A method, system, equipment and medium for preventing and controlling heatstroke based on workload
By acquiring the health records and real-time data of field personnel, and combining them with environmental and operational characteristics, the predicted heat load and risk index are calculated, solving the problem of inaccurate heatstroke risk assessment in existing technologies, and realizing personalized prevention and control strategies and improved safety.
Patent Information
- Application Number
- CN202611296905.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-25
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies lack accuracy in assessing and warning of heatstroke risks for field workers in high-temperature working environments, and fail to incorporate individual dynamic workloads, resulting in inaccurate risk assessments.
By acquiring health record data of field workers, combined with environmental meteorological data and business characteristic data, the preliminary heat load forecast value is calculated, and risk screening is carried out before dispatching orders; during the operation, microenvironmental and physiological data are monitored in real time to calculate the comprehensive heatstroke risk index and generate personalized prevention and control strategies.
It improved the accuracy of heatstroke risk assessment, and reduced the overwork hours of field staff in high-temperature environments by dynamically adjusting dispatch strategies and real-time prevention and control measures, thereby enhancing safety.
Smart Images

Figure CN123120174A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of monitoring equipment technology, and in particular to a method, system, equipment and medium for the prevention and control of heatstroke based on workload. Background Technology
[0002] With the continuous development of the on-demand delivery industry, the number of employees in field operations such as food delivery and express delivery continues to rise, and the coverage of delivery services is constantly expanding. How to monitor and prevent the risk of heatstroke among field workers in high-temperature environments is a current focus in the field of dispatch management.
[0003] In existing technologies, the identification and early warning functions for heatstroke risk during high-temperature operations are typically achieved through monitoring systems based on environmental meteorological parameters or single physiological indicators. For example, an alarm is triggered when the ambient air temperature or the instantaneous body surface temperature collected by the temperature measuring device exceeds a preset limit. However, in practical applications, existing technologies based on macroscopic environmental parameters or single physiological thresholds fail to consider the actual dynamic workload of field personnel during mobile operations. This results in insufficient accuracy in assessing and warning of heatstroke risk for individuals under different task intensities, particularly regarding their actual heat stress levels. Summary of the Invention
[0004] In view of this, this application provides a method, system, device and medium for preventing and controlling heatstroke based on workload, in order to solve the above problems.
[0005] Firstly, a method for preventing and controlling heatstroke based on workload is provided, the method comprising:
[0006] Obtain health record data of field workers and determine individual health correction coefficients for field workers based on the health record data;
[0007] Before dispatching target delivery orders to field workers, obtain environmental meteorological data of the target delivery area and business characteristic data of the target delivery orders. Based on individual health correction coefficients, environmental meteorological data, and business characteristic data, calculate the predicted value of the preceding heat load.
[0008] When the current heat load forecast value is greater than or equal to the preset interception threshold, the dispatch instruction for sending the target delivery order to the field operation personnel is intercepted;
[0009] When the current heat load forecast value is less than the interception threshold, the target delivery order is dispatched to the field operation personnel. During the execution of the target delivery order by the field operation personnel, the system receives the micro-environmental data of the current location of the field operation personnel and the real-time physiological heat stress data uploaded by the data collection device worn by the field operation personnel, and obtains the real-time work load data.
[0010] The environmental heat load sub-index is calculated based on microenvironment data, the individual physiological heat stress sub-index is calculated based on real-time physiological heat stress data, and the work load sub-index is calculated based on real-time work load data.
[0011] Based on the environmental heat load sub-index, individual physiological heat stress sub-index, work load sub-index, and individual health correction coefficient, the real-time comprehensive heatstroke risk index of field workers is calculated.
[0012] The system compares the real-time comprehensive heatstroke risk index with multiple preset risk classification thresholds in real time, and generates corresponding heatstroke risk management strategies based on the comparison results, which are then distributed to the mobile terminals of field workers.
[0013] Optionally, based on individual health correction factors, environmental meteorological data, and operational characteristic data, the predicted forward heat load is calculated, specifically including:
[0014] Extract temperature and humidity meteorological data from environmental meteorological data;
[0015] Extract order delivery distance and order weight from business characteristic data;
[0016] The environmental severity weights are determined based on temperature and humidity meteorological data.
[0017] Determine the order load weight based on the order delivery distance and the order weight;
[0018] The preceding heat load forecast is obtained by multiplying the environmental severity weight, the order load weight, and the individual health correction factor.
[0019] Optionally, environmental severity weights can be determined based on temperature and humidity meteorological data, specifically including:
[0020] Obtain the preset temperature mapping reference table and the preset humidity mapping reference table;
[0021] The temperature influence baseline is obtained by comparing temperature meteorological data with a temperature mapping benchmark table;
[0022] By comparing humidity meteorological data with a humidity mapping benchmark table, the humidity influence baseline is obtained;
[0023] The environmental severity weight is obtained by adding the base values for temperature and humidity.
[0024] Optionally, the environmental heat load sub-index is calculated based on microenvironment data, the individual physiological heat stress sub-index is calculated based on real-time physiological heat stress data, and the work load sub-index is calculated based on real-time work load data, specifically including:
[0025] Local real-time temperature and local solar radiation intensity are extracted from microenvironment data, and environmental heat load sub-index is determined based on local real-time temperature and local solar radiation intensity.
[0026] Body surface temperature and real-time heart rate are extracted from real-time physiological heat stress data, and individual physiological heat stress sub-indices are determined based on body surface temperature and real-time heart rate.
[0027] Extract the remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery from the real-time workload data, and determine the workload sub-index based on the remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery.
[0028] Optionally, the workload sub-index can be determined based on the remaining total delivery mileage, the current total delivery load, and the average number of floors climbed per delivery, specifically including:
[0029] Obtain the preset mileage load weight, preset heavy load weight, and preset floor load weight;
[0030] Multiply the remaining total delivery mileage by the mileage load weight to obtain the mileage energy consumption component;
[0031] Multiply the current total delivery load by the load weight to obtain the load energy consumption component;
[0032] Multiply the average number of floors climbed per delivery by the floor load weight to obtain the floor energy consumption component;
[0033] The energy consumption components of mileage, load, and floor are added together to obtain the work load sub-index.
[0034] Optionally, based on the comparison results, a corresponding heatstroke risk management strategy is generated and distributed to the mobile terminals of field workers, specifically including:
[0035] When the real-time comprehensive heatstroke risk index is greater than or equal to the preset extremely high risk classification threshold, the real-time location coordinates of the field workers are obtained.
[0036] Based on real-time location coordinates, multiple candidate online field personnel within a preset distance range are identified;
[0037] Obtain the candidate support risk index for each of the multiple candidate online field staff;
[0038] By comparing the candidate support risk index with the preset safety dispatch threshold, the target support personnel are selected from multiple candidate online field personnel.
[0039] Generate heatstroke prevention and self-rescue strategies for field workers, as well as emergency support strategies that include real-time location coordinates and physical cooling first aid guidelines;
[0040] Heatstroke prevention and self-rescue strategies were distributed to the mobile terminals of field workers, and emergency support strategies were distributed to the mobile terminals of the target support personnel.
[0041] Optionally, by comparing the candidate support risk index with a preset security dispatch threshold, target support personnel can be selected from multiple candidate online field personnel, specifically including:
[0042] From multiple candidates for online field personnel, candidates with a support risk index lower than the safety dispatch threshold are selected to obtain a set of safety candidates.
[0043] Obtain the real-time order acceptance load status and actual path distance from the real-time location coordinates of each candidate in the set of safety candidates;
[0044] From the set of safety candidates, select candidates whose real-time order-taking load status is not full to obtain the set of available candidates;
[0045] Sort the available candidate set in ascending order based on the actual path distance;
[0046] The top-ranked available candidate is identified as the target support personnel.
[0047] Secondly, a heatstroke prevention and control system based on workload is provided, the system comprising:
[0048] The coefficient determination module is used to acquire the health record data of field workers and determine the individual health correction coefficients of field workers based on the health record data.
