Container logistics safety monitoring and early warning platform based on the Internet of Things

Abstract

The present invention discloses a container logistics safety monitoring and early warning platform based on the Internet of Things, which relates to the field of logistics information technology. The method includes: firstly interacting with the relevant information of logistics tasks to determine the monitoring needs, then mining the logistics risk event chain, then building a multi-step decision model, and then monitoring the risk event chain to obtain data and transmit it back. After receiving the data, the logistics management center determines the decision matrix and probability, determines the probability over-limit, and locates the risk list if it is greater than the threshold. Finally, an alarm signal is generated based on the early warning module for alarm management. The technical problem that the existing technology cannot achieve comprehensive, accurate and efficient risk identification, evaluation, decision-making and early warning in container logistics safety monitoring is solved. By comprehensively and accurately monitoring container logistics risks, timely and accurately locating risk sources and issuing effective alarms, the technical effect of reducing logistics risks and effectively ensuring the safety of logistics and goods is achieved.

Description

Technical Field

[0001] The present application relates to the field of logistics information technology, and in particular to a container logistics safety monitoring and early warning platform based on the Internet of Things. Background Art

[0002] In the field of container logistics, with the continuous growth of its scale and complexity, logistics safety issues have become increasingly prominent, and the demand for logistics monitoring and early warning has become increasingly urgent. Traditional container logistics monitoring methods have many defects. The logistics monitoring needs are not clear and accurate enough, the mining of logistics risk event chains is not comprehensive and in-depth enough, there is a lack of effective decision-making models to scientifically respond to risks, the utilization efficiency of monitoring data is low, and the early warning mechanism is not perfect and timely enough. These problems make it difficult to effectively identify and control logistics risks, which has brought huge economic losses and safety hazards to logistics companies and related industries. It has become an urgent task to build a comprehensive, accurate and efficient container logistics safety monitoring and early warning platform based on the Internet of Things.

[0003] At the current stage, relevant technologies exist that cannot achieve comprehensive, accurate and efficient risk identification, assessment, decision-making and early warning in container logistics safety monitoring. Summary of the invention

[0004] The present application provides a container logistics safety monitoring and early warning platform based on the Internet of Things. It first uses the relevant information of interactive logistics tasks to determine the monitoring needs, then mines the logistics risk event chain, and then builds a multi-step decision-making model. The risk event chain is then monitored to obtain data and transmitted back. After receiving the data, the logistics management center determines the decision matrix and probability, determines the probability of exceeding the limit, and locates the risk order if it is greater than the threshold. Finally, an alarm signal is generated based on the early warning module for alarm management. By comprehensively and accurately monitoring the risks of container logistics, the risk sources are located in a timely and accurate manner and effective alarms are issued, thereby achieving the technical effect of reducing logistics risks and effectively ensuring the safety of logistics and goods.

[0005] This application provides a container logistics safety monitoring and early warning platform based on the Internet of Things, including:

[0006] A demand determination module, which is used to interact with the logistics network and container type of the target logistics task, and determine the logistics monitoring requirements based on logistics safety and cargo safety; a risk event chain mining module, which is used to mine the logistics risk event chain based on the logistics monitoring requirements, wherein each link node is marked with a logistics network node; a decision model building module, which is used to traverse the logistics risk event chain transformation to determine the event matrix set based on the event decomposition elements and the element equivalence relationship, perform data-driven modeling, and build a multi-step decision model, wherein the multi-step decision model is based on the matrix decision of event decomposition and dimensionality reduction as the principle, and the greedy decision based on the decomposition matrix as the decision-making method; a multi-source monitoring data acquisition module, which is used to acquire data from the multi-source monitoring data acquisition module. The block is used to traverse the logistics risk event chain, conduct monitoring based on the front-end monitoring equipment for the triggered logistics risk events, obtain multi-source monitoring data and transmit it back; the occurrence probability determination module, the occurrence probability determination module is used for the logistics management center to receive the multi-source monitoring data, combine the multi-step decision model to perform event matrix decomposition and sub-event concurrent decision-making, determine the event decision matrix and calculate the probability of event occurrence; the risk list determination module, the risk list determination module is used to make an over-limit judgment on the probability of occurrence of the event, if it is greater than the probability threshold, traverse the event decision matrix to trace the risk point and determine the risk list; the safety alarm management module, the safety alarm management module is used to generate a risk alarm signal based on the risk list based on the early warning module, and perform container logistics safety alarm management.

