Intelligent supply chain transportation management platform based on Internet of Things

Abstract

The invention discloses an intelligent supply chain transportation management platform based on the Internet of Things, and relates to the technical field of the Internet of Things. Historical transportation data, real-time order information and production progress demands are utilized to construct an input feature matrix, and a time sequence prediction method based on a recurrent neural network is adopted to improve the accuracy of transportation demand prediction; low-efficiency distribution and lagging response of transportation resources are avoided, a multi-agent collaborative scheduling module sets a transportation agent agent based on a prediction result, the real-time position of each component is monitored, and possible transportation delay is predicted in combination with a time sequence model; the coordination agent dynamically adjusts the transportation plan based on a shortest path optimization algorithm; a customs agent processes cross-border customs clearance affairs in advance, and registers customs clearance states under the support of a block chain smart contract, so that customs clearance automatic check and an exception triggering mechanism are realized, and the stability and controllability of cross-border transportation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to an intelligent supply chain transportation management platform based on the Internet of Things. Background Art

[0002] In large-scale transportation scenarios, such as the aviation equipment manufacturing industry, the manufacturing of super-large components such as aircraft fuselages, engines, and wings involves multiple suppliers, and ultimately needs to be precisely assembled at the final assembly plant. The components are produced by manufacturers in different regions and transported by various means such as sea, rail, road, or air, and ultimately assembled at the final assembly plant. Due to the large size of the components, high transportation costs, and the need to arrive at the assembly line simultaneously, their transportation scheduling is much more complicated than conventional industrial logistics.

[0003] Current intelligent supply chain management methods have introduced GPS positioning, Internet of Things monitoring, big data analysis and blockchain technology, which can realize the visualization of the transportation process. However, when it comes to multi-node synchronous transportation of oversized components, traditional supply chain management systems are mostly based on single transportation task optimization. The transportation of oversized components requires multiple key components to arrive at the assembly line synchronously within a precise time window. Once a transportation link is delayed, the entire assembly rhythm will be affected.

[0004] In addition, large parts often involve cross-border manufacturing, and the customs clearance process has become an important factor affecting transportation timeliness. Although traditional intelligent supply chain management systems can track the location of goods, it is difficult to proactively predict and avoid customs clearance risks. Once customs clearance delays or policy changes occur, supply chain managers can usually only wait passively and lack the ability to make real-time adjustments, which increases the uncertainty of transportation plans. Therefore, an intelligent supply chain transportation management platform based on the Internet of Things is urgently needed to solve such problems. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides an intelligent supply chain transportation management platform based on the Internet of Things to solve the problems that traditional solutions cannot efficiently cope with the synchronous transportation of oversized components, have limited path optimization, delayed response to emergencies, and low transportation efficiency.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] The embodiment of the present invention provides an intelligent supply chain transportation management platform based on the Internet of Things, which includes:

[0009] Intelligent forecasting module, which is used to collect historical transportation data, real-time order information and production schedule requirements; it uses the long short-term memory network deep learning model to predict future transportation needs;

[0010] Multi-agent collaborative scheduling module, which builds multi-agents based on transportation demand data, including transportation agent agents, coordination agents, and customs agents;

[0011] Smart contract management module, used to register transportation routes and customs clearance status. During the customs clearance process, the smart contract automatically performs customs clearance progress checks. If a delay is detected, the multi-agent collaborative scheduling module is triggered to re-plan the transportation plan;

[0012] The dynamic path optimization module uses graph neural network (GNN) to build a transportation task graph, model different transportation nodes, and calculate dynamic transportation paths based on weather, traffic flow, and transportation mode availability;

[0013] The monitoring feedback module uses IoT devices to monitor the transportation process in real time. If the transportation environment is abnormal, it will automatically go back to the multi-agent collaborative scheduling module, and the multi-agent system will re-optimize the transportation task.

[0014] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, wherein: the transportation agent intelligent agent monitors the real-time location of each component and predicts possible transportation delays;

[0015] Coordinate agents to adjust transportation plans based on the real-time location of each component;

[0016] Customs intelligent body handles cross-border customs clearance matters in advance and monitors the progress of customs clearance.