[0049] The load forecasting module is used to obtain environmental meteorological data of the target delivery area and business characteristic data of the target delivery order before dispatching the target delivery order to the field workers. Based on the individual health correction coefficient, environmental meteorological data and business characteristic data, it calculates the forward heat load forecast value.
[0050] The order dispatch control module is used to intercept the dispatch instruction for sending target delivery orders to field workers when the current heat load forecast value is greater than or equal to the preset interception threshold; when the current heat load forecast value is less than the interception threshold, the target delivery order is dispatched to the field workers. During the execution of the target delivery order by the field workers, the module receives the microenvironmental data of the current location of the field workers and real-time physiological heat stress data uploaded by the data collection device worn by the field workers, and obtains real-time work load data.
[0051] The index calculation module is used to calculate the environmental heat load sub-index based on microenvironment data, the individual physiological heat stress sub-index based on real-time physiological heat stress data, and the work load sub-index based on real-time work load data.
[0052] The risk assessment module is used to calculate the real-time comprehensive heatstroke risk index of field workers based on the environmental heat load sub-index, the individual physiological heat stress sub-index, the work load sub-index, and the individual health correction coefficient.
[0053] The strategy distribution module is used to compare the real-time comprehensive heatstroke risk index with multiple preset risk classification thresholds in real time, and generate corresponding heatstroke risk management strategies based on the comparison results, which are then distributed to the mobile terminals of field workers.
[0054] Thirdly, an electronic device is provided, including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform any of the methods described above.
[0055] Fourthly, a computer-readable storage medium is provided, which stores instructions that, when executed, perform the method described in any of the above.
[0056] In summary, implementing one or more technical solutions provided in this application has at least the following technical effects or advantages:
[0057] By acquiring health record data of field workers, microenvironmental data collected by wearable devices, and real-time physiological heat stress data, the system adds the monitoring dimension of local microclimate and individual physiological state to the regional macro-meteorological environmental parameters, making the data basis for the system's heatstroke risk assessment more multidimensional.
[0058] By extracting business characteristic data and real-time workload data and calculating the corresponding workload sub-indices, the physical energy consumption characteristics of field personnel during mobile operations (such as delivery mileage, load, and number of floors climbed) are transformed into quantitative indicators and participated in comprehensive risk assessment. This enables the risk assessment results to change dynamically according to the intensity of specific delivery tasks, enhancing the matching degree between the assessment model and actual working conditions.
[0059] By comparing the predicted heat load with the preset interception threshold before dispatching orders to control the order dispatch instructions, and issuing corresponding heatstroke prevention and self-rescue strategies or triggering the dispatch instructions of target support personnel based on the comparison results of the real-time comprehensive risk index during order execution, a real-time data-driven order allocation control and terminal response mechanism has been constructed. This objectively reduces the overload working time of field personnel in high-temperature environments and improves the safety of field dispatch. Attached Figure Description
[0060] Figure 1 This is an exemplary system architecture diagram of a heatstroke prevention and control method or a heatstroke prevention and control system based on workload according to this application;
[0061] Figure 2 This is a flowchart illustrating a heatstroke prevention and control method based on workload disclosed in this application;
[0062] Figure 3 This is a schematic diagram of a heatstroke prevention and control system based on workload disclosed in this application;
[0063] Figure 4 This is a schematic diagram of the structure of an electronic device disclosed in this application.
[0064] Explanation of reference numerals in the attached diagram: 301, Coefficient determination module; 302, Load forecasting module; 303, Order dispatch control module; 304, Index calculation module; 305, Risk assessment module; 306, Strategy distribution module; 401, Processor; 402, Communication bus; 403, User interface; 404, Network interface; 405, Memory. Detailed Implementation
[0065] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.
[0066] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.
[0067] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.
[0068] Figure 1 An exemplary system architecture diagram is shown, illustrating an embodiment of a heatstroke prevention and control method or a heatstroke prevention and control system based on workload to which this application can be applied.
[0069] like Figure 1 As shown in the embodiments of this application, the system architecture is hierarchically divided based on the data flow sequence and control logic, constructing a five-layer prevention and control architecture from top to bottom, including an external data source layer, a cloud-based intelligent management and control platform, a transmission and interaction layer, a terminal acquisition and display layer, and an emergency response layer. The coordinated operation of this five-layer architecture provides underlying hardware and software support for the execution of subsequent methods and processes in this application.
[0070] In the specific system operation and hierarchical interaction process, the external data source layer and the terminal acquisition and display layer together constitute the data input end of the entire prevention and control system. Specifically, the basic business and meteorological database in the external data source layer (which includes meteorological databases, health information databases, and platform order dispatch databases, etc.) is used to provide the system with environmental meteorological data of the target delivery area, business characteristic data of the target delivery orders, and health record data of field workers, providing basic support for the system's pre-risk assessment. At the same time, the acquisition devices worn by field workers in the terminal acquisition and display layer act as edge sensing antennas, sensing and reporting micro-environmental data and individual physiological heat stress data in real time during order execution, thereby providing real-time data input for dynamic risk assessment during the process. The data generated by these two layers are all aggregated and reported through the transmission and interaction layer composed of communication networks.
[0071] The cloud-based intelligent management and control platform, serving as the central hub for the entire system's computation and scheduling, can be physically deployed as a backend server cluster. This platform houses the core risk assessment algorithm model of this application, as well as... Figure 3The various functional modules shown include the coefficient determination module, load forecasting module, dispatch control module, index calculation module, risk assessment module, and strategy distribution module. After receiving the input basic data and reported data, the cloud-based intelligent management and control platform uses its computing power cluster to execute core computing logic, transforming multi-dimensional discrete input parameters into quantitative results such as pre-calculated heat load values and real-time comprehensive heatstroke risk indices.
[0072] Subsequently, the system enters the control command issuance phase. The cloud-based intelligent control platform generates warnings and control commands based on real-time risk classification comparison results, which are then distributed downwards via the transmission and interaction layer. Under normal risk conditions, downlink commands flow to mobile terminals (e.g., smartphones running field applications) in the terminal acquisition and display layer or offline display terminals to directly execute order interception commands or issue heatstroke prevention and self-rescue strategies. When the risk index reaches the extremely high risk classification threshold, the control command will simultaneously trigger an extremely high-risk emergency dispatch, i.e., trigger the emergency linkage layer. At this time, the system activates the corresponding collaborative support and emergency rescue terminals. That is, the mobile terminals of the target support personnel (nearby online field personnel candidates) and the medical emergency rescue platform within the emergency linkage layer will receive the emergency support strategy, thus completing a complete prevention and control closed loop from front-end data perception and cloud algorithm decision-making to multi-cross collaborative intervention.
[0073] Figure 2 This is a flowchart illustrating a heatstroke prevention method based on workload in an embodiment of this application. This method can be implemented using a computer program, a microcontroller, or run on a workload-based heatstroke prevention system. The computer program can be integrated into the application or run as a standalone utility application. The specific steps of a workload-based heatstroke prevention method are described in detail below.
[0074] S201: Obtain health record data of field workers and determine the individual health correction coefficient of field workers based on the health record data.
[0075] In this embodiment, health record data refers to a set of basic data used to record the physical health status and medical history of field workers. It reflects the inherent physiological tolerance of workers when facing high-temperature environments. For example, it may include information such as age, body mass index (BMI), underlying disease markers, and historical heatstroke records. The individual health correction coefficient is used to represent a numerical parameter for quantitatively assessing the individual physiological vulnerability of field workers. It aims to personalize the general heatstroke assessment algorithm to objectively distinguish the differences in the risk of disease among different physical groups in the same high-temperature environment. For example, for young people without underlying medical history, the coefficient can be taken as a base value of 1.0, while for elderly people with underlying cardiovascular diseases, the coefficient can be adjusted upward to 1.5.
