Intelligent logistics emergency risk early warning system

Abstract

The invention discloses an intelligent logistics emergency risk early warning system, which relates to the technical field of intelligent logistics, and is characterized in that real-time data acquisition is carried out by deploying Internet of Things equipment at each node of a logistics chain, and the data is transmitted to a data center by using a wireless communication technology, so that the continuity and the real-time performance of logistics data are ensured; the data analysis module compares real-time data with historical data and monitors the operation state and the performance stability of the Internet of Things equipment, and the comprehensive analysis module further evaluates data differences and equipment performance and judges the decision dependence accuracy when the system deals with emergencies; the prediction and early warning module predicts and warns the risk in advance when the decision accuracy is low, and automatically executes the emergency decision when the accuracy is high, thereby effectively improving the stability, risk response capability and overall operation efficiency of the logistics system, and reducing decision errors caused by equipment faults or data interruption.

Description

Technical Field

[0001] The present invention relates to the field of intelligent logistics technology, and in particular to an intelligent logistics emergency risk early warning system. Background Art

[0002] Intelligent logistics emergency risk warning refers to the use of advanced intelligent technology to monitor, analyze and warn potential risks in the logistics process in real time. By integrating the Internet of Things, big data, artificial intelligence and other technical means, the system can automatically collect and process massive information related to the logistics process, such as traffic conditions, weather changes, supply chain disruptions, etc., so as to identify risks that may affect logistics operations in advance and issue alarms in time. The purpose of the system is to provide risk warnings to logistics companies in advance before emergencies (such as natural disasters, traffic accidents, supply chain failures, etc.) occur, so as to help them respond quickly, adjust transportation plans, reallocate resources or choose alternative routes. This intelligent emergency management mechanism not only improves the overall efficiency of the logistics system, but also effectively reduces the impact of emergencies on the supply chain. However, if the core Internet of Things devices or artificial intelligence algorithms fail, the entire logistics early warning system may crash or misreport. For example, sensor failure or network connection interruption may cause key data to fail to be transmitted in time, resulting in the company's inability to make accurate judgments in critical moments and miss the best time to deal with risks. It may also reduce the ability of manual emergency response, making the company more vulnerable when technology fails. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent logistics emergency risk warning system to solve the shortcomings of the background technology.

[0004] In order to achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent logistics emergency risk warning system, including a data acquisition module, a data analysis module, a comprehensive analysis module and a prediction and warning module;

[0005] Data collection module: IoT devices are deployed at various nodes in the logistics chain to monitor data in real time and transmit it to the data center through wireless communication technology;

[0006] Data analysis module: compares the real-time data in the logistics process with the historical data, determines the difference of the real-time data based on the comparison results, monitors the operating status of the IoT devices in real time, and analyzes the performance stability of the IoT devices;

[0007] Comprehensive analysis module: conducts comprehensive analysis on the differences between real-time data and historical data in the logistics process and the performance stability of IoT devices to determine the accuracy of the decision-making reliance of the system in response to emergencies;

[0008] Prediction and warning module: When the accuracy of the decision-making reliance of the system in response to emergencies is low, the risk of system decisions within a fixed time period will be predicted and warned in advance. When the accuracy of the decision-making reliance of the system in response to emergencies is high, the warning decision will be automatically executed.

[0009] Preferably, in the data analysis module, the real-time data is matched with the historical data item by item according to the preset comparison dimension, and the abnormal data accumulation index is generated according to the comparison result to determine the difference of the real-time data. The method for obtaining the abnormal data accumulation index is:

[0010] Get the value D of the i-th data point at time t in the real-time data set t,i , and the reference value H of the i-th data point in the historical data set i ; Set a fixed time interval T, obtain the total number of data points N collected within the time interval T, and calculate the difference ΔD between the i-th real-time data and its historical reference value t,i , the calculation expression is: ΔD t,i =|D t,i -H i |; Each ΔD t,i Compare with the preset threshold θ to determine whether the data point is abnormal: If ΔD t,i >θ, then the data point ΔD t,i is considered anomaly; for each abnormal data point, a weight w is applied i To reflect the importance of different types of data, the total outlier value A i It is calculated by weighted accumulation of the difference values ​​of all abnormal data, and the expression is: A i =w i ·(ΔD t,i -θ); within the time interval T, calculate the cumulative value of all abnormal data points, that is, calculate the abnormal data cumulative index AT, the expression is:

[0011] Preferably, in the data analysis module, a signal strength deviation index is generated according to the quality change of the network connection of the IoT device within the H time period to determine the performance stability of the IoT device. The signal strength deviation index is obtained by:

[0012] Collect signal strength data S of IoT devices during the H period t , calculate the average signal strength of the device in the H time period The formula is: Among them, S t is the signal strength collected at time point t, N is the total number of collections, and the signal strength S at each time point t is calculated. t With the mean The difference ΔS t, expressed as: The difference value is used to measure the fluctuation range of the signal strength at each time point relative to the overall level, and obtain the standard range of signal strength fluctuation under normal working conditions of IoT devices in historical data: S min ~S max ; If ΔS t Greater than S max , the calculation expression of signal strength deviation index is: zx=|ΔS t -S max |; If ΔS t Less than S min , the calculation expression of signal strength deviation index is: zx=|ΔS t -S min |; where zx is the signal strength deviation index.

[0013] Preferably, in the comprehensive analysis module, a comprehensive analysis is performed on the difference between real-time data and historical data in the logistics link and the performance stability of the IoT devices to determine the accuracy of the decision-making reliance of the system when dealing with emergencies;

[0014] The abnormal data accumulation index and the signal strength deviation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the accuracy value label of the decision-making dependence of the system in response to emergencies as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the decision-making dependence of all systems in response to emergencies as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy value of the decision-making dependence of the system in response to emergencies is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0015] Preferably, the acquired accuracy value of the decision-making reliance of the system in response to an emergency is compared with a preset reference threshold of the accuracy value. If the accuracy value of the decision-making reliance of the system in response to an emergency is greater than or equal to the reference threshold of the accuracy value, then the accuracy of the decision-making reliance of the system in response to an emergency is high, and an accurate signal is generated at this time, and it is classified as a high-accuracy decision-making reliance; if the accuracy value of the decision-making reliance of the system in response to an emergency is less than the reference threshold of the accuracy value, then the accuracy of the decision-making reliance of the system in response to an emergency is low, and an inaccurate signal is generated at this time, and it is classified as a low-accuracy decision-making reliance.

[0016] Preferably, in the prediction and warning module, when the accuracy of the decision-making reliance of the system in response to an emergency is low, an early prediction and warning of the risk of the system decision within a fixed time period is performed;

[0017] When the accuracy of the decision-making dependence of the system in response to emergencies is low, that is, the accuracy value of the decision-making dependence of the system in response to emergencies generated within a fixed time period is less than the reference threshold of the accuracy value, the accuracy values ​​of the decision-making dependence of the system in response to emergencies generated within a subsequent fixed time period that are less than the reference threshold of the accuracy value are collected, the accuracy values ​​in the data set are compared and analyzed with the reference threshold of the accuracy value, the abnormal index of the accuracy value is calculated, that is, the standard deviation of the data set is calculated, the risk of the system decision is predicted, and early warning is issued.

[0018] Preferably, if the calculated abnormality index is greater than or equal to the preset reference threshold of the abnormality index, it means that the risk of the system decision is high, and a warning signal is generated at this time; if the calculated abnormality index is less than the preset reference threshold of the abnormality index, it means that the risk of the system decision is low, and no warning signal is generated at this time.