[0007] The container logistics safety monitoring and early warning platform based on the Internet of Things proposed in this application is intended to first interact with the relevant information of the logistics tasks to determine the monitoring needs, then explore the logistics risk event chain, and then build a multi-step decision-making model. The risk event chain is then monitored to obtain data and transmitted back. After receiving the data, the logistics management center determines the decision matrix and probability, and determines the probability of exceeding the limit. If it is greater than the threshold, the risk list is located. Finally, an alarm signal is generated based on the early warning module for alarm management, thereby achieving the technical effect of reducing logistics risks and effectively ensuring the safety of logistics and goods through comprehensive and accurate monitoring of container logistics risks, timely and accurate location of risk sources and issuing effective alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in this application to illustrate the operations performed by the platform according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0009] Figure 1 A schematic diagram of the structure of a container logistics safety monitoring and early warning platform based on the Internet of Things provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of the module structure of the decision model building of the container logistics safety monitoring and early warning platform based on the Internet of Things provided in an embodiment of the present application.

[0011] Explanation of the accompanying drawings: demand determination module 10, risk event chain mining module 20, decision model building module 30, multi-source monitoring data acquisition module 40, occurrence probability determination module 50, risk order determination module 60, safety warning management module 70. DETAILED DESCRIPTION

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0013] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0014] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0015] The present application embodiment provides a container logistics safety monitoring and early warning platform based on the Internet of Things, such as Figure 1 As shown, the method includes:

[0016] The demand determination module 10 is used to interact with the logistics network and container type of the target logistics task, and determine the logistics monitoring requirements based on logistics safety and cargo safety. Specifically, the demand determination module 10 conducts a comprehensive and in-depth analysis of the target logistics task, identifies key information such as route planning, transportation methods, and transit nodes of the logistics network, and uses data interaction and integration technology to obtain detailed information such as specifications, materials, and sealing of the container type, and analyzes the association and adaptability between them. Based on the analysis results of the logistics task, the logistics monitoring requirements are determined. The logistics monitoring requirements include information such as logistics methods, logistics status, and cargo denaturation, and also cover special situations that may need to be focused on (such as temperature control of cold chain logistics, environmental protection of living logistics, etc.), which are used to guide subsequent monitoring work. The characteristics of the obtained logistics monitoring requirements are extracted and classified. The characteristics refer to key factors that affect logistics safety and cargo safety, such as the particularity of the logistics method, the fragility of the cargo, the complexity of the transportation route, etc., and the requirements with similar characteristics are classified into one category through the classification algorithm. According to the classification results and the characteristics of logistics, the corresponding monitoring focus and strategy are clarified. Each focus and strategy represents a group of logistics tasks with the same or similar requirements.

[0017] The risk event chain mining module 20 is used to mine the logistics risk event chain based on the logistics monitoring needs, wherein each link node is marked with a logistics network node. Specifically, the risk event chain mining module 20 conducts a detailed analysis of the logistics monitoring needs, identifies key links and potential risk points in the logistics process, such as the choice of transportation mode, the adaptability of container types, the complexity of transportation routes and other key information, and uses risk analysis and assessment techniques to mine various risk events that may occur in the logistics process, and analyze the relationship and influence between them. Based on the analysis results of the logistics monitoring needs, the logistics risk event chain is determined. The logistics risk event chain covers the occurrence links, triggering conditions, impact scope and possible response measures (such as emergency plans, risk warnings, etc.) of each risk event, and is used to guide subsequent risk prevention and control work. The risk characteristics of the obtained logistics risk event chain are extracted and classified. The risk characteristics refer to the key factors that affect the risk level and response strategies, such as the probability of risk occurrence, the possible losses, the controllability of risks, etc. Through the classification algorithm, risk events with similar risk characteristics are classified into one category. According to the classification results and the characteristics of the risk events, the corresponding risk response categories are configured. Each risk response category represents a group of risk events with the same or similar risk characteristics.

[0018] In a possible implementation, the risk event chain mining module 20 includes: a logistics risk event chain component unit, the logistics risk event chain component unit is used for the logistics risk event chain to include long-time events and short-time events, and to set the risk decision frequency based on event fluctuations. Specifically, the logistics risk event chain component unit clarifies the composition of the logistics risk event chain, and the logistics risk event chain covers long-time events and short-time events. For long-time events, such as the entire long-distance transportation process, the duration is long, but the risk changes relatively slowly and steadily. In this case, a relatively low risk decision frequency is set based on the fluctuation of the event. This means that there is no need to conduct too frequent monitoring and decision-making, avoiding unnecessary resource consumption and reactive work. For example, during logistics transportation, if the state of the goods is relatively stable and the environmental conditions have not changed significantly, data can be regularly sent back for judgment without continuous monitoring. For short-time events, such as container loading and unloading, the conversion of logistics modes (such as the conversion from sea, land and air transportation modes) and other key links, the risk changes rapidly and has a large impact. In these scenarios, it is necessary to set a higher risk decision frequency for continuous monitoring and judgment. Because any small mistakes or changes in these links may lead to greater logistics risks, it is necessary to maintain a high degree of vigilance and make timely decisions to ensure the safety and smoothness of the logistics process. For example, during the long-distance sea transportation of a batch of goods, due to the stability of the ship and the predictability of the environment, it may be sufficient to send back the cargo status and transportation data every 24 hours for risk assessment. However, when the goods arrive at the port for loading and unloading, due to the complex operation, large number of personnel and the easy damage of the goods, continuous real-time monitoring is required, and decisions must be made at any time according to the situation to prevent risks such as loss, damage or delay of goods.