[0076] Specifically, health record data of field workers is obtained by connecting to a pre-set business database or personnel registration information database. After obtaining the health record data, core characteristic parameters such as age, number of previous heatstroke incidents, and types of underlying diseases are extracted. A basic health correction coefficient is set to a preset standard value (e.g., 1.0), and punitive compensation calculations are performed based on the core characteristic parameters: if the age exceeds a preset age threshold, the corresponding first compensation weight is matched; if there is a history of heatstroke, the second compensation weight is accumulated based on the number of heatstroke incidents; if specific underlying diseases such as cardiovascular diseases are included, a third compensation weight is matched. The standard value is summed with all triggered compensation weights to obtain the individual health correction coefficient of the field worker (its underlying logic can be expressed as: ,in For individual health correction factor, This is the standard value. (These are the compensation weights corresponding to each characteristic parameter). Through the above-mentioned quantization and superposition mechanism, multi-dimensional abstract health record information was successfully transformed into multiplier factors that can be directly used in subsequent pre-heat load prediction calculations.
[0077] S202: Before dispatching target delivery orders to field workers, obtain environmental meteorological data of the target delivery area and business characteristic data of the target delivery orders, and calculate the predicted value of the preceding heat load based on the individual health correction coefficient, environmental meteorological data and business characteristic data.
[0078] For example, this step aims to build a pre-emptive risk screening mechanism in the order dispatching process, significantly shifting the heatstroke prevention checkpoint from the traditional "passive early warning during the event" to "proactive avoidance before the event." In the underlying business scheduling workflow, the system, by embedding this safety verification node, breaks away from the traditional order dispatching engine's single logic of considering only capacity status and physical distance. Its core execution logic lies in: before the actual physical delivery action occurs, it integrates across dimensions the objective severity of the macro-meteorological environment, the expected physical exertion associated with the order itself, and the inherent physiological tolerance of the workers, proactively simulating and predicting the comprehensive heat stress level that will result from performing the task. This prediction mechanism transforms abstract environmental and business pressures into concrete, quantifiable risk indicators, thereby providing a quantitative decision-making basis for subsequent source control strategies such as high-risk order interception, order splitting, or staggered scheduling, effectively preventing field personnel from passively bearing excessive health risks under extreme high-temperature conditions.
[0079] In one possible implementation, the predicted heat load is calculated based on individual health correction coefficients, environmental meteorological data, and business characteristic data. Specifically, this includes: extracting temperature and humidity meteorological data from the environmental meteorological data; extracting order delivery distance and order weight from the business characteristic data; determining environmental severity weights based on the temperature and humidity meteorological data; determining order load weights based on the order delivery distance and order weights; and multiplying the environmental severity weights, order load weights, and individual health correction coefficients to obtain the predicted heat load.
[0080] In this embodiment, the environmental severity weight refers to a numerical parameter used to quantify the degree of aggravation of the impact of external meteorological conditions on human heat stress. It reflects the objective obstacles caused by natural environments such as high temperature and high humidity to the heat dissipation of workers. For example, when the temperature reaches 39 degrees Celsius and the relative humidity reaches 80%, this weight can be represented as a significantly amplified penalty coefficient. The order load weight is used to represent a quantitative indicator of the intensity of physical exertion associated with the delivery task itself. It reflects the additional physical exertion of workers due to the distance and weight of the order. For example, the load weight of an order with a delivery distance of 5 kilometers and a weight of 10 kilograms will be much higher than that of a short-distance light order. The pre-determined heat load value refers to the overall heatstroke risk load that workers are expected to bear, which is calculated in advance based on multi-dimensional constraints before the actual execution of the order dispatch action. It serves as the core criterion for whether to intercept the current order. For example, the predicted value can be a specific risk score between 0 and 100.
[0081] Specifically, after entering the pre-delivery task heat load screening stage, the environmental meteorological data of the target delivery area is first structured and analyzed to extract temperature and humidity meteorological data that directly affect human heat exchange. Simultaneously, the business characteristic data of the target delivery orders is read at the field level to extract the order delivery distance and order weight, which determine the amount of physical exertion. Based on the extracted temperature and humidity meteorological data, a preset meteorological assessment algorithm or mapping table is used for joint calculation to determine the environmental severity weight, objectively quantifying the degree of heat stress deterioration under current meteorological conditions. Furthermore, based on the extracted order delivery distance and order weight, combined with a preset physical exertion conversion logic, an evaluation is performed to determine the order load weight. The determined environmental severity weight, order load weight, and the individual health correction coefficient obtained in the previous steps are input into a preset risk multiplication model for multiplication. Through the interaction and superposition of parameters from the three dimensions of environment, business, and individual physiology, a preliminary heat load prediction value is finally obtained, providing quantitative data support for subsequent order interception or release decisions.
[0082] In one possible implementation, the environmental severity weight is determined based on temperature and humidity meteorological data, specifically including: obtaining a preset temperature mapping benchmark table and a preset humidity mapping benchmark table; obtaining a temperature influence base by comparing the temperature meteorological data with the temperature mapping benchmark table; obtaining a humidity influence base by comparing the humidity meteorological data with the humidity mapping benchmark table; and obtaining the environmental severity weight by adding the temperature influence base and the humidity influence base.
[0083] In this embodiment, the temperature mapping benchmark table refers to a pre-constructed data dictionary reflecting the correspondence between physical temperature ranges and the degree of human heat stress deterioration. It aims to discretize continuous temperature meteorological parameters into standard scores that can be used for algorithm calculation. For example, the benchmark table can record a score of 2 for the range of 35°C to 37°C. The humidity mapping benchmark table refers to a pre-set set of mapping rules between relative humidity and the evaporation resistance of sweating on the body surface. It reflects the objective degree of inhibition of heat dissipation efficiency by high humidity environment. For example, the benchmark table can record a penalty score of 1.5 for relative humidity reaching 80% or above. The temperature influence base and humidity influence base are used to represent the specific dimensionless risk penalty value obtained after table lookup matching. They independently quantify the individual contribution of specific environmental factors to the overall climate heat load. For example, under specific high temperature and high humidity weather, the temperature influence base may be 2 and the humidity influence base may be 1.5.
[0084] Specifically, when assessing the severity of current weather conditions, the underlying storage resources are first accessed to obtain a preset temperature mapping benchmark table and a preset humidity mapping benchmark table. The temperature and meteorological data of the target delivery area parsed in the previous steps are used as the retrieval basis. By comparing the temperature and meteorological data with the preset temperature intervals in the temperature mapping benchmark table, the associated risk values corresponding to the weather conditions are extracted, resulting in the temperature impact base. Using the same mapping logic, the extracted humidity and meteorological data are compared with the preset humidity intervals in the humidity mapping benchmark table to find and extract the associated risk values corresponding to the intervals they fall into, resulting in the humidity impact base. Considering the significant combined negative effects of high temperature and high humidity on human body heat load, the underlying logic directly adds the temperature impact base and humidity impact base obtained from the table lookup to achieve linear fusion of individual risk values, ultimately yielding the environmental severity weight.
[0085] S203: When the current heat load forecast value is greater than or equal to the preset interception threshold, the dispatch instruction for sending the target delivery order to the field operation personnel is intercepted.
[0086] In this embodiment, the interception threshold refers to a pre-set critical numerical boundary for determining whether the risk of heatstroke has reached a dangerous level. As a safety red line in the scheduling business logic, it aims to block the issuance of high-risk work tasks that are very likely to induce severe heatstroke. For example, the interception threshold can be set to 85 points corresponding to the real-time comprehensive heatstroke risk index. The dispatch instruction refers to the underlying system data signal or control logic used to trigger the task allocation action. It transforms the business decision of allocating orders into computer communication actions that can be received and executed by the terminal. For example, a network data packet containing order details is sent to the mobile terminal of the field worker and forces him to accept the order.
[0087] Specifically, after calculating the preliminary heat load forecast, the automated risk verification and safety blocking process begins. In the underlying logic, the real-time calculated preliminary heat load forecast is compared with a preset interception threshold. If the preliminary heat load forecast is greater than or equal to the preset interception threshold, it indicates that considering the current severe weather conditions, heavy order workload, and the physiological tolerance of the workers, forcibly dispatching the order would lead to an extremely high probability of heatstroke. At this point, the safety scheduling circuit breaker mechanism of the business platform is automatically triggered, directly intercepting the dispatch instruction for the target delivery order to the field workers at the data distribution interface. By interrupting the network transmission of this dispatch instruction or suspending the instruction in the process, it ensures that the target delivery order is not pushed to the field workers' mobile terminals, thereby preventing the occurrence of high-intensity heat exposure risks at the very source of the task flow.