[0019] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0020] 1. The present invention collects data in real time by deploying IoT devices at each node of the logistics chain, and transmits it to the data center using wireless communication technology. The data analysis module monitors the operating status and performance stability of IoT devices by comparing real-time data with historical data. The comprehensive analysis module determines the accuracy of the decision-making dependence of the system in response to emergencies based on data differences and device status. The prediction and warning module predicts risks in advance and generates warnings when the decision accuracy is low, and automatically executes emergency decisions when the accuracy is high, thereby ensuring the stability and efficiency of the logistics system.

[0021] 2. The present invention can quickly identify system anomalies and issue early warnings through the analysis of abnormal data accumulation index and signal strength deviation index, combined with the accuracy of machine learning model prediction system decision dependence, thereby reducing false alarms or system crashes caused by equipment failure or network interruption. When the risk of system decision is high, an early warning signal is automatically generated to ensure timely response to emergencies; when the risk is low, the system automatically executes the decision, improves the emergency response capability of the logistics system, and enhances the robustness and operational stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0023] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] For examples, see Figure 1 As shown, the intelligent logistics emergency risk warning system described in this embodiment includes a data collection module, a data analysis module, a comprehensive analysis module and a prediction and warning module;

[0026] Data collection module: IoT devices are deployed at various nodes in the logistics chain to monitor data in real time and transmit it to the data center through wireless communication technology;

[0027] Data analysis module: compares the real-time data in the logistics process with the historical data, determines the difference of the real-time data based on the comparison results, monitors the operating status of the IoT devices in real time, and analyzes the performance stability of the IoT devices;

[0028] Comprehensive analysis module: conducts comprehensive analysis on the differences between real-time data and historical data in the logistics process and the performance stability of IoT devices to determine the accuracy of the decision-making reliance of the system in response to emergencies;

[0029] Prediction and warning module: When the accuracy of the decision-making reliance of the system in response to emergencies is low, the risk of system decisions within a fixed time period will be predicted and warned in advance. When the accuracy of the decision-making reliance of the system in response to emergencies is high, the warning decision will be automatically executed.

[0030] Among them, in the data acquisition module, IoT devices are deployed at various nodes in the logistics chain to monitor the data in the logistics links in real time and transmit them to the data center through wireless communication technology.

[0031] Deploy IoT devices at each node of the logistics chain and use wireless communication technology to monitor and transmit data to the data center in real time:

[0032] First, identify the most important nodes in the logistics chain and ensure that the deployment of equipment can cover the key parts of the entire logistics process. These nodes usually include: Warehousing center: where goods are stored and managed, monitoring temperature, humidity, inventory status, etc. Transport vehicles: the location of vehicles, the safety of goods, etc. need to be tracked in real time. Sorting center: the place where goods are transferred, monitoring the handling of goods and circulation efficiency. Distribution point: the place where goods are finally delivered, tracking the arrival time and delivery status of goods.

[0033] Select the corresponding IoT devices according to the needs of different nodes. For example: RFID tags: used for identification and tracking of goods, capable of reading cargo information. Sensors (temperature, humidity, vibration, etc.): used to monitor the environmental conditions of goods, especially during storage and transportation, to ensure the safety of goods. GPS devices: used to track the real-time location of vehicles and goods, facilitating monitoring of transportation routes. Cameras: installed in warehouses and distribution points to monitor the loading and unloading and storage environment of goods.

[0034] Install the selected equipment at key locations of logistics nodes to ensure the coverage and accuracy of data collection. Warehousing center: Install sensors on shelves and storage areas to monitor environmental data (such as temperature, humidity, air quality, etc.) in real time. Transport vehicles: Install GPS devices and sensors in vehicles to track vehicle location information and the status of goods in real time. Sorting center: Deploy RFID readers in the goods circulation area to ensure the identification and tracking of each piece of goods. Distribution point: Install cameras at the delivery site or use handheld devices to scan RFID to ensure final confirmation of the status of the goods.

