Whole-process collaborative production logistics management optimization system and method

Abstract

The invention discloses a whole-process collaborative production logistics management optimization system and method. The core of the system is intelligent inquiry and blockchain cost management. The inquiry module fuses a dynamic weight algorithm and a machine learning model, intelligently screens goods opportunities, pushes an AES encrypted inquiry sheet, and optimizes a quotation scheme in a multi-objective manner. And the delivery module is in butt joint with the ERP through the OData API, automatically analyzes the order, generates a delivery order, integrates weighing data for fee calculation, and carries out early warning on an abnormal order. And the customs declaration module identifies invoice information through OCR, automatically extracts customs declaration fields, verifies compliance through API, and generates a standardized customs declaration. The in-transit tracking module integrates a multi-platform API, pulls a logistics state in real time, pushes the logistics state to a client through WebSocket, and transparentizes an order state. The cost management module constructs a distributed account book based on a PBFT consensus mechanism block chain, ensures that cost information is safe and credible, and supports multi-currency settlement and dispute arbitration. The system comprehensively optimizes the production logistics link, realizes full-process collaboration, improves the efficiency, reduces the cost, and enhances the enterprise competitiveness.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and specifically to an optimized system and method for full-process collaborative logistics management for manufacturing enterprises. Background Art

[0002] With the rapid development of globalization and e-commerce, enterprises are facing increasingly complex logistics management challenges. Traditional logistics management often has problems such as information silos, process fragmentation, and low collaboration efficiency. Specifically, it is manifested as follows: the manual participation rate in the inquiry process is high, which easily leads to selection deviations; the integration degree of shipping management and the ERP system is insufficient, resulting in information asynchrony; the document processing in the customs declaration link is cumbersome and error-prone; logistics tracking is scattered on multiple platforms, and complete information cannot be obtained in real time; the transparency of logistics cost management is insufficient, and the reconciliation is complex.

[0003] In the prior art, although there are various logistics management systems, most of them optimize a single link in the logistics chain and lack a systematic solution for full-process collaboration. Especially in aspects such as cross-system data integration, multi-party collaborative decision-making, and the level of automation and intelligence, there is still much room for improvement.

[0004] Therefore, there is an urgent need for an intelligent logistics management system that can achieve full-process collaboration in inquiry, shipping, customs declaration, logistics tracking, and cost management, improve logistics management efficiency, and reduce the logistics costs of enterprises. Summary of the Invention

[0005] The object of the present invention is to provide an optimized system and method for full-process collaborative production logistics management, solve the problems of information silos, process fragmentation, and low collaboration efficiency in existing logistics management, and achieve full-process intelligent management from inquiry to cost settlement.

[0006] To achieve the above object, the present invention provides an optimized system for full-process collaborative production logistics management, which is characterized by including:

[0007] An inquiry management module, which is used to receive inquiry requirements, screen candidate freight forwarders by using a dynamic weight algorithm in combination with a machine learning model, push an encrypted inquiry form generated by using the AES encryption algorithm, receive quotation information, and generate an optimal quotation plan based on a multi-objective optimization algorithm;

[0008] A shipping management module, which is used to automatically parse order data in the ERP system through the API interface of the OData protocol, generate a shipping order, integrate weighing sensor data, automatically calculate logistics costs, and trigger an early warning mechanism for abnormal orders;

[0009] An agency customs declaration module, which is used to identify cargo invoice information by using OCR technology, automatically extract customs declaration fields, perform compliance verification through the customs system API, and generate a standardized customs declaration form;

[0010] The in-transit order tracking module is used to integrate the API interfaces of multiple logistics platforms, pull the logistics status data in real time, and push it to the client in real time through WebSocket technology to update the order status; and

[0011] The logistics cost management module is used to build a distributed ledger based on blockchain technology, record the details of logistics costs, support automatic conversion and settlement of multiple currencies, and provide a function for cost dispute arbitration. The distributed ledger adopts the PBFT consensus mechanism.

[0012] Optionally, the inquiry management module includes:

[0013] The inquiry demand receiving unit is used to receive the inquiry demands input by the salesperson;

[0014] The freight forwarder screening unit is used to screen candidate freight forwarders from the freight forwarder database based on the dynamic weight algorithm of the machine learning model. The weight factors of the dynamic weight algorithm include the historical service quality, price competitiveness, service matching degree, transport capacity guarantee and compliance performance of the freight forwarder;

[0015] The inquiry form pushing unit is used to generate an encrypted inquiry form using the AES encryption algorithm and push it to the candidate freight forwarders;

[0016] The quotation receiving unit is used to receive the quotation information fed back by the freight forwarder system; and

[0017] The price comparison and analysis unit is used to automatically conduct price comparison and analysis. The price comparison and analysis is based on a multi-objective optimization algorithm, comprehensively considering the total cost, transportation timeliness, service quality and risk factors, and generating the optimal quotation plan.