[0088] S204: When the current heat load forecast value is less than the interception threshold, the target delivery order is dispatched to the field operation personnel. During the execution of the target delivery order by the field operation personnel, the micro-environmental data of the current location of the field operation personnel and the real-time physiological heat stress data uploaded by the data collection device worn by the field operation personnel are received, and the real-time operation load data is obtained.
[0089] In this embodiment, microenvironmental data refers to a set of parameters reflecting the specific meteorological and thermodynamic state within a very small space surrounding the worker. It is used to represent the actual severity of localized weather conditions that macro-weather forecasts cannot cover, such as real-time local air temperature on unshaded road sections and the intensity of solar radiation reflected from the road surface. Real-time physiological heat stress data is used to represent dynamic physiological compensation indicators generated by the human body to maintain core body temperature homeostasis under extreme heat exposure conditions. It aims to objectively quantify the current level of heat damage risk and pressure the body is experiencing, such as continuously high-frequency collected skin temperature and real-time physiological heat stress data. Characteristic values such as heart rate; real-time workload data refers to indicators used to evaluate the intensity of physical work continuously output by field workers at a specific time point. It reflects the driving effect of physical labor on the body's internal heat production rate. For example, it may include data such as the remaining total delivery mileage or current delivery load. Data collection equipment refers to smart hardware terminals worn by field workers that have IoT sensing, positioning and wireless communication capabilities. It acts as the physical sensing antennae for collecting objective data from the work site. For example, it may be a smart badge or temperature measuring wristband that integrates a heart rate sensor and a body temperature probe.
[0090] Specifically, in the underlying risk verification logic, if the predicted current heat load is less than the interception threshold, it is determined that the current overall environment and business conditions are within a safe and controllable range. This allows the order dispatch restriction to be lifted, and the target delivery order to be normally dispatched to the field staff for execution. Throughout the entire process of the field staff receiving and executing the target delivery order (i.e., during the cycling or walking route), a continuous data communication link is maintained with the underlying sensing hardware and business platform. On one hand, the system receives data from the field staff's wearable devices via wireless network, continuously sensing and uploading high-frequency data on the micro-environment and real-time physiological heat stress of the staff's current location. This enables dynamic tracking of local micro-climate changes along the route (such as moving from shaded areas to sunny sections) and fluctuations in the staff's physical characteristics. On the other hand, it synchronously connects to the backend scheduling interface to read the order backlog and flow status of the field staff during that period and obtain real-time work load data. Through these parallel external reception and internal acquisition actions, a real-time and highly individualized data foundation is provided for subsequent dynamic risk index sub-item calculations during order execution.
[0091] S205: Calculate the environmental heat load sub-index based on microenvironment data, calculate the individual physiological heat stress sub-index based on real-time physiological heat stress data, and calculate the work load sub-index based on real-time work load data.
[0092] For example, within the dynamic cycle of order execution, the system employs a multi-dimensional decoupled computational architecture to independently reduce and map the real-time aggregated multi-source heterogeneous data. The core objective of this step is to separate the complex factors inducing heatstroke into three relatively independent influencing dimensions: objective thermal stress from external natural conditions, compensatory responses within the body to maintain thermal balance, and the additional physical labor exerted by business tasks. By transforming discrete physical meteorological parameters, physiological signs, and business flow characteristics into standardized sub-item scores with unified scale, the system can not only intuitively identify the main sources of current surges in heat stress risk (e.g., whether it is due to extreme weather or excessive climbing leading to a sudden increase in load), but also complete the logical abstraction from raw perceived data to high-order feature indicators, thereby providing a structured and standardized basic input matrix for subsequently building a global comprehensive risk assessment model.
[0093] In one possible implementation, the environmental heat load sub-index is calculated based on microenvironment data, the individual physiological heat stress sub-index is calculated based on real-time physiological heat stress data, and the work load sub-index is calculated based on real-time work load data. Specifically, this includes: extracting local real-time temperature and local solar radiation intensity from microenvironment data, and determining the environmental heat load sub-index based on local real-time temperature and local solar radiation intensity; extracting body surface temperature and real-time heart rate from real-time physiological heat stress data, and determining the individual physiological heat stress sub-index based on body surface temperature and real-time heart rate; and extracting the remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery from real-time work load data, and determining the work load sub-index based on the remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery.
[0094] In this embodiment, the environmental heat load sub-index refers to a numerical value used to quantify the external heat pressure caused by the current micro-natural physical environment of field workers. It reflects the degree of objective obstruction of heat dissipation by the human body due to local high temperature and sunlight conditions. For example, this sub-index can be a specific score that integrates street surface sun exposure and local high temperature conditions. The individual physiological heat stress sub-index is a numerical parameter used to represent the intensity of the compensatory physiological response of the field worker's body when resisting the external heat environment and internal heat production. It aims to intuitively reflect whether the body is close to the critical state of heat exhaustion or heatstroke. For example, a high index means that the person has experienced abnormally high body surface temperature or excessively rapid heart rate. The work load sub-index is a quantitative indicator to assess the intensity of physical work that field workers need to continuously exert in the current delivery task flow. It reflects the driving effect of heavy business tasks under the new employment form on the accumulation of internal heat in the human body. For example, this index converts the high energy consumption state caused by carrying multiple heavy objects at the same time and frequently climbing stairs into a specific risk score.
[0095] Specifically, during the monitoring period when an order is in the execution phase, structured analysis is performed on the real-time aggregated multi-dimensional data stream. On one hand, local real-time temperature and local solar radiation intensity, which directly determine the heat exchange efficiency of the human body environment, are extracted from the received microenvironment data. Based on these local real-time temperature and local solar radiation intensity, the environmental heat load sub-index is determined. The calculation logic can be expressed as follows:
[0096]
[0097] in, This refers to the environmental heat load sub-index; This is a local real-time temperature, in degrees Celsius. ); This refers to the local solar radiation intensity, expressed in watts per square meter (W / m²). ); and These are the pre-calibrated temperature weighting coefficient and radiation intensity weighting coefficient, respectively.
[0098] On the other hand, the real-time physiological heat stress data collected and uploaded is simultaneously analyzed for features. From this data, body surface temperature, reflecting the body's core thermometabolic state, and real-time heart rate, representing cardiovascular compensatory load, are extracted. Based on body surface temperature and real-time heart rate, individual physiological heat stress sub-indices are determined. The calculation logic can be expressed as follows:
[0099]
[0100] in, This refers to the sub-index of individual physiological heat stress. Body surface temperature, in degrees Celsius (°C) ); This is the real-time heart rate, measured in beats per minute (Hz). ); and These are the pre-calibrated body temperature weighting coefficient and heart rate weighting coefficient, respectively.
[0101] Simultaneously, fields are extracted from the real-time workload data continuously flowing in the background, including the remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery, which are directly related to the physical exertion of field staff. Preset mileage-load weights are then obtained. Preset load weight and preset floor load weights ; the remaining total delivery mileage (unit: km) and Multiply by the product to obtain the energy consumption per mileage; then multiply by the total current delivery load. (Unit: kg) and Multiply by the product to obtain the energy consumption component for the load; calculate the average number of floors climbed per delivery. (Unit: layer) and Multiplying these components yields the floor-level energy consumption component. Considering that in real-world field operations, distance, load, and stair climbing often act simultaneously and cumulatively on the worker's body, these three energy consumption components are linearly added together to obtain the workload component index. The calculation logic can be expressed as follows:
[0102]
[0103] Through the above decomposition, the abstract business and physical parameters are specifically transformed into risk values with unified dimensions, laying a mathematical foundation for the subsequent output of comprehensive and integrated heatstroke risk assessment results.