[0019] The decision model building module 30 is used to traverse the logistics risk event chain transformation to determine the event matrix set based on the event decomposition elements and element equivalence relations, perform data-driven modeling, and build a multi-step decision model. The multi-step decision model is based on the matrix decision of event decomposition and dimensionality reduction as the principle, and the greedy decision based on the decomposition matrix as the decision-making method. Specifically, the decision model building module 30 conducts a detailed element analysis of the logistics risk event chain, identifies important information such as the key features, influencing factors, and correlations of the event, uses data mining and processing technology to extract the various components of the logistics risk event, and analyzes the logical relationship and influence mode between them. Based on the element analysis results of the logistics risk event chain, the event matrix set is determined. The event matrix set covers information such as event decomposition elements, element equivalence relations, and possible special circumstances (such as emergencies, extreme conditions, etc.), which is used to support the construction of a multi-step decision model. The decision features of the acquired event matrix set are extracted and classified. The decision features refer to the key factors that affect the decision effect, such as risk probability, degree of loss, response cost, etc. Through the classification algorithm, the event matrices with similar decision features are classified into one category. According to the classification results and the characteristics of the event matrix, the corresponding decision model category is configured. Each decision model category represents a group of logistics risk events with the same or similar decision features, and has the same decision logic and response strategy.

[0020] In one possible implementation, Figure 2 As shown, the decision model building module 30 includes: an event characteristic determination unit, which is used to traverse the logistics risk event chain and determine the event characteristics of the first event. Specifically, the event characteristic determination unit traverses the logistics risk event chain, focusing on the first key event, and determines the essential attributes of the event through data collection, analysis and logical reasoning, considering the specific time and space scope of the event, the details of the logistics facilities and operation processes involved, the direct and indirect consequences that may be caused, and the impact level of the event in the entire logistics system. For example, in the container transportation scenario, if the first event is the collision risk of the container during loading and unloading, the unit will clarify the specific terminal location where the collision occurs, the model and performance parameters of the loading and unloading equipment involved, the degree of damage to the container structure that may be caused by the collision, and the degree of delay in the subsequent transportation plan.

[0021] An element relationship determination unit, the element relationship determination unit is used to determine the event decomposition element and the element equivalence relationship based on the event characteristics, wherein the element equivalence relationship is an equivalent substitution relationship between the upper and lower elements before and after decomposition, and the decomposition layer contains at least one upper and lower layer, and the event decomposition element includes at least one item. Specifically, the element relationship determination unit uses logical reasoning, statistical methods and professional field rules based on the acquired event characteristics to accurately determine the event decomposition element and the element equivalence relationship, and the element equivalence relationship is manifested as the substitutability and association logic between the upper and lower elements before and after decomposition, and the decomposition layer contains at least one upper and lower layer to ensure that the analysis of the event has sufficient depth and accuracy. For example, for the above-mentioned container loading tilt event, the decomposition elements may include the structural strength of the container, the accuracy of the loading equipment, the flatness of the ground, etc. The element equivalence relationship is that a certain degree of change in the structural strength of the container is equivalent to the tolerable change in the tilt angle, or a specific loading equipment accuracy corresponds to a different loading error range.

[0022] The first event matrix determination unit is used to decompose the first event based on the event decomposition elements and the element equivalent relationship, perform matrix conversion on the event decomposition result, and determine the first event matrix. Specifically, based on the event decomposition elements and the element equivalent relationship determined by the element relationship determination unit, the first event matrix determination unit uses mathematical modeling and matrix theory to convert the complex first event into a quantifiable and computable matrix form, and eliminates the ambiguity and uncertainty in the event description by matrix conversion of the event decomposition result, thereby obtaining a clear and accurate first event matrix. For example, for the container loading tilt event, the above-mentioned decomposition elements and equivalent relationships are converted into elements and relationships in the matrix to form a matrix that can accurately describe the event status and potential risks.

[0023] A mapping establishment unit, the mapping establishment unit is used to establish a mapping between the first event matrix and the first greedy target matrix. Specifically, after the first event matrix is ​​determined, the mapping establishment unit clarifies the structure and element meanings of the first event matrix and the first greedy target matrix, and seeks potential associations and correspondences between the two based on linear programming, dynamic programming, etc. For example, assuming that the first greedy goal is to reduce the risk of cargo loss to a minimum at the minimum cost, then each element in the first event matrix (such as the personnel operation standardization score, the monitoring equipment coverage percentage, etc.) will be established with the cost and risk reduction degree. By continuously adjusting and optimizing the parameters of these functional relationships, a clear and accurate mapping relationship is gradually established. In the process of establishing the mapping relationship, a sensitivity analysis will also be introduced to evaluate the impact of different parameter changes on the mapping relationship, thereby ensuring the stability and reliability of the mapping relationship. Through the above detailed and operational steps, the functions of each module and unit can be realized, providing strong support for the effective management and decision-making of logistics risks.