[0018] Optionally, the shipping management module includes:

[0019] The order parsing unit is used to automatically parse the order data in the ERP system through the API interface of the OData protocol;

[0020] The shipping order generating unit is used to generate a shipping order according to the parsed order data;

[0021] The weighing integration unit is used to integrate the data of weighing sensors, automatically calculate the logistics cost, and the weighing integration unit supports integration with platform scales, truck scales and DWS systems; and

[0022] The exception warning unit is used to trigger an early warning mechanism for abnormal orders and push early warning information through the mobile application. The abnormal situations detected by the early warning mechanism include weight abnormality, address abnormality, logistics delay, special goods and compliance risks.

[0023] Optionally, the customs broker module includes:

[0024] An invoice recognition unit for identifying goods invoice information using OCR technology, automatically extracting customs declaration fields. The OCR technology is based on a deep learning model and is trained and optimized for commercial invoice formats;

[0025] A compliance verification unit for performing compliance verification through the customs system API; and

[0026] A customs declaration form generation unit for generating standardized customs declaration forms.

[0027] Optionally, the in-transit order tracking module includes:

[0028] A logistics data integration unit for integrating multi-logistics platform API interfaces, pulling logistics status data in real time. The logistics data integration unit integrates data in a hybrid mode of active pulling + passive reception;

[0029] A real-time push unit for pushing logistics status data to the client in real time through WebSocket technology; and

[0030] An order status update unit for updating the order status.

[0031] Optionally, the logistics cost management module includes:

[0032] A blockchain ledger unit for building a distributed ledger based on blockchain technology to record logistics cost details. The blockchain ledger adopts a consortium chain architecture;

[0033] A multi-currency settlement unit for supporting automatic conversion and settlement of multiple currencies; and

[0034] A cost arbitration unit for providing a cost dispute arbitration function, using blockchain deposit technology to verify the authenticity and accuracy of cost details.

[0035] The present invention also provides a full-process collaborative production logistics management optimization method, including the following steps:

[0036] Step S1: Receive inquiry requirements, screen candidate freight forwarders using a dynamic weight algorithm combined with a machine learning model, push an encrypted inquiry form generated using the AES encryption algorithm, receive quotation information, and generate an optimal quotation plan based on a multi-objective optimization algorithm;

[0037] Step S2: Automatically parse order data in the ERP system through the API interface of the OData protocol to generate a shipping order, integrate weighing sensor data, automatically calculate logistics costs, and trigger an early warning mechanism for abnormal orders;

[0038] Step S3: Identify goods invoice information using OCR technology, automatically extract customs declaration fields, perform compliance verification through the customs system API, and generate a standardized customs declaration form;

[0039] Step S4: Integrate the API interfaces of multiple logistics platforms, pull the logistics status data in real time, and push it to the client in real time through WebSocket technology to update the order status; and

[0040] Step S5: Build a distributed ledger based on blockchain technology, record the details of logistics fees, support automatic conversion and settlement of multiple currencies, and provide a function for fee dispute arbitration, where the distributed ledger adopts the PBFT consensus mechanism.

[0041] Optionally, Step S1 includes:

[0042] Step S11: Receive the inquiry requirements input by the salesman;

[0043] Step S12: Screen candidate freight forwarders from the freight forwarder database based on the dynamic weight algorithm of the machine learning model. The weight factors of the dynamic weight algorithm include the historical service quality, price competitiveness, service matching degree, transport capacity guarantee, and compliance performance of the freight forwarder;

[0044] Step S13: Generate an encrypted inquiry form using the AES encryption algorithm and push it to the candidate freight forwarders;

[0045] Step S14: Receive the quotation information fed back by the freight forwarder system; and

[0046] Step S15: Automatically conduct price comparison and analysis. The price comparison and analysis are based on the multi-objective optimization algorithm, comprehensively considering the total cost, transportation timeliness, service quality, and risk factors, and generate the optimal quotation plan.

[0047] Optionally, Step S2 includes:

[0048] Step S21: Automatically parse the order data in the ERP system through the API interface of the OData protocol;

[0049] Step S22: Generate a shipping order according to the parsed order data;

[0050] Step S23: Integrate the data of the weighing sensor, automatically calculate the logistics fee, and the integration of the weighing sensor data supports integration with platform scales, truck scales, and DWS systems; and

[0051] Step S24: Trigger an early warning mechanism for abnormal orders, and push early warning information through the mobile application. The abnormal situations detected by the early warning mechanism include weight abnormalities, address abnormalities, logistics delays, special goods, and compliance risks.