[0104] In one possible implementation, the workload sub-index is determined based on the remaining total delivery mileage, the current total delivery load, and the average number of floors climbed per delivery. Specifically, this includes: obtaining preset mileage load weights, preset load weights, and preset floor load weights; multiplying the remaining total delivery mileage by the mileage load weights to obtain the mileage energy consumption component; multiplying the current total delivery load by the load weight weights to obtain the load energy consumption component; multiplying the average number of floors climbed per delivery by the floor load weights to obtain the floor energy consumption component; and adding the mileage energy consumption component, the load energy consumption component, and the floor energy consumption component to obtain the workload sub-index.
[0105] In this embodiment, mileage load weight refers to an evaluation coefficient used to quantify the impact of unit delivery distance on the physical exertion of field workers. It reflects the objective driving effect of spatial displacement on the body's metabolic heat production. For example, for normal urban flat road conditions, this weight can be set to a preset basic constant. Weight load weight is used to represent a standard parameter for measuring the intensity of additional physical burden generated by carrying a unit weight of goods. It aims to convert the physical quantity of load into a physiological energy consumption score. For example, when delivering large, heavy items, this weight will be proportionally amplified to reflect the high-intensity muscle work pressure. Floor load weight refers to the quantification of the impact of vertical movement on physical exertion. When workers climb stairs, the conversion coefficient of the severe physical exertion caused by a single floor highlights the special high heat generation characteristics of non-flat ground operations. For example, for old residential areas without elevators, the load weight of that floor will be set to a punitive value that is much higher than walking on flat ground. The energy consumption component (including mileage energy consumption component, load energy consumption component, and floor energy consumption component) is used to represent the intermediate quantitative result derived by combining actual business parameters with corresponding benchmark weights. They independently characterize the expected physical load generated under each single business dimension. For example, the mileage energy consumption component is an absolute value that converts the actual "kilometers" into a specific heat stress risk score.
[0106] Specifically, when determining the overall task's workload status, the system first obtains pre-calibrated preset mileage load weights, preset heavy load weights, and preset floor load weights by calling the underlying baseline parameter configuration library or algorithm rule library. Then, in the data fusion multiplication stage, the remaining total delivery mileage extracted in real-time from the previous steps is multiplied by the obtained mileage load weights. This multiplication operation converts the remaining spatial distance into a mileage energy consumption component representing the cost of displacement work. Using the same underlying logic, the current total delivery load extracted in the previous steps is multiplied by the heavy load weight to obtain a precisely quantified heavy load energy consumption component. Finally, the average number of floors climbed per delivery is multiplied by the floor load weight to obtain a floor energy consumption component measuring the extreme work intensity in the vertical dimension. Considering that in real field operations, distance, load, and stair climbing often act simultaneously on the operator's body, the algorithm logic in the background directly adds the calculated energy consumption components of mileage, load, and floors to achieve linear summation and fusion of the three core business energy consumption dimensions, and finally obtains the work load sub-index that can comprehensively reflect the total physical burden of field tasks.
[0107] S206: Based on the environmental heat load sub-index, individual physiological heat stress sub-index, work load sub-index, and individual health correction coefficient, calculate the real-time comprehensive heatstroke risk index for field workers.
[0108] In this embodiment, the real-time comprehensive heatstroke risk index refers to the final output index used to comprehensively and quantitatively assess the probability of field workers suffering from severe heatstroke or heatstroke at the current work time. It fully integrates multiple influencing factors such as the severity of the external environment, the body's own compensatory state, the current physical exertion of the work, and the individual's inherent physical condition. It is the sole core criterion for triggering subsequent graded control and emergency rescue strategies. For example, the index can be a normalized comprehensive risk score value from 0 to 100. When the value climbs to above 90 points, it means that the field worker is in an extremely dangerous critical state of heatstroke onset.
[0109] Specifically, after independently decomposing and calculating the three dynamic dimensions of external weather, physiological characteristics, and operational energy consumption, the final stage of global risk aggregation and integration begins. In the underlying risk assessment model, pre-configured environmental weight coefficients, physiological weight coefficients, and operational weight coefficients are first obtained (these weight coefficients can be pre-calibrated based on historical epidemiological injury data). The environmental heat load sub-index, individual physiological heat stress sub-index, and operational load sub-index, calculated in real-time in the preceding steps, are extracted. These three dynamic sub-indices are multiplied by their corresponding weight coefficients and then weighted and summed to preliminarily calculate the absolute heat stress pressure baseline under the current dynamic operating conditions. The individual health correction coefficient determined in the pre-initialization stage is extracted and added as an adjustment constant to the above weighted summation result (its calculation logic can be expressed as: ,in To provide a real-time comprehensive heatstroke risk index, , , These are the sub-indices for environmental, physiological, and work load, with A, B, and C being the corresponding weighting coefficients, and M being the individual health correction coefficient. Through the above-mentioned multivariate linear fusion calculation with specific weighting ratios and individual correction compensation, the real-time comprehensive heatstroke risk index of field workers is finally calculated, thus transforming complex and multidimensional time-series data into a specific single quantitative score.
[0110] S207: Real-time comparison of the real-time comprehensive heatstroke risk index with multiple preset risk classification thresholds, and generation of corresponding heatstroke risk management strategies based on the comparison results, which are then distributed to the mobile terminals of field workers.
[0111] For example, the system is pre-configured with multiple risk classification thresholds distributed in a tiered manner (e.g., divided into low-risk, medium-risk, high-risk, and extremely high-risk thresholds) to achieve dynamic and tiered control over the risk of heatstroke among field personnel. In the core execution logic, the system automatically matches and generates differentiated intervention strategies, ranging from gentle prompts and rigid blocking to emergency rescue, based on the specific threshold range into which the real-time calculated index falls. For instance, when the risk index is in the low or medium-risk range, the control strategy mainly manifests as sending preventative instructions such as "timed hydration voice prompts" or "suggest resting nearby and intelligently postponing subsequent orders" to mobile terminals; when the index climbs to the high-risk range, the control strategy will be upgraded to "forcibly locking order-taking permissions for 30 minutes" and simultaneously sending a business blocking instruction of "navigating to the nearest cooling station"; and when the risk index exceeds the extremely high-risk threshold, it is determined that the field personnel are in an extremely dangerous critical state. At this time, the system will directly generate heatstroke prevention and self-rescue strategies and emergency support strategies, and trigger collaborative rescue at the emergency linkage layer. Through this multi-level comparison and pre-intervention mechanism, the system can effectively resolve the accumulation of heat load on operators without completely interrupting business operations. Furthermore, combined with the external mutual assistance and rescue mechanism triggered by extremely dangerous situations (described below), the system control process realizes the execution of corresponding intervention strategies based on different risk classification thresholds.
[0112] In one possible implementation, a corresponding heatstroke risk management strategy is generated based on the comparison results and distributed to the mobile terminals of field workers. Specifically, this includes: obtaining the real-time location coordinates of the field workers when the real-time comprehensive heatstroke risk index is greater than or equal to a preset extremely high risk classification threshold; identifying multiple candidate online field workers within a preset distance range based on the real-time location coordinates; obtaining the candidate support risk index for each of the multiple candidate online field workers; selecting target support personnel from the multiple candidate online field workers by comparing the candidate support risk index with a preset safety dispatch threshold; generating a heatstroke prevention and self-rescue strategy for the corresponding field workers, as well as an emergency support strategy including real-time location coordinates and physical cooling first aid guidelines; distributing the heatstroke prevention and self-rescue strategy to the mobile terminals of the field workers, and distributing the emergency support strategy to the mobile terminals of the target support personnel.