[0024] In a possible implementation, the mapping establishment unit further includes: a first event matrix identification subunit, the first event matrix identification subunit is used to identify the first event matrix and determine the first greedy target matrix based on the event risk threshold, wherein the greedy target is the risk decision point of the matrix sub-event. Specifically, the first event matrix identification subunit accurately identifies the first event matrix that has been constructed. In the identification process, various relevant data features and patterns are fully considered. Based on the pre-set event risk threshold, the first event matrix is ​​analyzed and evaluated. Through calculation and comparison, the first greedy target matrix corresponding to it is determined. Each element and structure in the greedy target matrix is ​​constructed with the risk decision point of the matrix sub-event as the core. For example, in logistics transportation, for the first event matrix of damaged goods, the identification subunit will determine the corresponding greedy target matrix based on the set risk threshold, such as the loss of the value of the goods exceeding a certain proportion, which may focus on risk decision points such as reducing the degree of bumps during transportation and optimizing packaging materials.

[0025] The decision criterion subunit is used to traverse the first greedy target matrix, combine the greedy distribution results, and use the mean of the risk probability as the decision criterion for the first event. Specifically, the decision criterion subunit traverses the first greedy target matrix. During the traversal process, combined with the greedy distribution results obtained previously, the statistical method and the risk analysis model are used to calculate the mean of the risk probability, and the mean is used as the decision criterion for the first event. The mean will become the key basis for evaluating and comparing different decision plans. For example, in the case of the above-mentioned damage to the goods, the decision criterion subunit will calculate the mean of the probability of the risk of damage to the goods under specific conditions based on the relevant information about the transportation conditions and packaging measures in the greedy target matrix, combined with the greedy distribution results, so as to decide whether it is necessary to adjust the mode of transportation, strengthen packaging protection, and other decisions.

[0026] In a possible implementation, the decision criterion subunit includes: a distribution ratio initialization microunit, which is used to initialize the greedy distribution ratio based on the number of decomposition elements. Specifically, the distribution ratio initialization microunit obtains the number information of the decomposition elements, which include key elements in logistics risk assessment such as edges, faces, and corners of containers, and makes a preliminary setting for the greedy distribution ratio based on the number of decomposition elements. For example, assuming there are three decomposition elements, i.e., three faces, the greedy distribution ratio of each face may be initially set to 1 / 3.

[0027] A preset greedy ratio determination micro-unit is used to traverse the decomposition elements, adjust the greedy distribution ratio based on the primary and secondary relationship of the elements, and determine a layer of preset greedy ratio. Specifically, after the distribution ratio initialization micro-unit completes the preliminary setting, the preset greedy ratio determination micro-unit traverses each decomposition element in detail, and adjusts the initial greedy distribution ratio based on the primary and secondary relationship of the elements. For example, when evaluating the stability of a container, it is considered that the importance of the face is higher than that of the edge and the corner, so the greedy distribution ratio of the face will be increased accordingly. Taking the edge, face, and corner as an example, if the face is considered to be the main element, the edge is a secondary element, and the corner is an even more secondary element, the initial greedy distribution ratio will be adjusted to increase the proportion of the face, appropriately reduce the proportion of the edge, and make the proportion of the corner even smaller. Through such adjustment, a layer of preset greedy ratio is determined.

[0028] The greedy distribution result determination micro-unit is used to map and evenly divide the preset greedy ratio of the first decomposition element based on the number of decompositions of the lower elements of the first decomposition element, and determine the greedy distribution result of the first decomposition element. Specifically, after the preset greedy ratio determination micro-unit determines a preset greedy ratio, the greedy distribution result determination micro-unit maps and evenly divides the previously determined preset greedy ratio based on the number of decompositions of the lower elements of the first decomposition element. Assuming that the first decomposition element is a face, and the face can be further subdivided into multiple small areas or parts, the greedy ratio corresponding to the adjusted face is evenly distributed according to the number of decompositions of these lower elements. If there are multiple edges, the greedy ratio corresponding to the adjusted edges will be evenly divided in the same way, so as to finally determine the greedy distribution result of the first decomposition element. The result will provide accurate data support for subsequent logistics risk assessment and decision-making. For example, when determining the specific greedy distribution result of a certain face of a container, if this face is subdivided into five small areas, the adjusted greedy ratio of the face will be evenly divided into these five small areas.

