E-commerce big data logistics supply chain control system based on artificial intelligence and block chain

Abstract

The invention relates to the technical field of e-commerce logistics supply chain management, and particularly discloses an e-commerce big data logistics supply chain control system based on artificial intelligence and a block chain. The system comprises a block chain network formed by supply chain participant nodes, an artificial intelligence analysis engine and an intelligent contract system deployed on the chain. The intelligent contract system comprises a rule management contract and a task execution contract. The rule management contract maintains an updatable rule base defining decision logic and an evidence structure. The task execution contract is used for calling structured data provided by the artificial intelligence engine according to a rule in an actual logistics task, and generating a resource scheduling instruction by calculating or organizing a node dynamic weight consensus on a chain defined by the execution rule. According to the system, decision logic is transparentized, consensus and optimizable, so that supply chain cooperative control with credible data, balanced benefits and dynamic adaptation in a multi-party participation environment is realized.

Description

Technical Field

[0001] This invention relates to the field of e-commerce logistics supply chain management technology, and more specifically, to an e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain. Background Technology

[0002] As e-commerce continues to expand, its underlying logistics supply chain system is becoming increasingly complex, involving collaboration among multiple parties and stages, including warehousing, transportation, and delivery. Against this backdrop, efficiently integrating resources from all parties, ensuring reliable information flow, and achieving dynamic global optimization have become crucial for improving the efficiency and reliability of e-commerce logistics, and are also important directions for the industry's intelligent upgrade.

[0003] Currently, e-commerce logistics supply chain management largely relies on centralized platform systems or core enterprise-led collaborative models. While this model is efficient in data aggregation and instruction distribution, it also has significant limitations. First, there are inherent differences in interests and information asymmetry among the various participants (such as suppliers and logistics service providers). Centralized systems struggle to fairly balance the interests of all parties in decision-making, easily leading to localized resource misallocation and collaborative friction. Second, supply chain data is distributed across the internal systems of different entities. Centralized platforms cannot ensure the complete reliability and immutability of the aggregated data, lacking universally recognized authoritative evidence for dispute attribution and performance evaluation. Third, in the face of sudden fluctuations (such as order surges or capacity shortages), existing systems typically rely on preset rules or manual intervention for scheduling, lacking adaptive and refined decision-making capabilities based on real-time global status, resulting in limited response speed and optimization depth. Finally, the system's decision-making logic is usually statically set by the platform, making it difficult to continuously and automatically evolve and optimize based on historical performance.

[0004] Therefore, this paper proposes an e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain to address the above issues. The problems to be solved include: existing technical solutions face challenges in achieving reliable collaboration across entities, dynamic decision-making with balanced interests, and autonomous optimization of system rules, which restricts the further improvement of the overall efficiency and resilience of the e-commerce logistics supply chain. Summary of the Invention

[0005] The technical problem this invention aims to solve is that existing e-commerce logistics supply chain management systems mainly rely on centralized platforms for data collection and scheduling, resulting in drawbacks such as difficulties in multi-party collaboration, opaque decision-making processes, poor rule adaptability, and high dispute resolution costs. Specifically, data sharing and mutual trust among different participants are difficult; scheduling decisions often favor the platform's own interests, failing to fairly reflect the reasonable demands and capabilities of carriers, warehousing service providers, and other parties; pre-set static rules cannot cope with sudden fluctuations and complex network-based cascading effects; and dispute resolution relies on inefficient manual arbitration.

[0006] To address the aforementioned technical challenges, this invention provides an e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain. This system achieves automated, transparent, and adaptive collaborative control of the supply chain by constructing a technical architecture that integrates trusted data storage, intelligent decision support, programmable rule execution, and decentralized governance.

[0007] The technical solution of the present invention is as follows: An e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain, the system comprising: A permissioned blockchain network comprised of nodes from supply chain participants is used to record and verify logistics status events and decision data in a distributed manner, forming an immutable operational log.

[0008] An artificial intelligence analytics engine interacts with data from the blockchain network to process on-chain and off-chain supply chain big data and provide structured data support for the decision-making process.