[0017] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, the management method of the platform includes:

[0018] Step S1, collect historical transportation data, real-time order information and production schedule requirements, and predict future transportation demand through deep learning models to generate transportation demand data;

[0019] Step S2, based on the transportation demand data of step S1, setting a transportation agent intelligent agent to monitor the real-time location of each component and predict possible transportation delays;

[0020] The coordination agent adjusts the transportation plan based on the real-time location of each component and generates an optimized transportation path;

[0021] The customs agent handles cross-border customs clearance matters in advance, monitors customs clearance progress, and generates customs clearance status information;

[0022] Step S3, constructing a smart contract based on the optimized transportation route and customs clearance status information, and registering the transportation route and customs clearance status to the blockchain network;

[0023] Step S4, based on the updated transportation plan, a transportation task graph is constructed using a graph neural network, different transportation nodes are modeled, and the dynamic transportation path is calculated by combining weather, traffic flow, and transportation mode availability;

[0024] Step S5: Based on the dynamic transportation path, IoT devices are used to monitor the transportation process in real time and detect the transportation environment. An intelligent feedback mechanism is set up. If an abnormality occurs at a transportation node, it will automatically go back to step S2, and the multi-agent collaborative scheduling module will re-plan the transportation task.

[0025] In step S5, the transport environment detected includes temperature and humidity, vibration and tilt angle.

[0026] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, the step of predicting future transportation demand through a deep learning model and generating transportation demand data is as follows:

[0027] Collect historical transportation data, real-time order information and production schedule requirements, and construct an input feature matrix, which can be expressed as:

[0028] X={x 1 , x 2 , ..., x m},

[0029] Among them, X represents the input feature matrix, x j represents the jth feature, 1≤j≤m, m represents the total number of features,

[0030] A prediction model based on recurrent neural network (RNN) is used to define the state transfer equation, which is:

[0031] h t =f(W h h t-1 +W x X t +b h ),

[0032] Among them, h t represents the hidden state at time t, h t-1 Indicates the hidden state at the previous moment, W h and W x are the weight matrices of hidden state and input data, respectively, h is the bias term, f(·) represents the nonlinear activation function, X t is the input feature matrix at time t,

[0033] The fully connected layer is used for demand forecasting, and the prediction formula is:

[0034] y t=W y h t +b y ,

[0035] Among them, y t represents the transportation demand data predicted at time t, W y is the output layer weight matrix, b y is the output bias,

[0036] The mean square error MSE is used as the loss function, and the function formula is:

[0037]

[0038] Among them, L represents the loss value, T is the time step, For real transportation demand data,

[0039] The model parameters are updated through the back propagation algorithm, and the update formula is:

[0040]

[0041] Among them, θ is the model parameter, including W h , W x , W y , b h , b y , η is the learning rate.

[0042] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, the step of setting the transportation agent intelligent body based on the transportation demand data of step S1 is as follows:

[0043] Assume a set of transport agent agents, denoted as A:

[0044] A={a 1 , a 2 , ..., a p},

[0045] Among them, A represents the set of transport agent agents, a q represents the qth agent, 1≤q≤p, p represents the total number of agents,

[0046] Set up the component position matrix that the agent monitors:

[0047] P = {(x q ,y q , z q , t q )},

[0048] Where P represents the real-time position set of the components, (x q ,yq , z q ) represents the coordinates of the qth component in three-dimensional space, t q is the timestamp,

[0049] The time series model is used to predict possible transportation delays. The prediction formula is:

[0050] d q =g(P,T d ),

[0051] Among them, d q represents the expected transportation delay time of the qth component, g(·) is the delay prediction function, T d is the historical transportation time distribution,

[0052] Set up a coordination agent to adjust the transportation plan:

[0053] R = {r 1 , r 2 , ..., r s},

[0054] Among them, R represents the optimized transportation path set, r u represents the u-th path, 1≤u≤s, s is the total number of optional paths,

[0055] The shortest path optimization algorithm is used to optimize the transportation path. The optimization formula is:

[0056]

[0057] Among them, r * represents the optimal transportation path, C(r u , a q ) represents the transport agent agent a q On the path r u The transportation costs on.