[0035] Configure appropriate wireless communication methods for IoT devices to ensure that each device can transmit data to the data center in real time. Common communication methods include: Wi-Fi: Applicable to fixed locations such as storage centers and distribution points, providing a stable data transmission environment. 4G / 5G: Applicable to mobile nodes such as transport vehicles, and real-time tracking of vehicle location and cargo status through cellular networks. LoRa, NB-IoT: Applicable to remote and low-power environments, such as in dispersed storage locations or long-distance transport vehicles, ensuring low-power remote communication. Bluetooth / Wireless LAN: Applicable to short-distance data transmission, such as using RFID data transmission in sorting centers.

[0036] IoT devices begin to continuously collect relevant data in logistics nodes. Specific data include: Environmental data: temperature, humidity, vibration, air quality, etc., especially important for monitoring fragile goods. Location data: Track the real-time location of vehicles and goods through GPS to ensure the visualization of transportation routes. Cargo status data: Collect the unique identity information, location changes and processing status of goods through RFID or other sensor devices. Image data: The camera collects image or video information regularly or when triggered by specific events to assist in the management of cargo loading and unloading.

[0037] The collected data is transmitted to the data center in real time for processing through the wireless communication network. The specific steps are as follows: the IoT device transmits the data to the central server or cloud platform in real time through the configured network (Wi-Fi, 5G, LoRa, etc.). Ensure that the data is encrypted during transmission to prevent sensitive information from being illegally accessed or tampered with, especially for data such as transportation routes and cargo status. When the network signal is unstable or interrupted, the device should have the function of caching data. Once the network is restored, the cached data will be automatically uploaded to avoid data loss.

[0038] Data analysis module: compares the real-time data in the logistics process with the historical data, determines the difference in the real-time data based on the comparison results, monitors the operating status of the IoT devices in real time, and analyzes the performance stability of the IoT devices.

[0039] Retrieve historical data corresponding to the current logistics link from the system. Historical data usually includes logistics information under similar conditions (such as the same route, time period, type of goods, etc.). Filter and classify historical data to ensure that it is relevant to the current real-time data. For example, if the status of goods on a specific transportation route is being monitored, it is necessary to retrieve past transportation records on that route. Preprocess real-time and historical data to eliminate possible noise, outliers, or erroneous data. For example, remove missing data caused by incorrect sensor readings or network interruptions.

[0040] According to the preset comparison dimension, the real-time data is matched with the historical data item by item, and the abnormal data accumulation index is generated according to the comparison result to determine the difference of the real-time data. The method for obtaining the abnormal data accumulation index is as follows:

[0041] Get the value D of the i-th data point at time t in the real-time data set t,i , and the reference value H of the i-th data point in the historical data set i ; Set a fixed time interval T, obtain the total number of data points N collected within the time interval T, and calculate the difference ΔD between the i-th real-time data and its historical reference value t,i , the calculation expression is: ΔD t,i =|D t,i -H i |; Each ΔD t,i Compare with the preset threshold θ to determine whether the data point is abnormal: If ΔD t,i >θ, then the data point ΔD t,i is considered anomaly; for each abnormal data point, a weight w is applied i To reflect the importance of different types of data. Total outlier value A i It is calculated by weighted accumulation of the difference values ​​of all abnormal data, and the expression is: A i =wi ·(ΔD t,i -θ); within the time interval T, calculate the cumulative value of all abnormal data points, that is, calculate the abnormal data cumulative index AT, the expression is:

[0042] When the abnormal data accumulation index is larger, it means that the difference between real-time data and historical data is becoming more and more obvious, and this difference has accumulated to a high degree. This means that abnormal situations in the logistics system occur frequently, and the degree of deviation is large each time. It may be equipment failure, environmental changes (such as abnormal temperature or humidity), or other factors that affect the stability of the logistics chain. A larger abnormal data accumulation index indicates that the system needs to intervene quickly to prevent potential problems from further deteriorating.

