Collaborative management method of multimodal end-to-end supply chain based on container logistics

Abstract

The present invention relates to a method for collaborative management of an end-to-end supply chain of multimodal transport based on container logistics, and belongs to the field of supply chain management technology. The method comprises: virtualizing and integrating the dispersed cargo owner demand, carrier capacity, and transit point operation capacity resources through a cloud computing platform to form a shared resource pool, and building a dynamic collaborative network; intelligently matching and scheduling the shared resource pool through a contribution distribution mechanism to obtain the full-link status of the supply chain; dynamically optimizing routing based on the full-link status of the supply chain, and selecting the optimal path for the transportation mode of container logistics based on the business process of multimodal transport; automatically decomposing tasks through smart contracts and sending them to nodes, and automatically uploading credentials through RFID gates after the carrier completes the node task, and triggering the next node task after the contract verifies the credentials. Seamless connection and efficient collaboration between the nodes of the supply chain are achieved, and the transportation efficiency and reliability of container logistics are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of supply chain management, and in particular relates to a multimodal transport end-to-end supply chain collaborative management method based on container logistics. Background Art

[0002] Traditional container logistics management suffers from poor information exchange between shippers, carriers, and transshipment points. Shippers, carriers, and transshipment points use heterogeneous systems (EDI / email / paper documents), requiring manual data transcription. This leads to order processing delays, a lack of real-time status visibility, and a lack of cross-platform sharing of IoT data such as location and container temperature. Under the traditional bidding model, carrier selection is time-consuming, and capacity remains idle due to information asymmetry, resulting in low transshipment equipment utilization. Meanwhile, the vacancy rate of alternative terminals is high during certain periods, and regional resource coordination is severely insufficient. This results in inefficient resource matching, high transportation costs, and difficulty in responding to unexpected risk events. Furthermore, insufficient automation across various processes relies on manual coordination, increasing the risk of operational errors and delivery delays. Therefore, the market urgently needs innovative supply chain collaborative management methods to achieve efficient resource integration and optimal allocation, improving transportation efficiency and reliability. Summary of the Invention

[0003] To solve the above problems existing in the prior art, the present invention provides a multimodal transport end-to-end supply chain collaborative management method based on container logistics. The purpose of the present invention can be achieved through the following technical solutions:

[0004] S1: Through the cloud computing platform, the dispersed shipper demands, carrier capacity, and transit point operational capabilities are virtualized and integrated into a shared resource pool, and a dynamic collaborative network is built;

[0005] S2: Based on the dynamic collaborative network, the shipper demand, carrier capacity and port operation capacity in the shared resource pool are intelligently matched and scheduled using a non-dominated sorting genetic algorithm and a fuzzy comprehensive evaluation method to obtain the full-link status of the supply chain;

[0006] S3: Dynamically optimize the route based on the full-link status of the supply chain, select the optimal path for container logistics from the starting point to the destination based on the multimodal transport business process, and update the transportation plan;

[0007] S4: The task allocation logic is defined through the smart contract rule library. According to the task allocation logic, the tasks are automatically decomposed and sent to the nodes through the smart contract. After the carrier completes the node task, the certificate is automatically uploaded through the RFID gate. After the contract verifies the certificate, it triggers the next node task.

[0008] Specifically, the shared resource pool in S1 encapsulates shipper demands, carrier capacity, and transit point operation capabilities into standardized microservice modules through containerization technology, and supports elastic expansion.

[0009] Specifically, the method for constructing the dynamic collaborative network in S1 is:

[0010] S101: Based on a graph computing engine, it maps the multi-dimensional relationships between shippers, carriers, and transit points in real time, including geographic location matching, historical cooperation credit scores, and real-time service capability thresholds.

[0011] S102: Collaboratively modeling the multi-source heterogeneous data of the multi-dimensional association relationship to generate a resource matching weight matrix;

[0012] S103: Responding to resource status changes through an event-driven architecture, dynamically adjusting the collaborative network topology according to the resource matching weight matrix, and automatically triggering capacity reallocation when the transfer point's operating capacity is overloaded.

[0013] Specifically, S2 includes:

[0014] S201: Construct a multi-objective optimization function, where the target variables include minimizing transportation costs, optimizing carbon emission intensity, and maximizing on-time delivery rate;

[0015] S202: Solve the Pareto frontier through the non-dominated sorting genetic algorithm to generate a set of candidate solutions for resource matching;

[0016] S203: Selecting a matching result with the highest comprehensive score from the candidate solution set based on a fuzzy comprehensive evaluation method, and dynamically updating the contribution weight.