[0058] As a preferred solution of the IoT-based intelligent supply chain transportation management platform described in the present invention, in which: in step S3, during the customs clearance process, the blockchain smart contract automatically performs a customs clearance progress check. If a delay is detected, the coordinating intelligent entity in step S2 is triggered to re-plan the transportation plan and update the transportation plan.

[0059] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, in step S3, the step of building a smart contract and automatically executing customs clearance progress inspection includes:

[0060] Set the smart contract set C in the blockchain network:

[0061] C={c 1 , c2 , ..., c v},

[0062] Among them, C represents the smart contract set, c w represents the wth smart contract, responsible for managing customs clearance and transportation information, v is the number of smart contracts,

[0063] Set the clearance status set S:

[0064] S={s 1 ,s 2 , ..., s k},

[0065] Among them, S represents the clearance state set, s x represents the clearance status of the xth level, k is the total number of levels,

[0066] Set up the smart contract trigger mechanism and set the event trigger conditions as:

[0067] E={e 1 , e 2 , ..., e k},

[0068] Among them, E represents the trigger event set, e x Represents the event triggered when level x is delayed.

[0069] Set the automatic execution logic of the smart contract:

[0070] If x Normal, then A(s x ) = Update status,

[0071] If x Delay, then A(s x ) = trigger transport adjustment,

[0072] Among them, A(s x ) represents the execution action of the smart contract at level x.

[0073] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, wherein: in step S4, when a transportation route is detected to have a risk of delay, it is automatically adjusted to other transportation modes;

[0074] In step S4, the different transportation nodes include warehouses, ports, railway hubs and customs clearance stations.

[0075] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, the steps of constructing a transportation task graph based on the updated transportation plan using a graph neural network, modeling different transportation nodes, and calculating a dynamic transportation path in combination with weather, traffic flow and transportation mode availability are as follows:

[0076] Construct a transportation task graph and set the transportation network to have a graph structure of:

[0077] G=(N,E),

[0078] Among them, G represents the transportation task graph, N is the set of transportation nodes, and E is the set of transportation paths.

[0079] Model the transport node and set the transport node state vector:

[0080]

[0081] Among them, h n represents the state vector of node n at time step t, is the state at the previous moment, W h and W x is the weight matrix, b h is the bias term, σ(·) is the activation function,

[0082] Set the transport route nm Connect nodes n and m and calculate the weight. The calculation formula is:

[0083] e nm =α(h n ,h m ),

[0084] Among them, e nm represents the weight of the transport path, α(·) is the path weight calculation function,

[0085] Calculate the dynamic transportation path and use the path optimization algorithm to calculate the optimal path. The calculation formula is:

[0086]

[0087] Among them, P * is the optimal path, represents the set of all feasible paths, C(e nm ) represents the path e nm transportation costs.

[0088] As a preferred solution of the intelligent supply chain transportation management platform based on the Internet of Things described in the present invention, the steps of using the Internet of Things IoT device to monitor the transportation process in real time and detect the transportation environment are as follows:

[0089] Real-time monitoring of the transportation environment, using IoT devices to collect transportation environment data, and defining the transportation environment set as D:

[0090] D={d 1 , d 2 , ..., d y},

[0091] Where D represents the transportation environment data set, d z represents the zth monitoring indicator, y is the total number of monitoring indicators,

[0092] Build an anomaly detection model and set the threshold function A z :

[0093] A z =I(d z >τ z ),

[0094] Among them, A z represents the abnormal state of the zth monitoring indicator, I(·) is the indicator function, if d z Exceeding the threshold τ z , then A z =1 indicates abnormality,

[0095] When an anomaly is detected, go back to step S2 to re-plan the transportation task.

[0096] The beneficial effects of the present invention are as follows: the present invention utilizes historical transportation data, real-time order information and production schedule requirements to construct an input feature matrix, and adopts a time series prediction method based on a recurrent neural network (RNN) to improve the accuracy of transportation demand prediction and avoid inefficient allocation and delayed response of transportation resources. Based on the prediction results, a multi-agent collaborative scheduling module sets a transportation agent agent to monitor the real-time position of each component and predict possible transportation delays in combination with a time series model; the coordination agent dynamically adjusts the transportation plan based on the shortest path optimization algorithm; the customs agent handles cross-border customs clearance affairs in advance and registers the customs clearance status with the support of blockchain smart contracts, realizes automatic customs clearance inspection and abnormal triggering mechanism, and improves the stability and controllability of cross-border transportation.