[0113] In this embodiment, the extremely high risk grading threshold refers to a pre-set limit risk value boundary characterizing the pathological reaction of severe heatstroke that is about to occur or has already occurred in the human body. It is intended to serve as a red line for triggering the highest level of mandatory intervention and on-site emergency rescue procedures. For example, this threshold can be set to 95 points corresponding to the real-time comprehensive heatstroke risk index. Candidate online field personnel refer to fellow workers who are geographically close to high-risk heatstroke personnel and are currently connected to the system online. They constitute a potential pool of nearby emergency rescuers. For example, they can be five other delivery riders within 3 kilometers of the current location of the deliveryman in danger and whose mobile terminals are connected to the network. The candidate support risk index refers to an indicator value that quantifies the current risk of heatstroke onset borne by the potential rescuers. It is used to verify the physiological safety status of the rescuers before dispatching rescue to prevent secondary occurrences. Secondary heatstroke injuries, for example, a lower index indicates that the colleague is currently in good physical condition and not under severe heat load; heatstroke prevention and self-rescue strategies refer to a set of operational instructions issued to field workers facing severe heatstroke risks, guiding them to perform emergency self-physiological protection. It aims to guide high-risk personnel to cut off the heat source during the golden time before professional or external rescue arrives. For example, it may include mandatory voice instructions such as "stop work immediately", "move to a cool and ventilated place", and "loosen collar clothing to dissipate heat"; emergency support strategies refer to the dispatch and execution plan issued to selected rescuers, which includes the spatial location of the person in need of help and specific first aid methods. It is used to guide colleagues to quickly arrive at the scene and take standardized cooling and rescue measures. For example, this strategy may be a pop-up task card that includes GPS positioning and navigation routes for people suffering from heatstroke and standardized first aid graphic steps such as "do not blindly give water" and "use cold water for physical wiping".
[0114] Specifically, after completing the dynamic risk assessment, the resulting real-time comprehensive heatstroke risk index is compared in real time with multiple preset risk classification thresholds. When the real-time comprehensive heatstroke risk index is determined to be greater than or equal to the preset extremely high risk classification threshold, it indicates that the field worker has lost the physiological conditions to continue working and is in imminent danger of death. At this time, the underlying positioning service interface is immediately invoked to obtain the real-time location coordinates of the field worker. Using these real-time location coordinates as the center, combined with the pre-configured emergency search radius, multiple candidate online field workers within a preset distance range are identified in the geographic information database. To avoid the secondary injury risk of "high-risk rescue of high-risk," the candidate support risk index of each of the multiple candidate online field workers is obtained simultaneously through the background calculation engine. By comparing the candidate support risk index with the preset safety dispatch threshold, colleagues with higher risks are excluded, and target support personnel with the physical ability and safety conditions to provide rescue are selected from the multiple candidate online field workers. After the rescue force is locked, the background simultaneously generates the corresponding heatstroke prevention and self-rescue strategy for the field worker, as well as an emergency support strategy including real-time location coordinates and physical cooling first aid guidelines. Two-way command distribution is carried out through the communication link to urgently send heatstroke prevention and self-rescue strategies to the mobile terminals of field workers to guide them in carrying out initial self-protection, and to send emergency support strategies to the mobile terminals of the target support personnel to dispatch them to the site to carry out mutual assistance and rescue according to the guidance.
[0115] In one possible implementation, target support personnel are selected from multiple candidate online field personnel by comparing the candidate support risk index with a preset safety scheduling threshold. Specifically, this includes: selecting candidates whose candidate support risk index is less than the safety scheduling threshold from multiple candidate online field personnel to obtain a set of safe candidate personnel; obtaining the real-time order-taking load status and actual path distance from the real-time location coordinates of each candidate in the set of safe candidate personnel; selecting candidates whose real-time order-taking load status is not full from the set of safe candidate personnel to obtain a set of available candidate personnel; sorting the set of available candidate personnel in ascending order according to the actual path distance; and determining the available candidate personnel at the top of the sorted list as the target support personnel.
[0116] In this embodiment, the safety dispatch threshold refers to a pre-set critical numerical boundary used to measure whether the risk of heatstroke onset for potential rescuers is within a safe range. It aims to ensure that rescuers will not suffer secondary heatstroke due to additional physical exertion and heat exposure during their journey to the rescue. For example, the threshold can be set to 60 points corresponding to the real-time comprehensive heatstroke risk index. The real-time order load status is used to indicate the intensity of business tasks and capacity saturation of online field personnel. It reflects whether the personnel currently have the business redundancy to immediately interrupt their original delivery business to provide support. For example, the status can be divided into different levels of labels such as "idle", "normal transport" or "full load". The actual path distance refers to the physical trajectory length of the potential rescuer's current geographical location along the real passable road network to the location of the extremely high-risk personnel. It objectively reflects the spatial obstacles that the rescue force needs to overcome to actually reach the scene. For example, it is the actual cycling route distance planned by the navigation algorithm after considering one-way streets, overpasses and physical roadblocks.
[0117] Specifically, in dispatching and matching rescue forces, the first round of preliminary screening based on physiological safety is conducted. From the previously identified pool of online field personnel, those with a risk index strictly below the safety dispatch threshold are directly selected, thus excluding those already under high heat load from the rescue mission, resulting in a set of safe candidates who meet the physiological requirements. The next phase involves a multi-dimensional evaluation based on business availability and spatial proximity. By connecting to the backend order database and underlying map navigation service interface, the real-time order load status of each candidate in the safe candidate set, as well as their actual path distance from the real-time location coordinates of the heatstroke victim, are obtained. The second round of business interception screening is then performed. From the safe candidate set, those with a non-full order load status (i.e., no backlog of orders or very few orders, meeting the business conditions for suspending current delivery tasks to go to the rescue) are selected, eliminating fully loaded riders who cannot leave, resulting in a set of available candidates who are ready to provide rescue. To minimize the response time for heatstroke relief efforts, a low-level sorting algorithm is used to sort all available candidates in ascending order based on the calculated actual path distance, prioritizing the rescue force with the shortest physical route to the accident site. The first element of this sorted sequence is then directly selected and extracted, identifying the top-ranked candidate as the target support personnel. This achieves a closed-loop scheduling of rescue resources that balances physiological safety, operational feasibility, and spatial timeliness.
[0118] Furthermore, for the various preset benchmark tables (such as temperature mapping benchmark table and humidity mapping benchmark table) and various preset thresholds (such as interception threshold, risk classification threshold and security scheduling threshold) mentioned in the embodiments of this application, in specific implementation, they can be stored as underlying basic business dictionaries in a high-performance memory database (such as a Redis cluster) of the cloud management platform to support high-frequency and fast reading by the aforementioned risk assessment module.
[0119] Furthermore, the aforementioned benchmark tables and thresholds are not static constants; the system can also be configured with a dynamic threshold tuning submodule. This submodule can periodically (e.g., monthly or quarterly) connect to the latest regional heatstroke epidemiological database of the CDC, and based on the recent statistical characteristics of actual heatstroke incidence in the region, use machine learning algorithms to adaptively fine-tune the interception threshold and risk classification threshold. For example, if the system statistics find that a specific location has recently seen frequent reports of mild heat cramps at a risk score of 80, the system will automatically lower the extremely high risk classification threshold for that location from the global standard of 90 to 85, thereby achieving more agile and localized adaptive risk prevention and control.
[0120] Figure 3 This is a schematic diagram of a heatstroke prevention and control system based on workload, as described in an embodiment of this application. This system can be implemented through software, hardware, or a combination of both, forming all or part of the overall system. Figure 3 As shown, the system includes:
[0121] The coefficient determination module 301 is used to acquire the health record data of field workers and determine the individual health correction coefficient of the field workers based on the health record data.
[0122] The load forecasting module 302 is used to obtain environmental meteorological data of the target delivery area and business characteristic data of the target delivery order before dispatching the target delivery order to the field workers, and calculate the forward heat load forecast value based on the individual health correction coefficient, environmental meteorological data and business characteristic data.
[0123] The dispatch control module 303 is used to intercept the dispatch instruction for dispatching target delivery orders to field workers when the current heat load forecast value is greater than or equal to the preset interception threshold; when the current heat load forecast value is less than the interception threshold, the target delivery order is dispatched to the field workers. During the execution of the target delivery order by the field workers, the module receives the microenvironmental data of the current location of the field workers and the real-time physiological heat stress data uploaded by the data acquisition device worn by the field workers, and obtains the real-time work load data.
[0124] The index calculation module 304 is used to calculate the environmental heat load sub-index based on microenvironment data, calculate the individual physiological heat stress sub-index based on real-time physiological heat stress data, and calculate the work load sub-index based on real-time work load data.