[0017] Specifically, the full-link status of the supply chain in S2 is used to integrate the shipper's order data, the carrier's GPS positioning, and the real-time monitoring data of the IoT devices at the transit point, and store the dynamic data stream through the time series database, and use the streaming computing platform to calculate the key status parameters in real time; the key status parameters include the transportation time deviation rate, the container detention rate, and the equipment failure rate.

[0018] Specifically, the dynamic routing optimization method in S3 is:

[0019] A multimodal transport routing decision model is constructed based on reinforcement learning. Input parameters include real-time transportation cost, congestion probability of transit nodes, carbon emission intensity threshold, and shipper priority weight. The model outputs the optimal solution set for the combined road, rail, and waterway transport routes.

[0020] By monitoring sudden risk events, dynamic path switching is automatically triggered, adjustment instructions are sent to affected nodes through blockchain smart contracts, and the full-link status data is updated synchronously.

[0021] Specifically, the triggering conditions for the dynamic path switching include: the transfer point's operating capacity is overloaded; the transportation time deviation rate continuously exceeds the planned value; and the Internet of Things device detects the risk of path interruption due to extreme weather.

[0022] Specifically, the multimodal transport business process in S3 includes:

[0023] The order receipt, cargo loading, transshipment, transportation tracking and delivery confirmation of road-rail-water transport, all business process links are automated through Internet of Things technology and blockchain smart contracts.

[0024] Specifically, the task allocation logic in S4 adopts a layered rule engine architecture, which includes a transportation mode identification layer, a resource constraint verification layer, and a dynamic priority adjustment layer; the rule base has a built-in fuzzy logic controller to analyze shipper priorities, carrier service level agreements, and real-time load status of transit points. The rule trigger mechanism adopts an event-driven mode to decompose end-to-end transportation into atomic operation units, and each atomic task is encapsulated as a smart contract object.

[0025] Specifically, the smart contract in S4 automatically executes the process management after task decomposition, including task progress monitoring, exception handling and automatic settlement. The smart contract deconstructs the task allocation of multimodal transport and extracts trigger conditions, execution actions and constraint rules; the data structure of the smart contract is a data model for on-chain and off-chain collaboration. The on-chain storage layer stores key credentials, and the off-chain computing layer is used to access external data sources. The state migration path of quotation submission, matching, transaction settlement, and evaluation feedback is defined through a finite state machine diagram.

[0026] The beneficial effects of the present invention are:

[0027] By integrating advanced technologies such as cloud computing, the Internet of Things, blockchain, and reinforcement learning, efficient, collaborative management of the end-to-end supply chain for container logistics in multimodal transport scenarios is achieved. This approach not only improves the utilization of logistics resources and reduces transportation costs, but also significantly enhances the flexibility and risk resilience of the supply chain. In the face of emergencies such as overloaded transit points, delivery time deviations, or extreme weather, the system can automatically trigger dynamic route switching to ensure the continuity and punctuality of cargo transportation. Furthermore, through the application of smart contracts, automated management of task allocation, progress monitoring, exception handling, and automatic settlement is achieved, further improving the transparency and efficiency of logistics operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0029] Figure 1Schematic diagram of the processing sequence of the multimodal transport end-to-end supply chain collaborative management method based on container logistics of the present invention;

[0030] Figure 2 This is a schematic diagram of the multi-source data collaborative modeling process in the present invention;

[0031] Figure 3 This is a structural diagram of the on-chain and off-chain collaboration mechanism in the present invention. DETAILED DESCRIPTION

[0032] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0033] See also Figure 1-3 , a multimodal end-to-end supply chain collaborative management method based on container logistics, including:

[0034] S1: Through the cloud computing platform, the dispersed shipper demands, carrier capacity, and transit point operational capabilities are virtualized and integrated into a shared resource pool, and a dynamic collaborative network is built;

[0035] S2: Based on the dynamic collaborative network, the shipper demand, carrier capacity and port operation capacity in the shared resource pool are intelligently matched and scheduled using a non-dominated sorting genetic algorithm and a fuzzy comprehensive evaluation method to obtain the full-link status of the supply chain;

[0036] S3: Dynamically optimize the route based on the full-link status of the supply chain, select the optimal path for container logistics from the starting point to the destination based on the multimodal transport business process, and update the transportation plan;

[0037] S4: The task allocation logic is defined through the smart contract rule library. According to the task allocation logic, the tasks are automatically decomposed and sent to the nodes through the smart contract. After the carrier completes the node task, the certificate is automatically uploaded through the RFID gate. After the contract verifies the certificate, it triggers the next node task.