[0097] In the present invention, before the transportation is executed, the blockchain smart contract management module synchronizes the optimized transportation path and customs clearance status information to the blockchain network, and when a customs clearance delay is detected, it automatically triggers the smart contract to adjust the transportation plan, thereby reducing the response time of human intervention; during the transportation process, the graph neural network GNN path optimization module constructs a transportation task graph based on the updated transportation plan, models different transportation nodes, and calculates the optimal transportation path in combination with weather, traffic flow and transportation mode availability, thereby improving the flexibility and real-time adaptability of the transportation network; when it is detected that a certain transportation path has a delay risk, it automatically adjusts to other transportation modes to improve the robustness of path planning.

[0098] The present invention continuously monitors the temperature, humidity, vibration and tilt angle environment during the transportation process, establishes an abnormality detection model, and sets a threshold mechanism. If a parameter exceeds the safety range, it automatically backtracks to the multi-agent collaborative scheduling module, triggering the agent to re-plan the transportation task and achieve dynamic adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0100] Figure 1 It is a structural schematic diagram of the intelligent supply chain transportation management platform based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0101] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0102] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0103] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0104] Example 1, reference Figure 1, this embodiment provides an intelligent supply chain transportation management platform based on the Internet of Things, including:

[0105] Intelligent forecasting module, which is used to collect historical transportation data, real-time order information and production schedule requirements; it uses the long short-term memory network deep learning model to predict future transportation needs;

[0106] Multi-agent collaborative scheduling module, which builds multi-agents based on transportation demand data, including transportation agent agents, coordination agents, and customs agents;

[0107] A transport agent that monitors the real-time location of each component and predicts possible transport delays;

[0108] Coordinate agents to adjust transportation plans based on the real-time location of each component;

[0109] Customs agents handle cross-border customs clearance matters in advance and monitor the progress of customs clearance;

[0110] Smart contract management module, used to register transportation routes and customs clearance status. During the customs clearance process, the smart contract automatically performs customs clearance progress checks. If a delay is detected, the multi-agent collaborative scheduling module is triggered to re-plan the transportation plan;

[0111] The dynamic path optimization module uses graph neural network (GNN) to build a transportation task graph, model different transportation nodes, and calculate dynamic transportation paths based on weather, traffic flow, and transportation mode availability;

[0112] The monitoring feedback module uses IoT devices to monitor the transportation process in real time. If the transportation environment is abnormal, it will automatically go back to the multi-agent collaborative scheduling module, and the multi-agent system will re-optimize the transportation task.

[0113] This embodiment also provides a management method for the above-mentioned intelligent supply chain transportation management platform based on the Internet of Things, including:

[0114] Step S1, collect historical transportation data, real-time order information and production schedule requirements, and predict future transportation demand through deep learning models to generate transportation demand data;

[0115] The steps to generate transportation demand data by predicting future transportation demand through deep learning models are as follows:

[0116] Collect historical transportation data, real-time order information and production schedule requirements, and construct an input feature matrix, which can be expressed as:

[0117] X={x 1 , x 2 , ..., x m},

[0118] Among them, X represents the input feature matrix, x j represents the jth feature, 1≤j≤m, m represents the total number of features,

[0119] A prediction model based on recurrent neural network (RNN) is used to define the state transfer equation, which is:

[0120] h t =f(W h h t-1 +W x X t +b h ),

[0121] Among them, h t represents the hidden state at time t, h t-1 Indicates the hidden state at the previous moment, W h and W x are the weight matrices of hidden state and input data, respectively, h is the bias term, f(·) represents the nonlinear activation function, X t is the input feature matrix at time t,

[0122] The fully connected layer is used for demand forecasting, and the prediction formula is:

[0123] y t =W y h t +b y ,

[0124] Among them, y t represents the transportation demand data predicted at time t, W y is the output layer weight matrix, b y is the output bias,