[0029] In a possible implementation, the decision model building module 30 further includes: an indirect judgment standard determination unit, which is used to traverse the logistics risk event chain, determine the risk event blind spot, determine the indirect judgment standard based on the risk event blind spot, and constrain the decomposition elements based on the indirect judgment standard. Specifically, the indirect judgment standard determination unit traverses the entire logistics risk event chain, and uses various analysis methods and tools, including data analysis, process simulation, empirical models, etc., to deeply explore various links and potential risk points in the logistics process. Through investigation, the blind spots of risk events in the logistics risk event chain are determined. Blind spots may be risk areas that are difficult to directly observe and evaluate due to incomplete information, limited monitoring methods, complex process interactions, etc. Based on the accurate identification of the blind spots of risk events, the indirect judgment standards are further determined. The standards are usually derived through comprehensive analysis and derivation of peripheral factors, historical data, industry experience and related theories related to the blind spots. For example, if the damage to goods in a certain transportation link is difficult to monitor directly, the indirect judgment standards may be determined based on factors such as the road conditions of the transportation route, transportation time, and cargo packaging materials. After the indirect judgment standards are determined, they are applied to the constraints on the decomposition elements. This means that when assessing and handling risks, these standards will serve as important references and constraints to ensure that the analysis and processing of decomposition factors are in line with actual conditions and risk management requirements. Through constraints, risks can be assessed more accurately, effective response strategies can be formulated, and the possibility and impact of risks can be reduced. For example, in a cold chain logistics scenario, for certain areas in the cold storage where it is difficult to directly monitor the temperature (risk event blind spots), indirect judgment criteria can be determined based on factors such as the operating status of the overall refrigeration equipment of the cold storage and the frequency of door opening, thereby reasonably constraining and evaluating temperature-related decomposition factors (such as the shelf life of goods, the types of perishable goods, etc.).

[0030] A multi-source monitoring data acquisition module 40 is used to traverse the logistics risk event chain, monitor the triggered logistics risk events based on the front-end monitoring equipment, obtain multi-source monitoring data and transmit it back. Specifically, the multi-source monitoring data acquisition module 40 conducts a detailed combing and analysis of the logistics risk event chain, identifies important information such as the potential risk points of each key link in the logistics process, the operating status of logistics facilities, and the relevant characteristics of the goods. With the help of advanced monitoring technology and data collection methods, the specific manifestations of logistics risk events are extracted, and the correlation and influence patterns between them are analyzed. Based on the in-depth analysis results of the logistics risk event chain, the key directions and goals of multi-source monitoring are clarified. The multi-source monitoring data covers the physical state parameters of the goods, the dynamic changes of the transportation environment, the real-time situation of logistics operations, and possible abnormal warnings, etc., which are used to provide strong data support for subsequent risk assessment and decision-making. The key features of the collected multi-source monitoring data are extracted and classified. The key features refer to factors that have an important impact on logistics risk assessment, such as temperature fluctuations of goods, congestion of transportation routes, failure rate of logistics equipment, etc. Through efficient data analysis algorithms, monitoring data with similar features are classified into one category. According to the classification results and the characteristics of logistics risks, corresponding monitoring strategies and data processing methods are configured. Each monitoring strategy represents a group of logistics monitoring scenarios with the same or similar features.

[0031] The occurrence probability determination module 50 is used for the logistics management center to receive the multi-source monitoring data, combine the multi-step decision model to perform event matrix decomposition and sub-event concurrent decision-making, determine the event decision matrix and calculate the probability of event occurrence. Specifically, the occurrence probability determination module 50 conducts an in-depth analysis and interpretation of the multi-source monitoring data received by the logistics management center, identifies important information such as key indicators of logistics status, environmental change parameters, and cargo characteristics reflected therein, uses advanced data processing and analysis techniques, extracts the various key elements covered by the multi-source monitoring data, and analyzes the inherent correlation and interaction between them. Based on the detailed analysis results of the multi-source monitoring data, the ideas and directions of event matrix decomposition are clarified. The event matrix covers the key links of event decomposition, the correlation between sub-events, and risk nodes that may need to be focused on, etc., to support the subsequent concurrent decision-making process. The key features of the acquired event matrix information are extracted and classified. The key features refer to factors that have a significant impact on the probability of event occurrence, such as the severity of environmental changes, the degree of standardization of logistics operations, the stability of equipment operation, etc. Through a precise probability calculation algorithm, event matrices with similar features are classified into one category. According to the classification results and the characteristics of logistics events, the corresponding probability calculation model is configured. Each probability calculation model represents a group of logistics event probability assessment scenarios with the same or similar characteristics, and follows the same probability calculation logic and analysis strategy.