[0009] A smart contract system deployed on the blockchain network serves as the core execution layer for business logic; the smart contract system includes a rule management contract and multiple task execution contracts.

[0010] The working principle of the system and the collaborative relationship between its components are as follows: The rule management contract is used to store and maintain a dynamically updated decision rule library. Each decision rule in the library is strictly associated with a specific business decision scenario (e.g., "order diversion during warehouse overload", "multi-carrier bidding selection", "determination of responsibility for transportation delays").

[0011] Each rule not only defines its applicable scenario but also explicitly defines, in a machine-readable manner, the input data structure required for decision-making in that scenario (i.e., a standardized description of the input data format, fields, and types), as well as the logical processing procedure employed in that scenario. This logical processing procedure is essentially a piece of executable code that specifies how, based on compliant input data, a deterministic resource scheduling instruction or responsibility determination result is ultimately generated through computation or consensus.

[0012] The task execution contract is uniquely bound to each specific logistics task (such as a single express delivery order or a batch of freight consignments) created in the system, serving as its digital agent for its entire lifecycle on the blockchain. When the logistics task is actually executed, triggered by an IoT device or business system, and reaches a preset decision point (e.g., warehouse allocation after order creation, or route selection after goods leave the warehouse), the task execution contract bound to that task is activated. Its workflow is as follows: 1. Rule Query: Based on the current decision point type, the contract queries the rule management contract for the corresponding decision rule to find out what format of data is required for this decision and what logic to use to process it.

[0013] 2. Data Acquisition: Based on the queried input data structure description, the contract generates a standardized data request and sends it to the artificial intelligence analysis engine through the blockchain oracle mechanism.

[0014] 3. Data Supply and Processing: Upon receiving a request, the AI ​​analysis engine initiates the corresponding data processing pipeline. It retrieves relevant historical, reliable data from the blockchain network and real-time status information from authorized off-chain systems, processing this data to generate the first type of data (objective factual data, such as historical on-time delivery rates and current inventory levels). Simultaneously, the engine utilizes its built-in simulation and prediction models to extrapolate different potential decision options, assessing their potential systemic impact (such as the impact on the load of other nodes in the network and potential risks to efficiency in future periods), and quantifies these assessment results into the second type of data (predictive impact data). The engine then packages and signs these two types of data according to the requested format and returns them as pending data to the task execution contract.

[0015] 4. Decision Execution: After receiving and verifying the data to be processed, the task execution contract immediately executes the logical processing procedures obtained from the rule management contract. This process, based on the rule design, can be divided into two modes: Direct computation mode: Suitable for scenarios with clear objectives and quantifiable optimization. The contract calls an embedded algorithm (such as a multi-attribute utility function) to calculate the input data and directly outputs the optimal resource identifier (such as "select warehouse A").

[0016] Consensus Voting Mode: Suitable for scenarios involving the division of responsibilities or the balancing of interests. The contract generates a proposal containing several options based on the data to be processed and invites a group of relevant participating nodes (such as shippers, carriers, and consignee representatives) to form a temporary consensus group to vote. The weight of each voting node is not fixed but dynamically calculated based on the quality (credibility) of its historical data contributions and its relevance to the current event. Subsequently, the contract forms a final resolution based on the weighted voting results.

[0017] 5. Instruction Generation and Driving: After the decision is executed, the task execution contract confirms the result (whether calculated or reached through consensus) as a formal resource scheduling instruction and records it as an event on the blockchain. This event will be monitored and automatically executed by logistics operation systems (such as Warehouse Management System (WMS) and Transportation Management System (TMS), thereby driving physical logistics resources to complete the corresponding operations.

[0018] Furthermore, the logical processing includes direct computation logic and consensus voting logic.

[0019] The direct calculation logic is configured to process the data to be processed according to the algorithm embedded in the rules, and directly output a definite resource scheduling instruction. For example, in a path selection scenario, the algorithm can perform a weighted summation of multi-dimensional data such as cost, distance, and timeliness, and use the highest total score as the output.