[0043] On the contrary, when the abnormal data accumulation index is smaller, it means that the difference between real-time data and historical data is smaller, or although a small number of abnormalities have occurred, their degree is not large and has not accumulated to a significant level. This shows that the operation of the logistics system is relatively stable, the equipment, environment and process are roughly within the normal range, and the overall operation of the system is in line with historical expectations. A smaller index means that the risk in the current logistics link is low and no immediate intervention measures are required.

[0044] It also monitors the operating status of IoT devices in real time, analyzes the performance stability of IoT devices, monitors the quality and stability of network connections, ensures that devices can continuously transmit data, and avoids data interruptions due to poor signals.

[0045] The signal strength deviation index is generated according to the quality change of the network connection of the IoT device in the H time period to determine the performance stability of the IoT device. The method for obtaining the signal strength deviation index is:

[0046] Collect signal strength data S of IoT devices during the H period t , for example, once every minute or every second. The data can be obtained through the device's wireless module, usually expressed in dBm. Calculate the average signal strength of the device in the H time period The formula is: Among them, S t is the signal strength collected at time point t, N is the total number of collections, and the signal strength S at each time point t is calculated. t With the mean The difference ΔS t , expressed as: The difference value is used to measure the fluctuation range of signal strength relative to the overall level at each time point, and obtain the standard range of signal strength fluctuation under normal working conditions of IoT devices in historical data: S min ~S max ; If ΔS t Greater than Smax , the calculation expression of signal strength deviation index is: zx=|ΔS t -S max |; If ΔS t Less than S min , the calculation expression of signal strength deviation index is: zx=|ΔS t -S min |; where zx is the signal strength deviation index.

[0047] When the signal strength deviation index is larger, it means that the signal strength of the IoT device fluctuates more frequently and with greater amplitude, which means that the network connection of the device is unstable and the signal may be frequently weakened or interrupted. This instability can cause data transmission delays, packet loss, or connection interruptions, which will affect the real-time monitoring and operation capabilities of the device in the logistics system and reduce the overall performance stability. Frequent signal fluctuations may also increase the power consumption of the device because the device may need to continuously reconnect to the network or retransmit data.

[0048] On the contrary, the smaller the signal strength deviation index is, the more stable the signal strength of the device is, the less fluctuation is, and the more reliable the network connection is. This means that IoT devices can continue to maintain efficient communication, and the integrity and real-time performance of data transmission are better, thereby improving the performance stability of the device. Stable signal strength reduces the risk of network interruption, ensures that the device can continuously monitor and transmit data, optimizes the operational efficiency of the logistics chain and reduces potential operational risks.

[0049] Comprehensive analysis module: conducts a comprehensive analysis of the differences between real-time data and historical data in the logistics process and the performance stability of IoT devices to determine the accuracy of the decision-making reliance of the system when responding to emergencies.

[0050] The abnormal data accumulation index and the signal strength deviation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the accuracy value label of the decision-making dependence of the system in response to emergencies as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the decision-making dependence of all systems in response to emergencies as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy value of the decision-making dependence of the system in response to emergencies is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0051] The method for obtaining the accuracy value of the decision-making dependence when the system responds to emergencies is as follows: from the first eigenvector training data of the trained machine learning model, obtain the corresponding function expression: CM=F(AT,zx); wherein F is the output function of the model, AT is the abnormal data accumulation index, zx is the signal strength deviation index, and CM is the accuracy value of the decision-making dependence when the system responds to emergencies.

[0052] The acquired accuracy value of the decision-making dependency of the system in response to emergencies is compared with the preset reference threshold of the accuracy value. If the accuracy value of the decision-making dependency of the system in response to emergencies is greater than or equal to the reference threshold of the accuracy value, then the accuracy of the decision-making dependency of the system in response to emergencies is high, and an accurate signal is generated at this time, and it is classified as a high-accuracy decision-making dependency; if the accuracy value of the decision-making dependency of the system in response to emergencies is less than the reference threshold of the accuracy value, then the accuracy of the decision-making dependency of the system in response to emergencies is low, and an inaccurate signal is generated at this time, and it is classified as a low-accuracy decision-making dependency.