[0032] In a possible implementation, the occurrence probability determination module 50 includes: a trend prediction unit, which is used to identify the event decision matrix and predict the matrix evolution trend. Specifically, the trend prediction unit obtains the event decision matrix, analyzes the data and relationships in the matrix, and identifies the event development pattern and characteristics represented by the matrix by analyzing the change rules of historical data and current data and the dynamic association between matrix elements. On this basis, the future evolution trend of the matrix is ​​predicted by using time series analysis, machine learning and other prediction technologies. For example, for the decision matrix of cargo loss risk events in logistics transportation, the trend prediction unit will analyze the data such as the speed of cargo loss and changes in the causes of loss in the past period of time, and predict the development direction of future loss situations.

[0033] A decision node determination unit is used to determine the lower-level decision nodes by combining the risk decision frequency with the matrix evolution trend. Specifically, the decision node determination unit obtains relevant information about the risk decision frequency and combines it with the matrix evolution trend. By analyzing the frequency of risk events and the change trend of the matrix, the lower-level decision nodes are determined. The determination of the decision nodes is based on the assessment of the risk and the judgment of future development to ensure that timely and effective decisions can be made at critical moments. For example, in the risk of logistics and distribution delays, the decision nodes that need to adjust the transportation route or increase transportation resources are determined by combining the frequency of delays and the predicted trend of delay aggravation.

[0034] The risk decision unit is used to transmit back the multi-source monitoring data of the lower-level decision node for the same risk event, and make risk decisions in combination with the multi-step decision model. Specifically, the risk decision unit transmits back the multi-source monitoring data of the lower-level decision node for the same risk event. The data comes from various front-end monitoring devices and contains real-time information. In combination with the multi-step decision model constructed previously, the data is deeply analyzed and calculated, and the final risk decision is made through the evaluation and judgment of the model. For example, when facing the risk event of abnormal temperature of cargo storage, it decides whether to initiate emergency cooling measures or adjust the storage environment based on the transmitted temperature monitoring data and decision model. Through the collaborative work of the trend prediction unit, the decision node determination unit and the risk decision unit, the effective monitoring, prediction and decision-making of logistics risk events are realized to reduce the losses caused by risks and improve the efficiency and stability of logistics operations.

[0035] In a possible implementation, the risk decision unit further includes: a risk decision composition subunit, and the risk decision composition subunit is used for risk decision composition. Risk decision includes full event matrix decision and local event matrix decision. Specifically, the risk decision composition subunit clarifies that its task is to build the framework and content of risk decision and determine the specific form of risk decision. Full event matrix decision is a comprehensive decision-making method, which comprehensively analyzes all events involved in the entire logistics risk event chain and their interrelationships, and conducts a holistic analysis and evaluation of the entire event matrix. It needs to collect and process a large amount of data, including risk factors in each link, the development dynamics of events, and the correlation and impact between different events. Through comprehensive data analysis and complex model calculations, a comprehensive decision covering the entire logistics process is obtained. The local event matrix decision focuses on specific parts or key links in the logistics risk event chain, and conducts in-depth analysis of local event matrices with significant risk impact or key roles. The decision-making method will focus more on certain key areas, such as a section of road that is prone to problems during transportation, the storage link of a specific type of goods, etc. By collecting and analyzing data related to these local events in a targeted manner, using specific algorithms and models, a decision-making plan specifically for these local problems is formulated. For example, in a multinational logistics project, the full event matrix decision may comprehensively consider all risk factors in the entire process from goods production, packaging, transportation, warehousing to distribution, and formulate a set of overall optimization strategies. The local event matrix decision may specifically formulate detailed risk response measures for sections of road that pass through high-risk areas during transportation, or for special packaging and handling links of easily damaged goods.

[0036] The risk list determination module 60 is used to determine whether the probability of the event occurs beyond the limit. If it is greater than the probability threshold, the event decision matrix is ​​traversed to trace the risk point and determine the risk list. Specifically, the risk list determination module 60 performs a detailed over-limit determination operation on the probability of the event, identifies the numerical characteristics of the probability of the event, and compares it with the preset probability threshold. If the probability of the event is greater than the threshold, the deep traversal process of the event decision matrix is ​​immediately started. Using data mining and logical analysis techniques, the source of the risk point can be accurately traced and its specific position in the matrix can be accurately located. Based on the results of tracing and positioning, the key contents of the risk list are clarified. The risk list covers information such as the root cause of the risk, the specific manifestation form, the possible consequences, and related important details, aiming to provide an accurate and guiding basis for subsequent risk response strategies and disposal measures. The key features of the acquired risk-related information are extracted and classified. The key features refer to factors that have a significant impact on the formation and development of risks, such as the scope of risk impact, urgency, controllability, etc. Through a scientific classification algorithm, risk situations with similar characteristics are grouped together. According to the classification results and the characteristics of the risks, corresponding risk response categories are configured. Each risk response category represents a group of risk situations with the same or similar characteristics.