[0038] Specifically, the shared resource pool in S1 encapsulates shipper demands, carrier capacity, and transit point operation capabilities into standardized microservice modules through containerization technology, and supports elastic expansion.

[0039] In this embodiment, a three-in-one architecture of "microservices + container orchestration + elastic scheduling" is adopted to achieve dynamic integration of shipper demand, carrier capacity, and transit point operation capabilities. The microservice module includes a shipper demand analysis module (which analyzes shipper demand, outputs structured data, and encapsulates it as a Docker container), a capacity registration center (which integrates data streams from on-board IoT devices to build a dynamic capacity profile), and a transit operation adapter. The elastic expansion system of this embodiment includes a three-level response strategy to meet multi-level expansion requirements, from burst demand in seconds to seasonal fluctuations. When the regional resource pool is saturated, the load is migrated to other availability zones through Kubernetes Federation.

[0040] Specifically, the method for constructing the dynamic collaborative network in S1 is:

[0041] S101: Based on a graph computing engine, it maps the multi-dimensional relationships between shippers, carriers, and transit points in real time, including geographic location matching, historical cooperation credit scores, and real-time service capability thresholds.

[0042] S102: Collaboratively modeling the multi-source heterogeneous data of the multi-dimensional association relationship to generate a resource matching weight matrix;

[0043] S103: Responding to resource status changes through an event-driven architecture, dynamically adjusting the collaborative network topology according to the resource matching weight matrix, and automatically triggering capacity reallocation when the transfer point's operating capacity is overloaded.

[0044] In this embodiment, a three-in-one architecture of "graph computing engine + stream processing platform + dynamic weight matrix" is adopted to realize multi-dimensional resource association modeling and dynamic collaborative optimization. The multi-source data collaborative modeling process is as follows: Figure 2 As shown in the figure; Dynamically adjust the collaborative network topology structure based on the reinforcement learning decision model:

[0045] ,

[0046] Among them, α, β, and γ are multi-objective weights, the state space s contains 128-dimensional features such as the current topology structure and resource utilization, and the action space a includes operations such as node addition and deletion and edge weight adjustment.

[0047] Specifically, S2 includes:

[0048] S201: Construct a multi-objective optimization function, where the target variables include minimizing transportation costs, optimizing carbon emission intensity, and maximizing on-time delivery rate;

[0049] S202: Solve the Pareto frontier through the non-dominated sorting genetic algorithm to generate a set of candidate solutions for resource matching;

[0050] S203: Selecting a matching result with the highest comprehensive score from the candidate solution set based on a fuzzy comprehensive evaluation method, and dynamically updating the contribution weight.

[0051] In this embodiment, the three-dimensional objective space of the multi-objective optimization function can be expressed as:

[0052] ,

[0053] Where f1 represents the total transportation cost, including the fixed cost C fixed , variable cost C variable , transfer operation fee γ i , f2 is the on-time delivery rate; d represents the degree of deviation between the actual transportation time of the goods from the starting point to the destination and the planned transportation time, T transfer Indicates the time required for transfer operation, t actual Indicates the actual transportation time, t plan Indicates the planned transportation time;

[0054] The constraints are modeled as:

[0055] ,

[0056] Combined with NSGA-II Pareto front generation, SBX was used to simulate binary crossover (crossover rate 0.8) and polynomial mutation (mutation rate 0.1). 200 initial solutions were generated through Latin hypercube sampling to ensure uniform spatial distribution. Real number encoding was used to represent transport path selection. The gene structure example is as follows:

[0057] ,

[0058] The fuzzy membership function of fuzzy comprehensive evaluation and weight update can be constructed based on the fuzzy optimization method as follows:

[0059] ,

[0060] Among them, the ideal value Take the optimal value of each objective in the Pareto solution set;

[0061] The comprehensive scoring adopts the fuzzy decision matrix:

[0062] ,

[0063] Among them, w m is the dynamic weight, λ is the historical performance impact factor, Indicates the carrier's historical on-time performance.