[0053] Prediction and warning module: When the accuracy of the decision-making reliance of the system in response to emergencies is low, the risk of system decisions within a fixed time period will be predicted and warned in advance. When the accuracy of the decision-making reliance of the system in response to emergencies is high, the warning decision will be automatically executed.

[0054] When the accuracy of the decision-making dependence of the system in response to emergencies is low, that is, the accuracy value of the decision-making dependence of the system in response to emergencies generated within a fixed time period is less than the reference threshold of the accuracy value, the accuracy values ​​of the decision-making dependence of the system in response to emergencies generated within a subsequent fixed time period that are less than the reference threshold of the accuracy value are collected, the accuracy values ​​in the data set are compared and analyzed with the reference threshold of the accuracy value, the abnormal index of the accuracy value is calculated, that is, the standard deviation of the data set is calculated, the risk of the system decision is predicted, and early warning is issued.

[0055] If the calculated abnormality index is greater than or equal to the preset reference threshold of the abnormality index, it means that the risk of the system decision is high, and a warning signal is generated at this time; if the calculated abnormality index is less than the preset reference threshold of the abnormality index, it means that the risk of the system decision is low, and no warning signal is generated at this time.

[0056] When the system relies on high accuracy in decision-making when responding to emergencies, the system can automatically execute early warning decisions. At this time, based on the high-accuracy prediction of the model, the system will automatically adjust the logistics process according to the preset emergency plan, re-plan the transportation route, allocate resources, or take other preventive measures to ensure the smooth operation of the logistics chain. This automated decision-making reduces the need for manual intervention, improves the response speed and decision-making efficiency in emergencies, thereby effectively reducing risks and ensuring the stability of the system.

[0057] In this embodiment, real-time data collection is performed by deploying IoT devices at each node of the logistics chain, and the data is transmitted to the data center using wireless communication technology. The data analysis module compares real-time data with historical data, monitors the operating status of IoT devices, and analyzes their performance stability. The comprehensive analysis module further analyzes data differences and equipment performance to determine the accuracy of decision-making dependencies when the system responds to emergencies. The prediction and warning module predicts risks in advance and generates warnings when the decision accuracy is low, and automatically executes emergency decisions when the decision accuracy is high, ensuring the efficient operation of the logistics system.

[0058] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0059] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0060] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0061] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An intelligent logistics emergency risk early warning system, characterized by: It includes data collection module, data analysis module, comprehensive analysis module and prediction and early warning module; Data collection module: IoT devices are deployed at each node of the logistics chain to monitor data in real time and transmit it to the data center through wireless communication technology; Data analysis module: compares the real-time data in the logistics process with the historical data, determines the difference of the real-time data based on the comparison results, monitors the operating status of the IoT devices in real time, and analyzes the performance stability of the IoT devices; Comprehensive analysis module: conducts comprehensive analysis on the differences between real-time data and historical data in the logistics process and the performance stability of IoT devices to determine the accuracy of the decision-making reliance of the system in response to emergencies; Prediction and warning module: When the accuracy of the decision-making reliance of the system in response to emergencies is low, the risk of system decisions within a fixed time period will be predicted and warned in advance. When the accuracy of the decision-making reliance of the system in response to emergencies is high, the warning decision will be automatically executed.

2. According to claim 1, the intelligent logistics emergency risk warning system is characterized by: In the data analysis module, according to the preset comparison dimension, the real-time data is matched with the historical data item by item, and the abnormal data accumulation index is generated according to the comparison result to determine the difference of the real-time data. The method for obtaining the abnormal data accumulation index is as follows: Get the value D of the i-th data point at time t in the real-time data set t，i , and the reference value H of the i-th data point in the historical data set i ; Set a fixed time interval T, obtain the total number of data points N collected within the time interval T, and calculate the difference △D between the i-th real-time data and its historical reference value t，i , the calculation expression is: △D t，i =|D t，i -H i |; Each △D t，i Compare with the preset threshold θ to determine whether the data point is abnormal: If △D t，i >θ, then the data point △D t，i is considered abnormal; For each abnormal data point, apply the weight w i To reflect the importance of different types of data, the total outlier value A i It is calculated by weighted accumulation of the difference values ​​of all abnormal data, and the expression is: A i =w i ·(△D t，i -θ); within the time interval T, calculate the cumulative value of all abnormal data points, that is, calculate the abnormal data cumulative index AT, the expression is:

3. The intelligent logistics emergency risk early warning system according to claim 2 is characterized by: In the data analysis module, the signal strength deviation index is generated according to the quality change of the network connection of the IoT device in the H time period to determine the performance stability of the IoT device. The signal strength deviation index is obtained as follows: Collect signal strength data S of IoT devices during the H period t , calculate the average signal strength of the device in the H time period The formula is: Among them, S t is the signal strength collected at time point t, N is the total number of collections, and the signal strength S at each time point t is calculated. t With the mean The difference △S t , expressed as: The difference value is used to measure the fluctuation range of the signal strength at each time point relative to the overall level, and obtain the standard range of signal strength fluctuation under normal working conditions of IoT devices in historical data: S min ~S max ; If △S t Greater than S max , the calculation expression of signal strength deviation index is: zx=|△S t -S max |; If △S t Less than S min , the calculation expression of signal strength deviation index is: zx=|△S t -S min |; where zx is the signal strength deviation index.

4. The intelligent logistics emergency risk early warning system according to claim 3 is characterized by: In the comprehensive analysis module, the differences between real-time data and historical data in the logistics process and the performance stability of IoT devices are comprehensively analyzed to determine the accuracy of the decision-making reliance of the system when dealing with emergencies; The abnormal data accumulation index and the signal strength deviation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the accuracy value label of the decision-making dependence of the system in response to emergencies as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the decision-making dependence of all systems in response to emergencies as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy value of the decision-making dependence of the system in response to emergencies is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

5. The intelligent logistics emergency risk warning system according to claim 4 is characterized by: The obtained accuracy value of the decision dependency of the system in response to an emergency is compared with a preset reference threshold of the accuracy value. If the accuracy value of the decision dependency of the system in response to an emergency is greater than or equal to the reference threshold of the accuracy value, the accuracy of the decision dependency of the system in response to an emergency is high, and an accurate signal is generated at this time, and it is classified as a high-accuracy decision dependency; If the accuracy value of the decision dependency of the system when responding to an emergency is less than the reference threshold of the accuracy value, the accuracy of the decision dependency of the system when responding to an emergency is low. At this time, an inaccurate signal is generated and classified as a low-accuracy decision dependency.

6. The intelligent logistics emergency risk early warning system according to claim 1 is characterized by: In the prediction and warning module, when the accuracy of the decision-making reliance of the system in response to emergencies is low, the risk of system decisions within a fixed time period is predicted and warned in advance; When the accuracy of the decision-making dependence of the system in response to emergencies is low, that is, the accuracy value of the decision-making dependence of the system in response to emergencies generated within a fixed time period is less than the reference threshold of the accuracy value, the accuracy values ​​of the decision-making dependence of the system in response to emergencies generated within a subsequent fixed time period that are less than the reference threshold of the accuracy value are collected, the accuracy values ​​in the data set are compared and analyzed with the reference threshold of the accuracy value, the abnormal index of the accuracy value is calculated, that is, the standard deviation of the data set is calculated, the risk of the system decision is predicted, and early warning is issued.

7. The intelligent logistics emergency risk early warning system according to claim 6 is characterized by: If the calculated abnormality index is greater than or equal to the preset reference threshold of the abnormality index, it means that the risk of the system decision is high, and a warning signal is generated at this time; if the calculated abnormality index is less than the preset reference threshold of the abnormality index, it means that the risk of the system decision is low, and no warning signal is generated at this time.

Images