[0125] The mean square error MSE is used as the loss function, and the function formula is:

[0126]

[0127] Among them, L represents the loss value, T is the time step, For real transportation demand data,

[0128] The model parameters are updated through the back propagation algorithm, and the update formula is:

[0129]

[0130] Among them, θ is the model parameter, including W h , W x , W y , b h , b y , η is the learning rate;

[0131] Specifically, this step predicts future transportation demand based on a deep learning model, collects historical transportation data, real-time order information, and production schedule requirements, builds an input feature matrix, and models time series data through a recurrent neural network (RNN), calculates hidden states, and then predicts future transportation demand through a fully connected layer. The loss function uses mean square error (MSE), and the model parameters are optimized through the back propagation algorithm (BPTT), which effectively captures trend information in the time series and improves the prediction accuracy of future transportation demand.

[0132] Step S2, based on the transportation demand data of step S1, setting a transportation agent intelligent agent to monitor the real-time location of each component and predict possible transportation delays;

[0133] Coordinate intelligent agents to adjust transportation plans based on the real-time locations of each component and generate optimized transportation routes;

[0134] Customs agents handle cross-border customs clearance matters in advance, monitor customs clearance progress, and generate customs clearance status information;

[0135] Based on the transportation demand data of step S1, the steps of setting the transportation agent agent are:

[0136] Assume a set of transport agent agents, denoted as A:

[0137] A={a 1 , a 2 , ..., a p},

[0138] Among them, A represents the set of transport agent agents, a q represents the qth agent, 1≤q≤p, p represents the total number of agents,

[0139] Set up the component position matrix that the agent monitors:

[0140] P = {(x q ,y q , z q , t q )},

[0141] Where P represents the real-time position set of the components, (x q ,y q , z q ) represents the coordinates of the qth component in three-dimensional space, t q is the timestamp,

[0142] The time series model is used to predict possible transportation delays. The prediction formula is:

[0143] d q =g(P,Td ),

[0144] Among them, d q represents the expected transportation delay time of the qth component, g(·) is the delay prediction function, T d is the historical transportation time distribution,

[0145] Set up a coordination agent to adjust the transportation plan:

[0146] R = {r 1 , r 2 , ..., r s},

[0147] Among them, R represents the optimized transportation path set, r u represents the u-th path, 1≤u≤s, s is the total number of optional paths,

[0148] The shortest path optimization algorithm is used to optimize the transportation path. The optimization formula is:

[0149]

[0150] Among them, r * represents the optimal transportation path, C(r u , a q ) represents the transport agent agent a q On the path r u Transportation costs;

[0151] Specifically, this step sets up a transport agent based on transport demand data, monitors the real-time location of each component, and predicts possible transport delays. At the same time, the coordination agent adjusts the transport plan and optimizes the transport route based on the real-time location of each component. The transport agent is responsible for monitoring the real-time location of the component and predicting transport delays using time series analysis. The coordination agent selects the optimal path from the path set based on the shortest path optimization algorithm, dynamically adjusts the transport plan, and reduces the risk of transport delays.

[0152] Step S3, constructing a smart contract based on the optimized transportation route and customs clearance status information, and registering the transportation route and customs clearance status to the blockchain network;

[0153] In step S3, during the customs clearance process, the blockchain smart contract automatically performs a customs clearance progress check. If a delay is detected, it triggers the coordination agent in step S2 to re-plan the transportation plan and update the transportation plan;

[0154] In step S3, the steps of constructing a smart contract and automatically executing customs clearance progress check include:

[0155] Set the smart contract set C in the blockchain network:

[0156] C={c 1 , c 2 , ..., c v},

[0157] Among them, C represents the smart contract set, c w represents the wth smart contract, responsible for managing customs clearance and transportation information, v is the number of smart contracts,

[0158] Set the clearance status set S:

[0159] S={s 1 ,s 2 , ..., s k},

[0160] Among them, S represents the clearance state set, s x represents the clearance status of the xth level, k is the total number of levels,

[0161] Set up the smart contract trigger mechanism and set the event trigger conditions as:

[0162] E={e 1 , e 2 , ..., e k},

[0163] Among them, E represents the trigger event set, e x Represents the event triggered when level x is delayed.