[0037] The safety warning management module 70 is used to generate a risk warning signal based on the risk list based on the early warning module, and perform safety warning management on container logistics. Specifically, the safety warning management module 70 establishes a connection and data exchange with the early warning module. The early warning module will provide it with key information and data related to the risk list. Based on the detailed content of the received risk list, the safety warning management module 70 generates a corresponding risk warning signal, and comprehensively analyzes factors such as the severity, urgency, and scope of impact of the risk to determine the type, level, and specific form of the warning signal. The warning signal form includes sound alarms, flashing lights, text prompts, or message notifications sent to mobile devices of relevant personnel. After the alarm signal is generated, it is transmitted to various relevant links and systems of container logistics to achieve comprehensive safety warning management, including but not limited to container loading and unloading areas, cockpits of transport vehicles, logistics monitoring centers, etc. For example, when the risk list shows that the goods in a container may be damaged due to excessive temperature, the safety warning management module 70 will generate an urgent sound alarm and text prompt, and at the same time send a notification to the driver and monitoring center responsible for the container transportation, reminding them to take cooling measures in time to ensure the safety of the goods and the normal operation of logistics. The safety warning management module 70 can issue risk warnings to relevant personnel and systems in a timely and effective manner, prompting them to take corresponding measures to reduce risks and avoid losses.

[0038] In a possible implementation, the safety alarm management module 70 includes: a temporary connection establishment unit, the temporary connection establishment unit is used to establish a temporary connection between the alarm module and the logistics terminal, wherein the temporary connection is switched synchronously with the logistics transit. Specifically, the temporary connection establishment unit establishes a temporary communication connection between the alarm module and the logistics terminal, and the establishment of the connection is based on the consideration of the real-time demand and dynamic changes in the logistics transportation process. During the logistics transit, the logistics terminal may change, such as switching from the terminal at the place of shipment to the transit station terminal during transportation, and then to the terminal at the place of receipt. The temporary connection establishment unit can perceive these changes in real time, and switch the connection synchronously with the logistics transit, ensuring that the alarm module always maintains effective communication with the currently related logistics terminal. For example, when the goods depart from the factory, a connection with the factory logistics terminal is established; when the goods arrive at the transit station, it is quickly switched to the logistics terminal of the transit station.

[0039] The alarm signal conversion unit is used to identify the risk list, determine the risk plan, and convert it into a risk alarm signal. Specifically, after the temporary connection is established, the alarm signal conversion unit first identifies and analyzes the risk list, obtains key information such as the risk type, severity, and scope of impact described therein, and converts the information into a specific risk alarm signal according to the risk plan prepared in advance. The risk plan contains the response strategies and alarm method regulations for different types and degrees of risks. Through conversion, it is ensured that the generated alarm signal accurately and clearly reflects the essence of the risk and the response measures that need to be taken. For example, for serious cargo damage risks, the generated alarm signal may be a strong sound and light alarm and emergency prompt information.

[0040] The logistics risk warning unit is used to transmit the risk warning signal to the logistics terminal to warn about logistics risks. Specifically, after the conversion of the warning signal is completed, the logistics risk warning unit uses the established temporary connection to accurately transmit the risk warning signal to the corresponding logistics terminal through various communication means, such as wireless network, wired network, etc. The logistics terminal can be a computer terminal, a mobile device terminal, etc. After receiving the warning signal, the logistics terminal will display the warning information to the relevant personnel in an intuitive way, such as displaying a warning screen on the screen, issuing a sound prompt, etc., so as to achieve the purpose of logistics risk warning and remind the relevant personnel to take timely measures to deal with the risk. For example, after the driver on the transport vehicle receives the warning signal through the on-board terminal, he can immediately know the risk situation and take corresponding driving adjustments or emergency measures. Through the coordinated work of the temporary connection establishment unit, the warning signal conversion unit and the logistics risk warning unit, timely and accurate warnings of logistics risks are achieved, ensuring the safety and smooth progress of logistics transportation.

[0041] The embodiment of the present application first interacts with the relevant information of the logistics task to determine the monitoring needs, then mines the logistics risk event chain, and then builds a multi-step decision model. The risk event chain is then monitored to obtain data and transmitted back. After receiving the data, the logistics management center determines the decision matrix and probability, determines the probability over-limit, and locates the risk order if it is greater than the threshold. Finally, an alarm signal is generated based on the early warning module for alarm management. By comprehensively and accurately monitoring the container logistics risks, the risk sources are located in a timely and accurate manner and effective alarms are issued, thereby achieving the technical effect of reducing logistics risks and effectively ensuring the safety of logistics and goods.