[0052] Optionally, Step S3 includes:

[0053] Step S31: Use OCR technology to identify the information of the goods invoice, automatically extract the customs declaration fields. The OCR technology is based on a deep learning model and is trained and optimized for the commercial invoice format.

[0054] Step S32: Conduct compliance verification through the customs system API; and

[0055] Step S33: Generate a standardized customs declaration form.

[0056] The beneficial effects of the present invention are as follows:

[0057] Achieve the digitization and automation of the entire logistics management process, significantly improve work efficiency, reduce manual intervention and error rates; break information silos, achieve seamless integration between systems, and ensure data consistency and real-time performance; leverage cutting-edge technologies such as AI, blockchain, and WebSocket to enhance the system's intelligence level; build a multi-party collaboration mechanism to achieve information sharing and process collaboration among suppliers, producers, and logistics providers; provide multi-dimensional data analysis and visual display to assist enterprise decision-making and optimization.

[0058] Other features and advantages of the present invention will become clear from the following detailed description of the exemplary embodiments of the present invention with reference to the accompanying drawings. Description of the Drawings

[0059] The drawings incorporated in the specification and forming a part of the specification illustrate embodiments of the present invention and, together with the description, are used to explain the principles of the present invention.

[0060] Figure 1 is a structural block diagram of the full-process collaborative production logistics management optimization system provided by the embodiment of the present invention. Detailed Embodiments

[0061] Now, various exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that: Unless otherwise specifically stated, the relative arrangements, numerical expressions, and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0062] The following description of at least one exemplary embodiment is merely illustrative and in no way limits the present invention or its application or use.

[0063] Techniques, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods, and devices should be regarded as part of the specification.

[0064] In all the examples shown and discussed here, any specific value should be construed as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0065] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0066] System overall architecture:

[0067] As Figure 1 shown, the full-process collaborative production logistics management optimization system provided by the present invention includes five core modules: an inquiry management module 10, a shipment management module 20, a customs broker agency module 30, an in-transit order tracking module 40, and a logistics cost management module 50. Each module conducts data exchange and process collaboration through a data bus to form a complete logistics management closed loop. The system adopts a cloud-native microservices architecture, supporting horizontal expansion and flexible deployment.

[0068] The system front-end adopts a responsive design, supporting multi-terminal access on the PC side and mobile side. The back-end adopts a distributed architecture, combined with containerization technology and microservices design, to ensure the high availability and scalability of the system. Data storage adopts a hybrid storage strategy, with a relational database storing business data, a NoSQL database storing unstructured data, and a distributed file system storing document and image data.

[0069] Detailed implementation of the inquiry management module:

[0070] The inquiry management module 10 includes an inquiry requirement receiving unit, a freight forwarder screening unit, an inquiry form pushing unit, a quotation receiving unit, and a price comparison and analysis unit.

[0071] The inquiry requirement receiving unit is responsible for receiving inquiry requirements submitted by salesmen through the Web interface or mobile application. The inquiry requirements include, but are not limited to: information such as goods type, weight, volume, origin, destination, estimated shipment date, special requirements (such as temperature control, dangerous goods handling), etc. The system provides a standardized inquiry form, supporting batch import and automatic verification of information to ensure the integrity and accuracy of the input information.

[0072] The freight forwarder screening unit is based on a machine learning model of the gradient boosting decision tree (GBDT) model and combines an adaptive dynamic weight algorithm to screen candidate freight forwarders from the freight forwarder database. In specific implementation, the system maintains a freight forwarder database containing more than 50 fields such as freight forwarder basic information, historical performance, professional fields, service evaluations, etc.

[0073] The core of the adaptive dynamic weight algorithm lies in dynamically adjusting the proportion of various weight factors according to different inquiry requirements and market environments to more accurately screen out the optimal freight forwarder. Its weight factors mainly include:

[0074] Freight forwarder historical service quality (basic weight, dynamically adjusted according to cargo type): Comprehensively evaluate the service performance of freight forwarders in the past 12 months. The initial weight ratio is 30%, and it is adaptively adjusted within the range of 20%-40% according to the cargo type (e.g., fragile goods, general goods). Specific indicators include:

[0075] On-time rate: The deviation rate between the actual arrival time and the estimated arrival time.