[0125] The risk assessment module 305 is used to calculate and obtain the real-time comprehensive heatstroke risk index of field workers based on the environmental heat load sub-index, the individual physiological heat stress sub-index, the work load sub-index, and the individual health correction coefficient.
[0126] The strategy distribution module 306 is used to compare the real-time comprehensive heatstroke risk index with multiple preset risk classification thresholds in real time, and generate corresponding heatstroke risk management strategies based on the comparison results, and distribute them to the mobile terminals of field workers.
[0127] Based on the above embodiments, as an optional embodiment, the load forecasting module 302 is specifically used for: extracting temperature meteorological data and humidity meteorological data from environmental meteorological data; extracting order delivery distance and order cargo weight from business characteristic data; determining environmental severity weight based on temperature meteorological data and humidity meteorological data; determining order load weight based on order delivery distance and order cargo weight; and multiplying the environmental severity weight, order load weight, and individual health correction coefficient to obtain the forward heat load forecast value.
[0128] Based on the above embodiments, as an optional embodiment, the load prediction module 302 is specifically used to: obtain a preset temperature mapping benchmark table and a preset humidity mapping benchmark table; obtain a temperature influence base by comparing temperature meteorological data with the temperature mapping benchmark table; obtain a humidity influence base by comparing humidity meteorological data with the humidity mapping benchmark table; and obtain an environmental severity weight by adding the temperature influence base and the humidity influence base.
[0129] Based on the above embodiments, as an optional embodiment, the index calculation module 304 is specifically used to: extract local real-time temperature and local solar radiation intensity from microenvironment data, and determine the environmental heat load sub-index based on the local real-time temperature and local solar radiation intensity; extract body surface temperature and real-time heart rate from real-time physiological heat stress data, and determine the individual physiological heat stress sub-index based on the body surface temperature and real-time heart rate; extract the remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery from real-time work load data, and determine the work load sub-index based on the remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery.
[0130] Based on the above embodiments, as an optional embodiment, the index calculation module 304 is specifically used to: obtain preset mileage load weight, preset load weight, and preset floor load weight; multiply the remaining total delivery mileage by the mileage load weight to obtain the mileage energy consumption component; multiply the current total delivery load by the load weight to obtain the load energy consumption component; multiply the average number of floors climbed per delivery by the floor load weight to obtain the floor energy consumption component; and add the mileage energy consumption component, the load energy consumption component, and the floor energy consumption component to obtain the work load sub-index.
[0131] Based on the above embodiments, as an optional embodiment, the strategy distribution module 306 is specifically used for: obtaining the real-time location coordinates of field workers when the real-time comprehensive heatstroke risk index is greater than or equal to a preset extremely high risk classification threshold; determining multiple candidate online field workers within a preset distance range based on the real-time location coordinates; obtaining the candidate support risk index for each of the multiple candidate online field workers; selecting target support personnel from the multiple candidate online field workers by comparing the candidate support risk index with a preset safety scheduling threshold; generating a heatstroke prevention and self-rescue strategy for the corresponding field workers, as well as an emergency support strategy including real-time location coordinates and physical cooling first aid guidance; distributing the heatstroke prevention and self-rescue strategy to the mobile terminal of the field workers, and distributing the emergency support strategy to the mobile terminal of the target support personnel.
[0132] Based on the above embodiments, as an optional embodiment, the policy distribution module 306 is specifically used to: filter candidates whose support risk index is less than the safety scheduling threshold from multiple candidate online field personnel to obtain a set of safe candidates; obtain the real-time order-taking load status and actual path distance from the real-time location coordinates of each candidate in the set of safe candidates; filter candidates whose real-time order-taking load status is not full from the set of safe candidates to obtain a set of available candidates; sort the set of available candidates in ascending order according to the actual path distance; and determine the available candidate at the top of the sorted list as the target support personnel.
[0133] In a specific engineering implementation scheme, considering the characteristics of massive orders and high-concurrency data flow in field delivery operations, the above system embodiments and Figure 3Each functional module can be deployed on a cloud-based distributed server cluster based on a microservice architecture. Specifically, the index calculation module 304 can integrate a real-time stream processing engine to receive and process micro-environmental and physiological time-series data uploaded by tens of thousands of field workers in parallel with high throughput through message queues and sliding windows; the load prediction module 302 can be implemented as a microservice node equipped with a lightweight rule engine to achieve millisecond-level interception and verification of order load characteristics; the policy distribution module 306 can maintain full-duplex communication with the mobile terminals of field workers through a long connection protocol to ensure that when an extremely high risk is determined, heatstroke prevention and self-rescue strategies and emergency dispatch instructions can be pushed to the target location with extremely low latency of hundreds of milliseconds, thereby maximizing the seizing of the golden window for rescue.
[0134] Furthermore, regarding the preset benchmark tables (such as temperature mapping benchmark tables and humidity mapping benchmark tables) and preset thresholds (such as interception thresholds, risk classification thresholds, and safety scheduling thresholds) mentioned in the embodiments of this application, in specific implementations, they can be stored as underlying basic business dictionaries in the high-performance memory database of the cloud management platform to support high-frequency and fast reading by the aforementioned risk assessment module. Further, the aforementioned benchmark table and threshold system can also be configured with a threshold dynamic tuning submodule. This submodule can periodically connect to a regional heatstroke epidemiological database and, based on the recent statistical characteristics of actual heatstroke incidence in the region, use machine learning algorithms to adaptively fine-tune the interception thresholds and risk classification thresholds. For example, if the system statistics find that a specific location has frequently reported mild heat cramps at a specific risk score recently, the system will automatically lower the extremely high risk classification threshold for that location, thereby achieving more agile and localized adaptive risk prevention and control.
[0135] It should be noted that the above embodiments of the heatstroke prevention and control system based on workload are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different software or hardware modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same technical concept, and the specific implementation process and the resulting technical effects are detailed in the method embodiments, which will not be repeated here.
[0136] This embodiment also discloses an electronic device, which may specifically be a cloud server or a core scheduling center for implementing the aforementioned heatstroke prevention and control method. (Refer to...) Figure 4 The electronic device may include: at least one processor 401, at least one communication bus 402, user interface 403, network interface 404, and at least one memory 405.
[0137] The communication bus 402 is used to enable communication between these components. The network interface 404 is mainly used to maintain real-time communication with the mobile terminals and wearable data acquisition devices of field workers, such as receiving microenvironmental data and real-time physiological heat stress data uploaded at high frequency by the data acquisition devices, and issuing instructions or risk management strategies to the mobile terminals of field workers.
[0138] The memory 405 may include a program storage area and a data storage area. In this embodiment, the data storage area is primarily used to securely store health record data of field workers, environmental meteorological data of the target delivery area, business characteristic data of the target delivery orders, as well as underlying business data such as preset temperature mapping benchmark tables, humidity mapping benchmark tables, various load weights (such as mileage, load weight, floor load weight) and risk classification thresholds.
[0139] The processor 401 executes the core algorithms and control logic of this application by running or executing instructions, programs, code sets, or instruction sets stored in the memory 405, and by calling data stored in the memory 405. Specifically, the processor 401 executes instructions to complete the following core actions: determining individual health correction coefficients based on health record data; calculating the predicted heat load value to determine whether to intercept the dispatch instruction; calculating the environmental heat load sub-index, the individual physiological heat stress sub-index, and the work load sub-index during order execution; and finally, integrating and calculating the real-time comprehensive heatstroke risk index, and generating a heatstroke risk management strategy and intelligently screening and scheduling target support personnel based on the comparison results.
[0140] In actual field operations, considering special circumstances where field workers may enter basements, elevators, or other locations that could cause mobile terminal network connection interruptions, the electronic devices and wearable data acquisition devices (i.e., edge devices) of this application also have built-in edge computing degradation strategies. When the mobile terminal detects a network communication loss with the cloud server (intelligent management and control platform), the local processor within the mobile terminal will take over some of the functions of the "index calculation module." Specifically, the wearable data acquisition device will automatically cache real-time collected physiological heat stress data such as body surface temperature and heart rate in local memory; simultaneously, based on the cached last known weather data and local sensor data, the mobile terminal uses a simplified linear weighted formula to calculate a locally estimated physiological heat load warning value. Once this locally estimated value exceeds the limit, the mobile terminal can issue a local heatstroke prevention and self-rescue strategy (such as a forced voice broadcast "Abnormal vital signs detected, please stop delivery immediately and go to a shady place") directly through strong vibration or maximum volume broadcast at the operating system level, without relying on cloud commands. Once the network link is restored, the data collection device will then retransmit the underlying data recorded during the network outage to the cloud server to complete the long-term health record review, thereby improving the reliability of life safety prevention and control for field personnel in extremely weak network environments.