[0020] The consensus voting logic is configured as follows: a decision proposal containing at least two options is generated based on the data to be processed, and the decision proposal is submitted to a consensus group composed of relevant nodes for weighted voting to form a resolution; wherein, the voting weight value of each node in the consensus group is determined according to a preset dynamic calculation rule; the task execution contract generates resource scheduling instructions based on the weighted voting results.

[0021] Furthermore, the dynamic calculation rule for the voting weight of each node in the consensus group is determined based on at least one of the following: the number or proportion of data items provided by the node historically that have been verified as authentic by the blockchain network and adopted by the final decision; and whether the node has a direct connection with the resources or responsible parties involved in the current decision proposal. This design binds a node's voting rights to its historical integrity record and current conflict of interest status, incentivizing nodes to provide authentic data through algorithms and automatically reducing their influence in the event of a conflict of interest, thereby ensuring the fairness of the consensus at the program level.

[0022] Furthermore, the data to be processed generated by the artificial intelligence analysis engine includes a first type of data and a second type of data.

[0023] The first type of data is based on statistics of verified historical data in the blockchain network, or obtained by collecting and verifying authorized off-chain real-time data sources. Its purpose is to provide a verifiable and objective factual basis for decision-making.

[0024] The second type of data is generated by the AI ​​analysis engine through analysis of the first type of data and the supply chain network topology. It is used to predict the potential impact of different resource scheduling instructions on the overall subsequent state of the supply chain network. Its function is to extend the decision-making perspective from the current local state to the future global impact, supporting more resilient optimization decisions.

[0025] Furthermore, for decision-making scenarios aimed at optimizing the overall performance of the supply chain network, the logical processing defined by the associated decision rules is configured such that, when evaluating candidate instructions, indicators reflecting systemic impact in the second type of data are given higher priority. For example, in warehouse allocation decisions, a solution that could lead to severe congestion in a certain region's distribution network, even if its individual warehouse costs the lowest, will be preferentially excluded by the rules because it triggers systemic risk indicators.

[0026] The self-optimization and evolution mechanism of the system is as follows: The rule management contract is a static rule store and also a governance contract that supports dynamic upgrades. It is configured to periodically initiate an optimization process for the decision rule base based on feedback data on the actual execution effects of a large number of historical decision cases recorded on the blockchain (e.g., changes in average performance time after the application of a certain sub-account rule, or the frequency of subsequent arbitrations triggered by a certain liability determination rule).

[0027] Furthermore, the rule management contract optimizes and updates decision rules through an on-chain governance mechanism, including a simulation evaluation step: It invokes the artificial intelligence analysis engine to simulate the operation of candidate new rules and the original rules to be updated, based on historical data snapshots; it compares and analyzes the differences in key performance indicators generated by the candidate new rules and the original rules to be updated during the simulation; and it submits a simulation analysis report containing quantitative comparison results to the blockchain network for relevant nodes to vote on whether to adopt the candidate new rules. This step moves the risk of rule changes from the production environment to the sandbox environment and supports governance decisions through quantitative data, making the system's evolution process more rational and secure.

[0028] Furthermore, the conditions that trigger the rule management contract to initiate the optimization and update process for a specific decision rule are at least one of the following: detecting that the occurrence rate of subsequent on-chain dispute events associated with the decision rule exceeds a preset threshold; or, sensing a preset type of change in the supply chain network structure. This makes the system's optimization behavior responsive, automatically driven by actual operational problems or environmental changes.

[0029] Furthermore, at least some decision rules in the decision rule base are configured to include incentive parameters, which are used to adjust the evaluation of node behavior during logical processing; the rule management contract's optimization and updates of the decision rules include adjustments to the incentive parameters. For example, in the service provider rating rules, adjusting the weight coefficient of the "historical reputation" item can guide service providers to focus more on long-term reputation accumulation rather than short-term low-price competition. The adjustment of the incentive parameters is based on simulating the expected behavior choices of nodes in historical scenarios under different parameter values ​​and the resulting global effects through the artificial intelligence analysis engine, and selecting the parameter value combination that makes the preset global optimization goal better.