[0076] Cargo safety rate: The proportion of damaged or lost goods during transportation.

[0077] Response speed: The average response time to requests such as inquiries and exception handling.

[0078] User evaluation: User ratings and feedback based on historical cooperation orders.

[0079] Price competitiveness (basic weight, dynamically adjusted according to route and timeliness): Evaluate based on the historical quotation data of freight forwarders and the market average price level. The initial weight ratio is 25%, and it is adaptively adjusted within the range of 15%-35% according to the route (e.g., popular route, unpopular route) and timeliness requirements (e.g., urgent, general cargo).

[0080] Service matching degree (fixed weight 20%): Evaluate the matching degree between the professional fields and service scopes of freight forwarders and the current inquiry requirements. For example, for cold chain transportation requirements, give priority to freight forwarders with cold chain transportation qualifications and experience.

[0081] Transport capacity guarantee (basic weight, dynamically adjusted according to season and holidays): Evaluate the transport capacity resources (e.g., own vehicles, cooperative shipping schedules) and stability of freight forwarders. The initial weight ratio is 15%, and it is adaptively adjusted within the range of 10%-25% according to the season (e.g., peak season, off-season) and holidays (e.g., Spring Festival).

[0082] Compliance performance (fixed weight 10%): Evaluate the historical records of freight forwarders in terms of compliance, such as whether there have been customs penalties, illegal operations, etc.

[0083] Gradient Boosting Decision Tree (GBDT) model is used to further optimize the dynamic adjustment of weight factors. The model is trained using inquiry - quotation - transaction data from the past two years to learn the relationship between different combinations of weight factors and the final inquiry results (e.g., transaction price, service evaluation). The model outputs a dynamic adjustment coefficient, which is used to fine-tune the proportion of each weight factor in the adaptive dynamic weight algorithm, so as to achieve more accurate freight forwarder screening.

[0084] The algorithm first calculates the preliminary score through the adaptive dynamic weight algorithm, then fine-tunes the score using the dynamic adjustment coefficient output by the GBDT model, and finally comprehensively scores and ranks the candidate freight forwarders, selecting the top 3-8 (the quantity can be configured) with the highest scores as the final candidate freight forwarders. The system retrains the GBDT model monthly based on the newly added historical data and updates the parameters of the adaptive dynamic weight algorithm weekly to ensure the accuracy and timeliness of the screening results. The inquiry form push unit uses the AES-256 encryption algorithm to generate encrypted inquiry forms and push them to the candidate freight forwarders. The inquiry form contains information such as cargo information, logistics requirements, and inquiry deadline, but does not contain customer sensitive information. The push methods support three forms: email, API interface docking, and internal system messages. Freight forwarders can decrypt the content of the inquiry form with a dedicated key. The system automatically records the push time, receipt confirmation, and viewing status of the inquiry form.

[0085] The quotation receiving unit receives the quotation information fed back by the freight forwarder system. The quotation information includes basic freight, surcharges (such as fuel surcharge, security fee), transit fees, estimated transit time, service commitments, etc. The system supports multiple quotation receiving methods, including API automatic docking, email parsing, and manual entry. For non-standard format quotation forms, the system uses natural language processing technology to automatically extract key information and store it in a structured manner.

[0086] The price comparison and analysis unit automatically conducts price comparison and analysis to generate the optimal quotation plan. The price comparison and analysis is based on the weighted multi-objective optimization algorithm, comprehensively considering the following key objectives:

[0087] Total cost (core objective): including basic freight and various surcharges (fuel surcharge, security fee, port miscellaneous charges, etc.), with the highest cost weight, an initial value of 50%, and can be adjusted within the range of 40%-60% according to user preferences.

[0088] Transportation timeliness (important objective): including estimated transportation time and on-time rate, with the initial timeliness weight of 30%, and can be adjusted within the range of 20%-40% according to the urgency of the goods.

[0089] Estimated transportation time: the transportation duration promised by the freight forwarder from the place of departure to the destination.

[0090] On-time rate (historical data): evaluated based on the on-time delivery rate of the freight forwarder's historical orders.

[0091] Service quality (balanced objective): based on the service evaluation of historical data, with the initial service quality weight of 15%, and can be adjusted within the range of 10%-20% according to the user's service requirements.

[0092] Service evaluation indicators: including user ratings, customer service response speed, problem-solving efficiency, etc.

[0093] Risk factors (auxiliary objectives): Evaluate potential logistics risks with an initial risk weight of 5%, which can be adjusted within the range of 0% - 10% according to the route and freight forwarder situation.