[0141] This application also provides a computer-readable storage medium storing a computer program or instructions. When the computer program or instructions are executed by a computer (such as the processor of the aforementioned electronic device), the computer is able to implement the various steps of the workload-based heatstroke prevention method in any of the above method embodiments. Specifically, the computer-readable storage medium may include, but is not limited to, various physical media capable of persistently or non-persistently storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
Claims
1. A method for preventing and controlling heatstroke based on workload, characterized in that, The method includes: Obtain health record data of field workers, and determine the individual health correction coefficient of the field workers based on the health record data; Before dispatching the target delivery order to the field workers, environmental meteorological data of the target delivery area and business characteristic data of the target delivery order are obtained. Based on the individual health correction coefficient, the environmental meteorological data and the business characteristic data, the predicted value of the heat load is calculated. When the predicted value of the preheating load is greater than or equal to the preset interception threshold, the dispatch instruction to send the target delivery order to the field workers is intercepted. When the predicted value of the preheating load is less than the interception threshold, the target delivery order is dispatched to the field worker. During the execution of the target delivery order by the field worker, the microenvironmental data of the field worker's current location and real-time physiological heat stress data are uploaded by the data collection device worn by the field worker, and real-time work load data is obtained. The environmental heat load sub-index is calculated based on the microenvironment data, the individual physiological heat stress sub-index is calculated based on the real-time physiological heat stress data, and the work load sub-index is calculated based on the real-time work load data. Based on the environmental heat load sub-index, the individual physiological heat stress sub-index, the work load sub-index, and the individual health correction coefficient, the real-time comprehensive heatstroke risk index of the field workers is calculated and obtained. The system compares the real-time comprehensive heatstroke risk index with multiple preset risk classification thresholds in real time, and generates corresponding heatstroke risk management strategies based on the comparison results, which are then sent to the mobile terminals of the field workers.
2. The method according to claim 1, characterized in that, The calculation of the predicted heat load based on the individual health correction coefficient, the environmental meteorological data, and the business characteristic data specifically includes: Temperature and humidity data are extracted from the environmental meteorological data. Extract order delivery distance and order weight from the aforementioned business characteristic data; The environmental severity weight is determined based on the temperature and humidity meteorological data. The order load weight is determined based on the order delivery distance and the order goods weight; The environmental severity weight, the order load weight, and the individual health correction coefficient are multiplied together to obtain the predicted forward heat load value.
3. The method according to claim 2, characterized in that, The determination of environmental severity weights based on the temperature and humidity meteorological data specifically includes: Obtain the preset temperature mapping reference table and the preset humidity mapping reference table; The temperature influence baseline is obtained by comparing the temperature meteorological data with the temperature mapping benchmark table; By comparing the humidity meteorological data with the humidity mapping benchmark table, the humidity influence base is obtained; The environmental severity weight is obtained by adding the temperature influence base number and the humidity influence base number.
4. The method according to claim 1, characterized in that, The calculation of the environmental heat load sub-index based on the microenvironment data, the calculation of the individual physiological heat stress sub-index based on the real-time physiological heat stress data, and the calculation of the work load sub-index based on the real-time work load data specifically include: Local real-time temperature and local solar radiation intensity are extracted from the microenvironment data, and the environmental heat load sub-index is determined based on the local real-time temperature and the local solar radiation intensity. Body surface temperature and real-time heart rate are extracted from the real-time physiological heat stress data, and the individual physiological heat stress sub-index is determined based on the body surface temperature and the real-time heart rate. The remaining total delivery mileage, current total delivery load, and average number of floors climbed per delivery are extracted from the real-time work load data. Based on the remaining total delivery mileage, the current total delivery load, and the average number of floors climbed per delivery, the work load sub-index is determined.
5. The method according to claim 4, characterized in that, The determination of the workload sub-index based on the remaining total delivery mileage, the current total delivery load, and the average number of floors climbed per delivery specifically includes: Obtain the preset mileage load weight, preset heavy load weight, and preset floor load weight; Multiply the remaining total delivery mileage by the mileage load weight to obtain the mileage energy consumption component; Multiply the current total delivery load by the load weight to obtain the load energy consumption component; Multiply the average number of floors climbed per delivery by the floor load weight to obtain the floor energy consumption component; The work load sub-index is obtained by adding the mileage energy consumption component, the load energy consumption component, and the floor energy consumption component.
6. The method according to claim 1, characterized in that, The process of generating corresponding heatstroke risk management strategies based on the comparison results and distributing them to the mobile terminals of the field workers specifically includes: When the real-time comprehensive heatstroke risk index is greater than or equal to the preset extremely high risk classification threshold, the real-time location coordinates of the field workers are obtained. Based on the real-time location coordinates, multiple candidate online field personnel located within a preset distance range are identified; Obtain the candidate support risk index for each of the multiple candidate online field personnel; By comparing the candidate support risk index with the preset security scheduling threshold, target support personnel are selected from the multiple candidate online field personnel. Generate heatstroke prevention and self-rescue strategies for the field workers, as well as emergency support strategies that include the real-time location coordinates and physical cooling first aid guidelines; The heatstroke prevention and self-rescue strategy is sent to the mobile terminal of the field workers, and the emergency support strategy is sent to the mobile terminal of the target support personnel.
7. The method according to claim 6, characterized in that, The step of selecting target support personnel from a plurality of candidate online field personnel by comparing the candidate support risk index with a preset security scheduling threshold specifically includes: From the multiple candidate online field personnel, candidates whose candidate support risk index is less than the security scheduling threshold are selected to obtain a set of security candidates; Obtain the real-time order-acceptance load status and the actual path distance from the real-time location coordinates for each candidate in the set of safety candidates; From the set of safety candidates, candidates whose real-time order-taking load status is not full are selected to obtain a set of available candidates. Sort the set of available candidates in ascending order based on the actual path distance; The top-ranked available candidate is identified as the target support personnel.
8. A heatstroke prevention and control system based on workload, characterized in that, The system includes: The coefficient determination module is used to acquire the health record data of field workers and determine the individual health correction coefficient of the field workers based on the health record data. The load forecasting module is used to obtain environmental meteorological data of the target delivery area and business characteristic data of the target delivery order before dispatching the target delivery order to the field workers, and calculate the advance heat load forecast value based on the individual health correction coefficient, the environmental meteorological data and the business characteristic data; The dispatch control module is used to intercept the dispatch instruction for dispatching the target delivery order to the field worker when the predicted heat load value is greater than or equal to a preset interception threshold; when the predicted heat load value is less than the interception threshold, the target delivery order is dispatched to the field worker; and during the execution of the target delivery order by the field worker, the module receives microenvironmental data of the current location of the field worker and real-time physiological heat stress data uploaded by the data acquisition device worn by the field worker, and obtains real-time work load data. The index calculation module is used to calculate the environmental heat load sub-index based on the microenvironment data, calculate the individual physiological heat stress sub-index based on the real-time physiological heat stress data, and calculate the work load sub-index based on the real-time work load data. The risk assessment module is used to calculate and obtain the real-time comprehensive heatstroke risk index of the field workers based on the environmental heat load sub-index, the individual physiological heat stress sub-index, the work load sub-index, and the individual health correction coefficient. The strategy distribution module is used to compare the real-time comprehensive heatstroke risk index with multiple preset risk classification thresholds in real time, and generate corresponding heatstroke risk management strategies based on the comparison results, and distribute them to the mobile terminals of the field workers.
9. An electronic device, characterized in that, The device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. The user interface and the network interface are both used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-7.