[0164] Set the automatic execution logic of the smart contract:

[0165] If x Normal, then A(s x ) = Update status,

[0166] If x Delay, then A(s x ) = trigger transport adjustment,

[0167] Among them, A(s x ) represents the execution action of the smart contract at level x;

[0168] Specifically, this step builds smart contracts based on blockchain technology. The smart contract set is responsible for registering the customs clearance status and making dynamic adjustments based on the trigger mechanism. When the smart contract detects an abnormal customs clearance status, it automatically performs adjustment operations and triggers the re-planning of the transportation task.

[0169] Step S4, based on the updated transportation plan, a transportation task graph is constructed using a graph neural network, different transportation nodes are modeled, and the dynamic transportation path is calculated by combining weather, traffic flow, and transportation mode availability;

[0170] In step S4, when a transportation route is detected to have a risk of delay, it is automatically adjusted to another mode of transportation;

[0171] In step S4, different transport nodes include warehouses, ports, railway hubs and customs clearance stations;

[0172] Based on the updated transportation plan, a transportation task graph is constructed using a graph neural network to model different transportation nodes. The steps for calculating the dynamic transportation path are as follows:

[0173] Construct a transportation task graph and set the transportation network to have a graph structure of:

[0174] G=(N,E),

[0175] Among them, G represents the transportation task graph, N is the set of transportation nodes, and E is the set of transportation paths.

[0176] Model the transport node and set the transport node state vector:

[0177]

[0178] Among them, h n represents the state vector of node n at time step t, is the state at the previous moment, W h and W x is the weight matrix, b h is the bias term, σ(·) is the activation function,

[0179] Set the transport route nm Connect nodes n and m and calculate the weight. The calculation formula is:

[0180] e nm =α(h n ,h m ),

[0181] Among them, e nm represents the weight of the transport path, α(·) is the path weight calculation function,

[0182] Calculate the dynamic transportation path and use the path optimization algorithm to calculate the optimal path. The calculation formula is:

[0183]

[0184] Among them, P * is the optimal path, represents the set of all feasible paths, C(e nm ) represents the path e nm transportation costs;

[0185] Specifically, a graph neural network is used here to construct a transportation task graph, model different transportation nodes, and calculate dynamic transportation paths in combination with weather, traffic flow, and transportation mode availability. The set of transportation nodes is represented by a state vector, and the path set calculates the optimal transportation path through edge weights. The graph neural network can effectively model the transportation network structure and dynamically adjust the transportation plan when the traffic environment changes, thereby improving the computational efficiency of transportation path optimization.

[0186] Step S5: Based on the dynamic transportation path, IoT devices are used to monitor the transportation process in real time and detect the transportation environment. An intelligent feedback mechanism is set up. If an abnormality occurs at a transportation node, it will automatically go back to step S2, and the multi-agent collaborative scheduling module will re-plan the transportation task.

[0187] In step S5, the transport environment detected includes temperature, humidity, vibration and tilt angle;

[0188] Use IoT devices to monitor the transportation process in real time. The steps to detect the transportation environment are as follows:

[0189] Real-time monitoring of the transportation environment, using IoT devices to collect transportation environment data, and defining the transportation environment set as D:

[0190] D={d 1 , d 2 , ..., d y},

[0191] Where D represents the transportation environment data set, d z represents the zth monitoring indicator, y is the total number of monitoring indicators,

[0192] Build an anomaly detection model and set the threshold function A z :

[0193] A z =I(d z >τ z ),

[0194] Among them, A z represents the abnormal state of the zth monitoring indicator, I(·) is the indicator function, if d z Exceeding the threshold τ z , then A z =1 indicates abnormality,

[0195] When an anomaly is detected, go back to step S2 to re-plan the transportation task;

[0196] Specifically, IoT devices are used here to monitor the transportation environment in real time, and an intelligent feedback mechanism is set to detect abnormalities in transportation nodes. When an indicator in the monitored transportation environment data exceeds the threshold, an abnormal state is triggered, and the process goes back to step S2 to re-plan the transportation task and improve the reliability of supply chain transportation.