[0064] Specifically, the full-link status of the supply chain in S2 is used to integrate the shipper's order data, the carrier's GPS positioning, and the real-time monitoring data of the IoT devices at the transit point, and store the dynamic data stream through the time series database, and use the streaming computing platform to calculate the key status parameters in real time; the key status parameters include the transportation time deviation rate, the container detention rate, and the equipment failure rate.

[0065] This embodiment is compatible with shipper EDI / API order data (JSON / EDIFACT), vehicle-mounted GPS positioning messages (NMEA0183 standard), and transit point IoT device data (Modbus / OPC UA protocol); it uses the Flink+Spark StructuredStreaming hybrid engine to achieve real-time computing with sub-second latency.

[0066] Specifically, the dynamic routing optimization method in S3 is:

[0067] A multimodal transport routing decision model is constructed based on reinforcement learning. Input parameters include real-time transportation cost, congestion probability of transit nodes, carbon emission intensity threshold, and shipper priority weight. The model outputs the optimal solution set for the combined road, rail, and waterway transport routes.

[0068] By monitoring sudden risk events, dynamic path switching is automatically triggered, adjustment instructions are sent to affected nodes through blockchain smart contracts, and the full-link status data is updated synchronously.

[0069] In this embodiment, real-time parameters such as transportation cost (API connection to bank exchange rate), congestion probability (real-time traffic conditions on AutoNavi Maps), and shipper weight (EDI order metadata) are integrated. When a risk event is detected, a delay in route switching is triggered. The main chain stores route strategy metadata, and the side chain handles regional transport certificate verification. The multimodal transport route decision model is based on Markov decision process modeling:

[0070] ,

[0071] Among them, C t represents the transportation cost, P j represents the congestion probability of transit node j, W m Indicates the priority weight of the shipper.

[0072] Discrete transportation events are encoded into a continuous vector space. Without sharing the original data, the effectiveness of the path optimization plan is verified across enterprises. The reward function parameters are adjusted in real time based on the shipper's weight to achieve personalized optimization.

[0073] Specifically, the triggering conditions for the dynamic path switching include: the transfer point's operating capacity is overloaded; the transportation time deviation rate continuously exceeds the planned value; and the Internet of Things device detects the risk of path interruption due to extreme weather.

[0074] Specifically, the multimodal transport business process in S3 includes:

[0075] The order receipt, cargo loading, transshipment, transportation tracking and delivery confirmation of road-rail-water transport, all business process links are automated through Internet of Things technology and blockchain smart contracts.

[0076] Specifically, the task allocation logic in S4 adopts a layered rule engine architecture, which includes a transportation mode identification layer, a resource constraint verification layer, and a dynamic priority adjustment layer; the rule base has a built-in fuzzy logic controller to analyze shipper priorities, carrier service level agreements, and real-time load status of transit points. The rule trigger mechanism adopts an event-driven mode to decompose end-to-end transportation into atomic operation units, and each atomic task is encapsulated as a smart contract object.

[0077] Specifically, the smart contract in S4 automatically executes the process management after task decomposition, including task progress monitoring, exception handling and automatic settlement. The smart contract deconstructs the task allocation of multimodal transport and extracts trigger conditions, execution actions and constraint rules; the data structure of the smart contract is a data model for on-chain and off-chain collaboration. The on-chain storage layer stores key credentials, and the off-chain computing layer is used to access external data sources. The state migration path of quotation submission, matching, transaction settlement, and evaluation feedback is defined through a finite state machine diagram.

[0078] In this embodiment, the smart contract is written in Solidity language and supports blockchain platforms such as EVM (Ethereum Virtual Machine) and Hyperledger Fabric; the deployment and upgrade of the smart contract are completed automatically through the CI / CD pipeline to ensure the consistency and security of the code version. The task progress monitoring module integrates the blockchain event monitoring mechanism to capture task status change events in real time and trigger the corresponding processing logic. The exception handling module automatically diagnoses and repairs common problems in the transportation process, such as vehicle failure and cargo damage, based on the predefined exception handling strategy library. The automatic settlement module automatically calculates the transportation fee according to the settlement rules defined in the smart contract, and realizes the instant transfer of funds through the blockchain wallet. The evaluation feedback module collects service evaluation data of shippers, carriers and transit points, and provides a reference for subsequent transportation decisions through data analysis. On-chain and off-chain collaborative mechanisms such as Figure 3 As shown, the on-chain credentials are stored using Merkle Patricia Tree, and the off-chain is connected to the Amap real-time traffic API.