[0030] The system's efficient dispute resolution mechanism is as follows: The system incorporates a decentralized arbitration contract. When a participant disagrees with the instructions or status generated by the task execution contract, they can file a complaint with the arbitration contract.

[0031] Furthermore, the arbitration contract is configured to be initiated in response to a complaint and to perform the following operations: extract all data records related to the complaint from the blockchain network; organize an arbitration panel composed of nodes with no direct interest in the complaint to make a ruling; wherein the ruling weight of each member of the arbitration panel is determined based on their historical record of impartiality in arbitration; the ruling result and its basis are recorded on the blockchain, triggering an update to the status of the relevant task execution contract. This mechanism utilizes the notarization advantages of blockchain to solidify evidence, ensures the independence and authority of arbitrators through random selection and reputation weighting, and automatically executes the ruling result through smart contracts, achieving full automation and decentralization of dispute resolution.

[0032] The beneficial effects that can be achieved by adopting the technical solution of this invention include: 1. Establish a trusted foundation for collaboration: Through a permissioned blockchain network, all participants synchronize information on a shared, immutable ledger, fundamentally solving the problems of data consistency and trust in multi-party collaboration and reducing the cost of mutual trust.

[0033] 2. Achieving Intelligent and Balanced Decision-Making: The AI ​​analytics engine provides real-time, objective factual data and forward-looking impact prediction data, making decision-making more comprehensive and scientific. By codifying the decision-making logic into publicly auditable on-chain rules and introducing a consensus mechanism based on dynamic weights, the decision-making process is both automated and intelligent, and can more fairly reflect the interests and capabilities of multiple parties, reducing centralized bias.

[0034] 3. Improving System Adaptability: Through a dynamic optimization and update mechanism supported by rule-managed contracts, the system's decision-making rules are no longer fixed but can iteratively evolve based on actual operational data. This enables the system to adapt to market changes, business growth, and the evolution of participant behavior patterns, thus possessing long-term viability.

[0035] 4. Provides efficient dispute resolution methods: The dispute resolution mechanism based on blockchain evidence storage and procedural arbitration transforms the traditional lengthy and costly manual arbitration into a fast, transparent, and automatically executed on-chain process, improving the efficiency of dispute handling and ensuring the healthy operation of the ecosystem.

[0036] 5. Enhance overall supply chain resilience: Because the decision-making process takes into account the systemic impact at the network level (second type of data), the system can avoid the destruction of global stability by local optima, thus exhibiting stronger robustness in the face of fluctuations and shocks. Attached Figure Description

[0037] Figure 1 The flowchart illustrates the system initialization deployment and task creation process of this invention.

[0038] Figure 2 This is a detailed flowchart of the task execution and intelligent decision-making process of the present invention.

[0039] Figure 3 The flowchart illustrates the dynamic optimization and updating process of the decision rules in this invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] This invention provides an e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain. The system includes a permissioned blockchain network jointly maintained by supply chain participants, an artificial intelligence analysis engine that interacts with the blockchain network, and a smart contract system deployed on the blockchain network. The smart contract system further includes a rule management contract and multiple task execution contracts.

[0042] The following details the complete implementation process of this system, from deployment and operation to optimization.

[0043] 1. System initialization deployment Before the control system is put into operation, the following basic environment construction and core component deployment must be completed.

[0044] S101. Construct a permissioned blockchain network. Multiple participating entities in the supply chain each configure and run a blockchain node, collectively forming a consortium blockchain network. This network employs a consensus algorithm based on a voting mechanism to ensure that all nodes reach agreement on transaction order and state changes. Each node has a smart contract execution environment and a distributed ledger installed. After the network starts, an on-chain identity registration contract registers the identity of each node and the entity to which it belongs.

[0045] S102. Deploy the AI ​​analytics engine. Deploy the AI ​​analytics engine on the server. This engine includes a data interface module, a model management module, and a computation engine module. The data interface module is responsible for communicating with nodes in the blockchain network to obtain on-chain historical data and accessing authorized off-chain real-time data sources through application programming interfaces or oracle services. The model management module is responsible for storing and managing a series of pre-trained machine learning or operations research models. The computation engine module is responsible for loading the corresponding model and processing pipeline according to requests, and performing real-time computations.

