature and noise level of the working area, and calculating the situational fatigue risk value; calling the warehouse layout and goods distribution information of the warehouse, integrating the real-time readings of the fire smoke sensor and the toxic gas sensor, associating and assigning the situational fatigue risk value and the sensor readings on the corresponding grid cells of the warehouse layout, generating a partition risk element set, marking a high-risk area according to the partition risk element set, and establishing a fusion risk situation map; based on the fusion risk situation map, real-time obtaining the positions and working states of the Internet of Things positioning inspectors, forklifts and fire-fighting robots, and calling the fixed fire hydrant, fire extinguisher and emergency exit position information, establishing a resource availability state list, and calculating an emergency resource response capability matrix based on the resource availability state list; prioritizing the high-risk area in the fusion risk situation map, and selecting resources from the emergency resource response capability matrix for matching to generate an emergency resource scheduling instruction set; the situational fatigue risk value is obtained by: analyzing the scheduling table and current working time of the forklift driver or the inspector, extracting the start and end time of the continuous working section of the personnel, and calculating the total length of the continuous time period without alternating operation to obtain the personnel continuous working time value; according to the current task number, matching the corresponding task type and extracting the task load factor uniquely associated with the task type, synchronously calling the temperature sensor and noise sensor to obtain the temperature value and noise level value of the current working area, standardizing each environmental value with respect to the comfort threshold and danger threshold, and weighting each environmental factor with respect to the risk enhancement coefficient to form a comprehensive environmental stress parameter; calculating the situational fatigue risk value according to the personnel continuous working time value, reference working time value, task load factor and comprehensive environmental stress parameter; the acquisition step of the emergency resource response capability matrix is: based on the resource availability state list, calculating a comprehensive response capability score value; based on the comprehensive response capability score value, constructing a comprehensive response capability score value table according to the two-dimensional combination structure of the resource item and the high-risk area, indicating the emergency resource item by row and the high-risk area by list, and generating an emergency resource response capability matrix.

2. The warehouse environment risk monitoring method of claim 1, wherein, the acquisition step of the partition risk element set is: calling the warehouse layout and goods distribution information of the warehouse, dividing the layout into grid cells of a fixed size, analyzing the goods type, goods storage quantity and storage location in the goods distribution information, and mapping them to the corresponding grid cells one by one to generate a warehouse grid goods distribution mapping table; based on the warehouse grid goods distribution mapping table, calling the real-time readings of the fire smoke sensor and the toxic gas sensor, analyzing the installation position coordinates of each sensor and associating the corresponding grid cells, assigning the smoke concentration value and the toxic gas concentration value to each grid cell, and generating a warehouse grid environment sensor reading distribution table; Based on the warehouse grid environment sensor reading distribution table and the scenario fatigue risk value, the smoke concentration value, the toxic gas concentration value and the scenario fatigue risk value are merged and assigned according to the one-to-one correspondence of the warehouse grid unit coordinates to generate a partition risk element set.

3. The warehouse environment risk monitoring method of claim 1, wherein, The acquisition step of the fusion risk situation map is: Based on the partition risk element set, safety thresholds of smoke concentration value, toxic gas concentration value and scenario fatigue risk value are set, and the smoke concentration value, toxic gas concentration value and scenario fatigue risk value in each grid unit are compared with the corresponding safety threshold one by one. If any value exceeds the corresponding safety threshold, the grid unit is marked as a high-risk area. According to the risk marking result of the grid unit, the area risk is marked on the warehouse layout map by color classification to establish a fusion risk situation map.

4. The warehouse environment risk monitoring method of claim 1, wherein, The acquisition step of the resource availability state list is: The grid unit coordinate values of all high-risk areas marked in the fusion risk situation map are parsed to obtain the positioning coordinates and operation state identifiers of the inspection personnel, forklifts and fire-fighting robots. The positioning information of the fixed fire hydrant, fire extinguisher and emergency exit in the warehouse is called to form a resource availability state list classified by resource type.

5. The warehouse environment risk monitoring method of claim 1, wherein, The acquisition step of the emergency resource scheduling instruction set is: The fusion risk situation map is called to mark all grid units as high-risk areas. The scenario fatigue risk value, smoke concentration value and toxic gas concentration value in each grid unit are parsed one by one. The scenario fatigue risk value, smoke concentration value and toxic gas concentration value are compared with the corresponding safety risk threshold value one by one. The number of risk values exceeding the safety risk threshold in each grid unit is calculated. The risk value exceeding the limit number is arranged in descending order from more to less to form a high-risk area priority ranking table. Based on the ranking results of each grid unit in the high-risk area priority ranking table, the comprehensive response capability score values of all emergency resources corresponding to each grid unit in the emergency resource response capability matrix are called. The numerical values of all comprehensive response capability score values in each grid unit are compared one by one. The emergency resource item with the largest value is selected as the response resource of the corresponding grid unit to form a high-risk area emergency resource matching relationship table.

6. The warehouse environment risk monitoring method of claim 1, wherein, The acquisition step of the emergency resource scheduling instruction set also includes: based on the high-risk area emergency resource matching relationship table, the response resource type, the current location coordinates of the response resource and the grid unit coordinates of the target high-risk area corresponding to each grid unit are extracted one by one. The resource scheduling instruction is generated one by one according to the information structure of emergency resource type, starting position and target position to form an emergency resource scheduling instruction set.

7. The warehouse environment risk monitoring system of the warehouse environment risk monitoring method according to any one of claims 1 to 6, characterized in that, It includes: The scenario risk assessment module obtains the scheduling table of forklift drivers or inspection personnel in the target warehouse and the current operation time. The task load factor of the current task type and the temperature and noise level of the operation area are collected to calculate the scenario fatigue risk value. The fusion risk modeling module calls warehouse layout and cargo distribution information of the warehouse, integrates real-time readings of fire smoke sensors and toxic gas sensors, associates and assigns the situational fatigue risk value and the sensor readings on the corresponding grid cells of the warehouse layout map, generates a partition risk element set, labels a high-risk area according to the partition risk element set, and establishes a fusion risk situation map; The emergency resource assessment module, based on the fusion risk situation map, obtains the positions and working states of the Internet of Things positioning inspection personnel, forklifts and fire-fighting robots in real time, calls fixed fire hydrants, fire extinguishers and emergency exit position information, establishes a resource availability state list, and calculates an emergency resource response capability matrix based on the resource availability state list; The resource scheduling instruction module prioritizes the high-risk areas in the fusion risk situation map and selects resources from the emergency resource response capability matrix for matching to generate an emergency resource scheduling instruction set.

Images