[0079] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A multimodal transport end-to-end supply chain collaborative management method based on container logistics, characterized by: include: S1: Through the cloud computing platform, the dispersed shipper demands, carrier capacity, and transit point operational capabilities are virtualized and integrated into a shared resource pool, and a dynamic collaborative network is built; S2: Based on the dynamic collaborative network, the shipper demand, carrier capacity and port operation capacity in the shared resource pool are intelligently matched and scheduled using a non-dominated sorting genetic algorithm and a fuzzy comprehensive evaluation method to obtain the full-link status of the supply chain; S3: Dynamically optimize the route based on the full-link status of the supply chain, select the optimal path for container logistics from the starting point to the destination based on the multimodal transport business process, and update the transportation plan; The dynamic routing optimization method is to build a multimodal transport path decision model based on reinforcement learning, with input parameters including real-time transportation cost, transfer node congestion probability, carbon emission intensity threshold and shipper priority weight, and output the optimal solution set of combined road, rail and water transport paths; By monitoring sudden risk events, dynamic path switching is automatically triggered, adjustment instructions are sent to affected nodes through blockchain smart contracts, and the status data of the entire link is updated synchronously; S4: The task allocation logic is defined through the smart contract rule base. According to the task allocation logic, the smart contract automatically decomposes the tasks and distributes them to the nodes. After the carrier completes the node task, the certificate is automatically uploaded through the RFID gate. The contract verifies the certificate and triggers the next node task. The task allocation logic uses a layered rules engine architecture, including a transport mode identification layer, a resource constraint verification layer, and a dynamic priority adjustment layer. The rule base has a built-in fuzzy logic controller that analyzes shipper priorities, carrier service level agreements, and the real-time load status of transit points. The rule triggering mechanism uses an event-driven model, breaking down end-to-end transportation into atomic operation units, and encapsulating each atomic task as a smart contract object. The smart contract automatically executes process management after task decomposition, including task progress monitoring, exception handling and automatic settlement. The smart contract deconstructs the task allocation of multimodal transport and refines the trigger conditions, execution actions and constraint rules. The data structure of the smart contract is a data model for on-chain and off-chain collaboration. The on-chain storage layer stores key credentials, and the off-chain computing layer is used to access external data sources. The state transition path of quotation submission, matching, transaction settlement, and evaluation feedback is defined through a finite state machine diagram.

2. The method according to claim 1, characterized in that The shared resource pool in S1 encapsulates shipper demands, carrier capacity and transit point operation capabilities into standardized microservice modules through containerization technology, and supports elastic expansion.

3. The method according to claim 1, characterized in that The method for constructing the dynamic collaborative network in S1 is: S101: Based on a graph computing engine, real-time mapping of multi-dimensional relationships between shippers, carriers, and transit points, including geographic location matching, historical cooperation credit scores, and real-time service capability thresholds; S102: Collaboratively modeling the multi-source heterogeneous data of the multi-dimensional association relationship to generate a resource matching weight matrix; S103: Responding to resource status changes through an event-driven architecture, dynamically adjusting the collaborative network topology according to the resource matching weight matrix, and automatically triggering capacity reallocation when the transfer point's operating capacity is overloaded.

4. The method according to claim 1, wherein S2 include: S201: Construct a multi-objective optimization function, where the target variables include minimizing transportation costs, optimizing carbon emission intensity, and maximizing on-time delivery rate; S202: Solve the Pareto frontier through the non-dominated sorting genetic algorithm to generate a set of candidate solutions for resource matching; S203: Selecting a matching result with the highest comprehensive score from the candidate solution set based on a fuzzy comprehensive evaluation method.

5. The method according to claim 1, wherein The full-link status of the supply chain in S2 is used to integrate the shipper's order data, the carrier's GPS positioning, and the real-time monitoring data of the IoT devices at the transit point, and store the dynamic data stream through the time series database, and use the streaming computing platform to calculate the key status parameters in real time; the key status parameters include the transportation time deviation rate, the container detention rate, and the equipment failure rate.

6. The method according to claim 1, characterized in that The triggering conditions for the dynamic path switching include the following: the transfer point's operating capacity is overloaded; the transportation time deviation rate continuously exceeds the planned value; and the IoT device detects the risk of path interruption caused by extreme weather.

7. The method according to claim 1, characterized in that The multimodal transport business process in S3 includes: The order receipt, cargo loading, transshipment, transportation tracking and delivery confirmation of road-rail-water transport, all business process links are automated through Internet of Things technology and blockchain smart contracts.

Images