[0046] S103. Deploy the smart contract system. By sending a deployment transaction to the blockchain network, the core contracts constituting the smart contract system are created.

[0047] First, the rule management contract is deployed. Upon creation, this contract initializes a mapping data structure in its storage space to maintain the decision rule base. The key of this mapping is a string-type decision scenario identifier, and the value is a structure containing the address of a smart contract that processes the decision logic for that scenario, and a specification string describing the required input data format for that scenario's decision logic.

[0048] Secondly, a template contract for task execution is deployed. This template contract defines the basic data structure for representing logistics tasks on the blockchain and a set of standard interface functions.

[0049] Finally, deploy a reputation contract to record the reputation of node behavior, and an arbitration contract to handle disputes.

[0050] 2. Logistics Task Creation and On-Chain Instantiation When an e-commerce transaction generates a logistics order that needs to be fulfilled, the system executes the following process to create an on-chain instance for it.

[0051] S201. Create a task execution contract. After confirming an order, the e-commerce platform's backend system generates the corresponding logistics task data. Subsequently, the blockchain node operated by the platform calls the creation function of the task execution contract factory contract. This function uses the template contract as a basis and passes in the specific parameters of the current logistics task to generate a brand-new, independent task execution contract instance on the blockchain network. The contract address of this instance is associated with the task identifier.

[0052] 3. Task execution and intelligent decision-making process A successfully deployed task execution contract acts as an agent for the logistics task on the blockchain, driving it to execute step by step according to predefined logic.

[0053] S301. Physical State Synchronization and Decision Triggering. During the physical execution of a logistics task, the completion of each key step generates an electronic certificate with the operator's digital signature, which is sent to the blockchain network as a state event transaction. The pre-built event listening logic within the task execution contract can capture state events related to itself. The listening function verifies the validity of the event signature and the order of business logic. After successful verification, the contract updates its internal state. After a state transition, the contract code automatically determines whether a preset decision point has been reached.

[0054] S302. Query decision rules and construct a data request. Once a decision point is triggered, the task execution contract initiates a query to the rule management contract to retrieve the rule details corresponding to the current decision scenario. The rule management contract searches its decision rule base mapping table and returns the corresponding input data format specification and decision logic contract address. The task execution contract assembles a structured data request object according to the input data format specification.

[0055] S303. Obtain the data to be processed generated by the AI ​​analysis engine. The task execution contract securely sends the data request constructed in step S302 to the AI ​​analysis engine through the oracle mechanism of the blockchain network. After receiving the request, the AI ​​analysis engine performs authentication and parsing.

[0056] Furthermore, as a preferred embodiment, the data to be processed generated by the artificial intelligence analysis engine includes a first type of data and a second type of data.

[0057] The AI ​​analytics engine schedules the appropriate data processing pipeline based on the decision-making scenario identifier in the request: For the first type of data, the pipeline queries verifiable historical states from the blockchain network and / or obtains and verifies real-time information from authorized off-chain data sources. All of this data is based on objective facts that have occurred or can be verified in real time.

[0058] For the second type of data, the pipeline calls pre-built analytical models. These models take the first type of data and the supply chain network topology as input and perform extrapolation calculations. For example, a network simulation model will simulate and quantitatively assess the systematic impact of different decision options on the overall network load balance and the potential delay risks to other tasks over a future period.

[0059] The engine assembles and serializes all acquired and computed data strictly according to the format defined in the request, digitally signs the data using its private key, generates the final data packet to be processed, and returns it to the task execution contract. After verifying the validity of the signature, the task execution contract obtains the input data that can be used for decision-making.

[0060] S304. Execute the decision logic generation instruction. The task execution contract takes the data to be processed obtained in step S303 as a parameter and passes it to the decision logic contract pointed to by the decision logic contract address obtained from the rule management contract through cross-contract calls.