[0197] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent supply chain transportation management platform based on the Internet of Things, characterized by: include, Intelligent forecasting module, which is used to collect historical transportation data, real-time order information and production schedule requirements; it uses the long short-term memory network deep learning model to predict future transportation needs; Multi-agent collaborative scheduling module, which builds multi-agents based on transportation demand data, including transportation agent agents, coordination agents, and customs agents; Smart contract management module, used to register transportation routes and customs clearance status. During the customs clearance process, the smart contract automatically performs customs clearance progress checks. If a delay is detected, the multi-agent collaborative scheduling module is triggered to re-plan the transportation plan; The dynamic path optimization module uses graph neural network (GNN) to build a transportation task graph, model different transportation nodes, and calculate dynamic transportation paths based on weather, traffic flow, and transportation mode availability; The monitoring feedback module uses IoT devices to monitor the transportation process in real time. If the transportation environment is abnormal, it will automatically go back to the multi-agent collaborative scheduling module, and the multi-agent system will re-optimize the transportation task.

2. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 1, characterized in that: A transport agent that monitors the real-time location of each component and predicts possible transport delays; Coordinate agents to adjust transportation plans based on the real-time location of each component; Customs intelligent body handles cross-border customs clearance matters in advance and monitors the progress of customs clearance.

3. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 2, characterized in that: The platform's management approach includes: Step S1, collect historical transportation data, real-time order information and production schedule requirements, and predict future transportation demand through deep learning models to generate transportation demand data; Step S2, based on the transportation demand data of step S1, setting a transportation agent intelligent agent to monitor the real-time location of each component and predict possible transportation delays; The coordination agent adjusts the transportation plan based on the real-time location of each component and generates an optimized transportation path; The customs agent handles cross-border customs clearance matters in advance, monitors customs clearance progress, and generates customs clearance status information; Step S3, constructing a smart contract based on the optimized transportation route and customs clearance status information, and registering the transportation route and customs clearance status to the blockchain network; Step S4, based on the updated transportation plan, a transportation task graph is constructed using a graph neural network, different transportation nodes are modeled, and the dynamic transportation path is calculated by combining weather, traffic flow, and transportation mode availability; Step S5: Based on the dynamic transportation path, IoT devices are used to monitor the transportation process in real time and detect the transportation environment. An intelligent feedback mechanism is set up. If an abnormality occurs at a transportation node, it will automatically go back to step S2, and the multi-agent collaborative scheduling module will re-plan the transportation task. In step S5, the transport environment detected includes temperature and humidity, vibration and tilt angle.

4. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 3, characterized in that: The steps of predicting future transportation demand through a deep learning model and generating transportation demand data are as follows: Collect historical transportation data, real-time order information and production schedule requirements, and construct an input feature matrix, which can be expressed as: X={x1,x2,…,x m }, Among them, X represents the input feature matrix, x j represents the jth feature, 1≤j≤m, m represents the total number of features, The prediction model based on recurrent neural network RNN ​​is adopted to define the state transfer equation, which is: h t =f(W h h t-1 +W x X t +b h ), Among them, h t represents the hidden state at time t, h t-1 Indicates the hidden state at the previous moment, W h and W x are the weight matrices of hidden state and input data, respectively, h is the bias term, f(·) represents the nonlinear activation function, X t is the input feature matrix at time t, The fully connected layer is used for demand forecasting, and the prediction formula is: y t =W y h t +b y , Among them, y t represents the transportation demand data predicted at time t, W y is the output layer weight matrix, b y is the output bias, The mean square error MSE is used as the loss function, and the function formula is: Among them, L represents the loss value, T is the time step, For real transportation demand data, The model parameters are updated through the back propagation algorithm, and the update formula is: Among them, θ is the model parameter, including W h ,W x ,W y ,b h ,b y , η is the learning rate.

5. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 4, characterized in that: The steps of setting the transport agent agent based on the transport demand data of step S1 are: Assume a set of transport agent agents, denoted as A: <h2 style=";text-align:left;direction:ltr">A = {a1,a2,…,a<h2 style=";text-align:left;direction:ltr"> p <h2 style=";text-align:left;direction:ltr">}, Among them, A represents the set of transport agent agents, a q represents the qth agent, 1≤q≤p, p represents the total number of agents, Set up the component position matrix that the agent monitors: P={(x q ,y q ,z q ,t q )}, Where P represents the real-time position set of the components, (x q ,y q ,z q ) represents the coordinates of the qth component in three-dimensional space, t q is the timestamp, The time series model is used to predict possible transportation delays. The prediction formula is: d q =g(P,T d ), Among them, d q represents the expected transportation delay time of the qth component, g(·) is the delay prediction function, T d is the historical transportation time distribution, Set up a coordination agent to adjust the transportation plan: R＝{r1,r2,…,r s}， Among them, R represents the optimized transportation path set, r u represents the u-th path, 1≤u≤s, s is the total number of optional paths, The shortest path optimization algorithm is used to optimize the transportation path. The optimization formula is: Among them, r * represents the optimal transportation path, C(r u ,a q ) represents the transport agent agent a q On the path r u The transportation costs on.

6. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 5, characterized in that: In step S3, during the customs clearance process, the blockchain smart contract automatically performs a customs clearance progress check. If a delay is detected, it triggers the coordination agent in step S2 to re-plan the transportation plan and update the transportation plan.

7. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 6, characterized in that: In step S3, the steps of constructing a smart contract and automatically executing customs clearance progress check include: Set the smart contract set C in the blockchain network: C={c1,c2,…,c v }, Among them, C represents the smart contract set, c w represents the wth smart contract, responsible for managing customs clearance and transportation information, v is the number of smart contracts, Set the clearance status set S: S={s1,s2,…,s k }, Among them, S represents the clearance state set, s x represents the clearance status of the xth level, k is the total number of levels, Set up the smart contract trigger mechanism and set the event trigger conditions as: E={e1,e2,…,e k }, Among them, E represents the trigger event set, e x Represents the event triggered when level x is delayed. Set the automatic execution logic of the smart contract: If x Normal, then A(s x ) = Update status, If x Delay, then A(s x ) = trigger transport adjustment, Among them, A(s x ) represents the execution action of the smart contract at level x.

8. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 7, characterized in that: In step S4, when a transportation route is detected to have a risk of delay, it is automatically adjusted to another mode of transportation; In step S4, the different transportation nodes include warehouses, ports, railway hubs and customs clearance stations.

9. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 8, characterized in that: The steps of constructing a transportation task graph based on the updated transportation plan using a graph neural network, modeling different transportation nodes, and calculating a dynamic transportation path in combination with weather, traffic flow, and transportation mode availability are as follows: Construct a transportation task graph and set the transportation network to have a graph structure of: G=(N,E), Among them, G represents the transportation task graph, N is the set of transportation nodes, and E is the set of transportation paths. Model the transport node and set the transport node state vector: Among them, h n represents the state vector of node n at time step t, is the state at the previous moment, W h and W x is the weight matrix, b h is the bias term, σ(·) is the activation function, Set the transport route nm Connect nodes n and m and calculate the weight. The calculation formula is: in nm =α(h n ,h m ), Among them, e nm represents the weight of the transport path, α(·) is the path weight calculation function, Calculate the dynamic transportation path and use the path optimization algorithm to calculate the optimal path. The calculation formula is: Among them, P * is the optimal path, represents the set of all feasible paths, C(e nm ) represents the path e nm transportation costs.

10. The intelligent supply chain transportation management platform based on the Internet of Things as claimed in claim 9, characterized in that: The steps of using IoT devices to monitor the transportation process in real time and detect the transportation environment are as follows: Real-time monitoring of the transportation environment, using IoT devices to collect transportation environment data, and defining the transportation environment set as D: D={d1,d2,…,d y }, Where D represents the transportation environment data set, d z represents the zth monitoring indicator, y is the total number of monitoring indicators, builds an anomaly detection model, and sets the threshold function A z : A z =I(d z >t z ), Among them, A z represents the abnormal state of the zth monitoring indicator, I(·) is the indicator function, if d z Exceeding the threshold τ z , then A z =1 indicates abnormality, When an anomaly is detected, go back to step S2 to re-plan the transportation task.

Images