[0042] Although the present application makes various references to certain modules in the platform according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0043] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. The container logistics safety monitoring and early warning platform based on the Internet of Things is characterized by: The platform includes: A demand determination module, which is used to interact with the logistics network and container type of the target logistics task, and determine the logistics monitoring needs based on logistics safety and cargo safety; A risk event chain mining module, wherein the risk event chain mining module is used to mine logistics risk event chains based on the logistics monitoring requirements, wherein each link node is identified with a logistics network node; A decision model building module, which is used to decompose the elements and the equivalent relationship between the elements according to the events, traverse the logistics risk event chain transformation to determine the event matrix set, perform data-driven modeling, and build a multi-step decision model. The multi-step decision model is based on the matrix decision of event decomposition and dimensionality reduction as the principle, and the greedy decision based on the decomposition matrix as the decision-making method; A multi-source monitoring data acquisition module, which is used to traverse the logistics risk event chain, perform monitoring based on front-end monitoring equipment for triggering logistics risk events, acquire multi-source monitoring data and transmit it back; An occurrence probability determination module, wherein the occurrence probability determination module is used for the logistics management center to receive the multi-source monitoring data, perform event matrix decomposition and sub-event concurrent decision-making in combination with the multi-step decision model, determine the event decision matrix and calculate the probability of event occurrence; A risk list determination module, which is used to determine if the probability of the event is beyond the limit. If it is greater than the probability threshold, the event decision matrix is ​​traversed to trace the risk point and determine the risk list; A safety warning management module is used to generate a risk warning signal based on the risk list based on the early warning module, and perform safety warning management on container logistics.

2. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 1 is characterized in that: The decision model building module includes: An event characteristic determination unit, the event characteristic determination unit is used to traverse the logistics risk event chain and determine the event characteristic of the first event; An element relationship determination unit, the element relationship determination unit is used to determine the event decomposition element and the element equivalence relationship based on the event characteristics, wherein the element equivalence relationship is an equivalent substitution relationship between the upper and lower elements before and after decomposition, and the decomposition layer includes at least one upper and lower layer, and the event decomposition element includes at least one item; a first event matrix determining unit, the first event matrix determining unit being used to decompose the first event based on the event decomposition elements and the element equivalence relationship, perform matrix conversion on the event decomposition result, and determine a first event matrix; A mapping establishing unit, wherein the mapping establishing unit is used to establish a mapping between the first event matrix and a first greedy target matrix.

3. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 2 is characterized in that: The mapping establishment unit comprises: A first event matrix identification subunit, the first event matrix identification subunit is used to identify the first event matrix and determine a first greedy target matrix based on an event risk threshold, wherein the greedy target is a risk decision point of a matrix sub-event; A decision criterion subunit is used to traverse the first greedy target matrix, combine the greedy distribution results, and use the mean of the risk occurrence probability as the decision criterion for the first event.

4. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 3 is characterized in that: The decision criteria subunit includes: A distribution ratio initialization micro-unit, wherein the distribution ratio initialization micro-unit is used to initialize a greedy distribution ratio based on the number of decomposition elements; A preset greedy ratio determination micro-unit is used to traverse the decomposition elements, adjust the greedy distribution ratio based on the primary and secondary relationship of the elements, and determine a layer of preset greedy ratio; The greedy distribution result determines a micro-unit, and the greedy distribution result determines the micro-unit for mapping and evenly dividing the preset greedy ratio of the first decomposition element based on the number of lower-level element decompositions of the first decomposition element, to determine the greedy distribution result of the first decomposition element.

5. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 1 is characterized in that: The logistics risk event chain is traversed to determine the risk event blind spot, an indirect determination standard is determined based on the risk event blind spot, and the decomposition elements are constrained based on the indirect determination standard.

6. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 1 is characterized in that: The logistics risk event chain includes long-term events and short-term events, and the risk decision frequency is set based on event fluctuations.

7. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 6 is characterized in that: The occurrence probability determination module comprises: A trend prediction unit, the trend prediction unit is used to identify the event decision matrix and predict the matrix evolution trend; A decision node determination unit, the decision node determination unit is used to determine a lower-level decision node by combining the risk decision frequency with the matrix evolution trend; The risk decision unit is used to transmit back the multi-source monitoring data of the lower decision node for the same risk event and make risk decisions in combination with the multi-step decision model.

8. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 7 is characterized in that: Risk decision composition Risk decision includes full event matrix decision and local event matrix decision.

9. The container logistics safety monitoring and early warning platform based on the Internet of Things as claimed in claim 1, characterized in that: The safety warning management module includes: A temporary connection establishing unit, the temporary connection establishing unit is used to establish a temporary connection between the alarm module and the logistics terminal, wherein the temporary connection is switched synchronously with the logistics transit; An alarm signal conversion unit, the alarm signal conversion unit is used to identify the risk list, determine the risk plan, and convert it into a risk alarm signal; A logistics risk warning unit, wherein the logistics risk warning unit is used to transmit the risk warning signal to the logistics terminal to issue a logistics risk warning.

Images