[0061] Furthermore, as a preferred embodiment, the logical processing procedures defined in the decision logic contract include direct calculation logic and consensus voting logic.

[0062] Scenario 1: Execute direct computation logic.

[0063] If the logic processing bound to the current decision-making scenario is direct computation logic, then the entry function of the decision logic contract will execute a deterministic algorithm. For example, this algorithm calculates a utility value for each candidate object i. The calculation formula is: .

[0064] in: , , From the first type of data, respectively representing candidate objects Cost, distance, and inventory levels.

[0065] Data from the second category indicates the selection of candidate objects. Predicted impact on the network.

[0066] to This is a normalization function used to convert data with different dimensions into dimensionless scores.

[0067] to The preset weighting coefficients satisfy... .

[0068] The algorithm iterates through all candidate objects and selects the utility value. The highest-ranking object identifier is used as the output. The entire process is completed independently within the blockchain virtual machine.

[0069] Scenario 2: Execute consensus voting logic.

[0070] If the current decision-making scenario is bound to a consensus voting logic, then the decision-making logic contract will initiate a multi-party on-chain voting process: 1. Proposal Generation: Based on the data to be processed, the contract generates a decision proposal containing a finite number of disposal options.

[0071] 2. Consensus Group Formation and Weight Calculation: The contract determines the set of nodes participating in this vote according to the rules.

[0072] Furthermore, as a preferred implementation, the voting weight value of each node in the consensus group is determined according to a preset dynamic calculation rule. In specific implementation, the contract queries the reputation contract to obtain the voting weight value of each node. Basic credit score At the same time, determine the node. Whether the individual is a direct stakeholder in this event. If so, their effective weight in this round of voting. for: ,in For a penalty coefficient less than 1 (e.g.) If it is a neutral node, then its effective weight... .

[0073] 3. Weighted Voting and Decision Formation: The decision logic contract publishes the proposal and the calculated weight list on-chain and notifies the consensus group members to vote. Each member, within a specified time, signs their chosen option using their private key and submits it to the chain. After collecting all votes, the contract calculates the weights for each option. Calculate the total weighted votes it received. : ,in Indicates voting for option The set of nodes. Ultimately, it will... The option with the highest value is determined as the consensus resolution.

[0074] S305, On-chain Instruction and Physical Execution. After receiving the result returned by the decision logic contract, the task execution contract confirms it as the final resource scheduling instruction. The contract then updates its own state and emits an on-chain event containing the instruction. Listening services deployed on various execution nodes in the physical world continuously subscribe to the blockchain event stream. When a relevant event is captured, the listening service parses the instruction content and triggers the corresponding offline operation process.

[0075] 4. Dynamic optimization and updating of decision-making rules The system's decision-making rules have the ability to be iteratively optimized through on-chain governance.

[0076] S401. Optimize monitoring trigger conditions. During system operation, a separate monitoring service continuously analyzes blockchain data. The optimization process is triggered when one of the following conditions is met: Specific decision rules Recently invoked After that, the number of subsequent on-chain arbitration appeals resulting from the instructions generated was: When the dispute rate Exceeding the preset threshold Time (e.g.) ).

[0077] By listening to node or resource status change events, we can detect significant, pre-defined changes in the supply chain network topology or resource distribution.

[0078] S402. Initiate an optimization proposal. When the monitoring service detects that any of the above conditions are met, it can automatically, or have any authorized node manually, submit a rule optimization proposal to the rule management contract.

[0079] S403, Off-chain Simulation Verification. Before the proposal officially enters the on-chain voting phase, an off-chain simulation verification stage is initiated: 1. Instruct the AI ​​analytics engine to load a complete network state snapshot at a historical block height.

[0080] 2. The AI ​​analytics engine deploys the currently effective old rule logic within its sandbox environment. and the logic of the new rules in the proposal .

[0081] 3. Starting from this historical snapshot, the engine reconstructs all relevant transaction flows over a subsequent period that actually occurred. For each triggered decision point, the sandbox environment is used in parallel. and Two sets of rules are used to simulate two different branches of historical development.

[0082] 4. After the simulation, the engine calculates the performance of the old and new branches on a series of predefined key performance indicators. ( Differences in ) This generates a simulation verification report containing detailed data comparisons and statistical significance analysis. The cryptographic hash of this report is uploaded and stored on the blockchain.

[0083] S404, On-chain Governance Voting. A proposal and its associated mock verification report hash enter a voting period. Nodes holding governance tokens or with voting rights can view the full mock report and decide whether to support or oppose the proposal based on its content. Voting takes place within the rules management contract, which sets a voting deadline and a passing threshold.

[0084] Furthermore, as a preferred embodiment, at least some of the decision rules in the decision rule base are configured to include incentive parameters, which are used to adjust the evaluation of node behavior during logical processing; the rule management contract's optimization and update of the decision rules includes adjusting the incentive parameters.

[0085] For example, the decision-making logic in service provider rating rules relies on a formula: The coefficients among them , , These are the excitation parameters. In the simulation verification of step S403, the artificial intelligence analysis engine can systematically test multiple sets of parameter combinations. This simulation examines the long-term behavioral evolution of the service provider group and the ultimate system-level effects under various parameter settings. It aims to find a global objective function that achieves this. Maximize the combination of parameters This will be used as the content of the optimization proposal.

[0086] S405, Rule Takes Effect. If the optimization proposal receives the required number of votes during the voting period, the rule management contract automatically performs a status update operation. In its maintained decision rule base mapping table, it updates the corresponding decision logic contract address to the newly deployed logic contract address, or updates the parameter set in the rule structure. Thereafter, all newly triggered logistics tasks involving this decision scenario will use the optimized new rule for decision-making.

[0087] 5. Dispute Arbitration Process To address potential disputes during the execution process, the system provides an on-chain arbitration solution.

[0088] S501. Appeal Submission. If a participant has any objection to the final state of a task execution contract or the execution result of an instruction, they may submit an appeal to the arbitration contract and pledge a digital asset as a guarantee as stipulated in the contract.

[0089] S502, Automated Arbitration Enforcement. Once the arbitration contract is invoked, the following process is automatically executed: 1. Evidence Fixation: Based on the disputed task identifier, the contract automatically retrieves and reassembles all relevant data of the task from its creation to its current state from the blockchain network, forming a complete chain of evidence.

[0090] 2. Arbitration Tribunal Formation: The contract stipulates that arbitrators are randomly selected from a pre-registered pool using a verifiable random function. Name (e.g.) Arbitrators who have no direct interest in the parties to the complaint.

[0091] 3. Weighted Ruling: Selected arbitrators independently review the chain of evidence within a specified time and submit their rulings on the blockchain. The arbitration contract will query each arbitrator. Historical fairness score This will be used as the weighting for their opinion in this instance. Regarding the options... Its total weight for: ,in Indicates support option The arbitrators' assembly.

[0092] 4. Result Execution and Settlement: The arbitration contract summarizes the weighted opinions of all arbitrators and calculates the total weight of each award option. ,Will The option with the highest value will be determined as the final arbitration result. Once the result is determined, the contract will immediately and automatically invoke relevant tasks to execute the contract, perform status corrections or fund transfers, and process related margin deposits. Finally, based on the performance of each arbitrator in this arbitration, their historical impartiality scores will be updated. .

[0093] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0094] Finally, the following points should be noted: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installation", "connection", and "linkage" should be interpreted broadly, and can be mechanical or electrical connections, or internal connections between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may change. Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain, characterized in that: include: A permissioned blockchain network consisting of multiple supply chain participant nodes is used for distributed storage and verification of logistics status data. An artificial intelligence analysis engine that interacts with data from the blockchain network; and a smart contract system deployed on the blockchain network; wherein: The smart contract system includes a rule management contract and multiple task execution contracts; The rule management contract is used to maintain a decision rule base. Each decision rule in the decision rule base is associated with a decision scenario and defines the input data structure required for the decision scenario and the logical processing procedure for generating decision results based on the input data. The task execution contract is associated with a specific logistics task and is configured to: during the execution of the logistics task, when the decision scenario is triggered, obtain the data to be processed that meets the requirements of the input data structure from the artificial intelligence analysis engine, and then run the logical processing procedure defined by the associated decision rule to generate a resource scheduling instruction based on the data to be processed; The rule management contract is also configured to optimize and update the logical processing of decision rules in the decision rule base through an on-chain governance mechanism, based on historical decision cases and their execution effect data recorded in the blockchain network.

2. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 1, characterized in that, The logical processing includes direct calculation logic and consensus voting logic; The direct calculation logic is configured to: process the data to be processed according to the calculation method embedded in the rules, and directly output a definite resource scheduling instruction; The consensus voting logic is configured as follows: based on the data to be processed, a decision proposal containing at least two options is generated, and the decision proposal is submitted to a consensus group composed of relevant nodes for weighted voting; wherein, the voting weight value of each node in the consensus group is determined according to a preset dynamic calculation rule; the task execution contract generates resource scheduling instructions based on the weighted voting results.

3. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 2, characterized in that, The dynamic calculation rules are configured based on at least one of the following voting weight values ​​of the computing node: the number or proportion of data items provided by the node in history that have been verified as authentic by the blockchain network and adopted by the final decision; and whether the node has a direct association with the resources or responsible parties involved in the current decision proposal.

4. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 1, characterized in that, The data to be processed generated by the artificial intelligence analysis engine includes a first type of data and a second type of data. The first type of data is based on statistics of verified historical data in the blockchain network, or obtained by collecting and verifying authorized off-chain real-time data sources; The second type of data is generated by the artificial intelligence analysis engine by analyzing the first type of data and the supply chain network topology, and is used to predict the possible impact of different resource scheduling instructions on the subsequent overall state of the supply chain network.

5. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 4, characterized in that, For decision-making scenarios aimed at optimizing the overall performance of the supply chain network, the logical processing defined by the associated decision rules is configured to assign higher priority to indicators reflecting systemic impact in the second type of data when evaluating candidate instructions.

6. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 1, characterized in that, The process by which the rule management contract optimizes and updates decision rules through an on-chain governance mechanism includes a simulation evaluation step: The artificial intelligence analysis engine is invoked to simulate the execution of candidate new rules and original rules to be updated, based on historical data snapshots. A comparative analysis was conducted to examine the differences in key performance indicators generated by the candidate new rules and the original rules to be updated during the simulation. A simulation analysis report containing quantitative comparison results is submitted to the blockchain network, where relevant nodes vote on the report to decide whether to adopt the candidate new rule.

7. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 1 or 6, characterized in that, The conditions that trigger the rule management contract to initiate the optimization and update process for a specific decision rule are at least one of the following: the occurrence rate of subsequent on-chain dispute events associated with the decision rule is detected to exceed a preset threshold; or, a preset type of change is perceived in the supply chain network structure.

8. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 1, characterized in that, The smart contract system also includes arbitration contracts; The arbitration contract is configured to be initiated in response to an objection to the execution result of a resource scheduling instruction, and to perform the following operations: extract all data records related to the objection from the blockchain network, organize an arbitration panel composed of nodes with no direct interest in the objection to make a ruling; wherein the ruling weight of each member of the arbitration panel is determined based on their historical record of impartiality in arbitration; the ruling result and basis are recorded on the blockchain, and trigger an update to the status of the relevant task execution contract.

9. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 1, characterized in that, At least some of the decision rules in the decision rule base are configured to include incentive parameters, which are used to adjust the evaluation of node behavior during logical processing; the rule management contract's optimization and updates of the decision rules include adjustments to the incentive parameters.

10. The e-commerce big data logistics supply chain control system based on artificial intelligence and blockchain according to claim 9, characterized in that, The adjustment of the incentive parameters is based on simulating the expected behavior of nodes in historical scenarios and the resulting global effects under different parameter values ​​through the artificial intelligence analysis engine, and selecting the parameter value combination that makes the preset global optimization target better.

Images