[0094] Risk factor assessment: Include route stability (e.g., whether passing through war-torn areas or areas with frequent natural disasters), the financial status of the freight forwarder (e.g., whether there is a risk of bankruptcy), potential compliance risks, etc.

[0095] The weighted multi-objective optimization algorithm first standardizes the scores of each freight forwarder's quotation plan in the above four objective dimensions (e.g., the plan with the lowest cost gets the highest score in the cost dimension, and the plan with the fastest delivery time gets the highest score in the delivery time dimension), and then performs weighted summation according to the preset weight coefficients to obtain the comprehensive score of each plan. The system selects the plan with the highest comprehensive score as the optimal quotation plan.

[0096] The system generates a detailed price comparison analysis report, including:

[0097] Detailed comparison of each freight forwarder's quotation: Clearly show the specific quotations and scores of each freight forwarder in terms of cost, delivery time, service quality, and risk factors.

[0098] Comprehensive score ranking: Rank each plan according to the comprehensive score calculated by the weighted multi-objective optimization algorithm.

[0099] Optimal plan recommendation: The system recommends the plan with the highest comprehensive score as the optimal choice and provides a clear reason explanation.

[0100] Explanation of the decision-making basis: Explain in detail the advantages and disadvantages of the optimal plan, as well as the weight ratios of each objective dimension to help business personnel understand and verify the decision-making process.

[0101] Visual display: The report uses visual methods such as charts and bar graphs to intuitively show the advantages and disadvantages of each plan, facilitating business personnel to quickly understand and make decisions.

[0102] Users can customize and adjust the weight coefficients of each objective according to actual business needs to meet different inquiry preferences (e.g., being more concerned about price, being more concerned about delivery time, etc.). The system also supports the scenario analysis function, and users can simulate the changes in the optimal plan under different weight combinations to assist in more comprehensive decision-making analysis.

[0103] Detailed implementation of the shipping management module:

[0104] The shipping management module 20 includes an order parsing unit, a shipping note generation unit, a weighing integration unit, and an exception warning unit.

[0105] The order parsing unit automatically parses the order data in the ERP system through the API interface of the OData protocol. The system supports integration with mainstream ERP systems (such as SAP, Oracle, Microsoft Dynamics) to achieve real-time data synchronization. The interface is designed in the RESTful style and supports the incremental data pulling and change data capture (CDC) mechanism to ensure the real-time and accuracy of data synchronization.

[0106] In the specific implementation, the system polls the ERP interface every 5 minutes to obtain the newly added and changed order data. The received data is parsed in JSON format and cleaned, and then stored in the system database. In case of exceptions (such as interface timeout, data format error) during the data synchronization process, the system will automatically retry and record the log, and send an alarm notification to the administrator after exceeding the preset retry times.

[0107] The shipping note generation unit generates shipping notes based on the parsed order data. The generation of shipping notes follows the following process:

[0108] Data integrity check: Verify whether the order information is complete, including the information of the shipper and consignee, goods information, service requirements, etc.;

[0109] Business rule verification: Check whether the order meets the shipping conditions according to the preset business rules;

[0110] Logistics method matching: Based on the order requirements and the characteristics of the goods, match the most suitable logistics method;

[0111] Shipping note number generation: Generate a unique shipping note number according to the predetermined rules;

[0112] Shipping note document generation: Generate a standardized shipping note document, including PDF format and structured data.

[0113] The generated shipping note supports electronic signature and anti-tampering verification to ensure the legal effect and security of the document. The system also provides the function of customizing the shipping note template, and enterprises can adjust the format and fields of the shipping note according to their own needs.

[0114] The weighing integration unit integrates the weighing sensor data with high precision and multiple modes, and automatically and intelligently calculates the logistics cost. The system is widely compatible with various mainstream electronic weighing devices, and conducts real-time and stable data communication with the electronic weighing devices through various communication protocols such as the standard serial port protocol (RS-232 / RS-485), TCP / IP network protocol, and Modbus protocol to obtain accurate goods weight data.

[0115] The supported types of weighing devices include:

[0116] Bench Scale: Suitable for the precise weighing of small and light goods, with an accuracy of up to ±0.1 grams.

[0117] Floor Scale: Suitable for the weighing of medium and heavy goods, with a wide range of measuring capacities and an accuracy of up to ±1 kg.

[0118] DWS (Dimension Weight Scanner) System: An integrated device that combines dimension measurement, weight weighing, and barcode scanning functions. It is suitable for automated logistics scenarios, capable of quickly and accurately obtaining the dimension and weight data of goods, and automatically scanning barcodes to improve logistics efficiency. The system supports integration with the API interface of the DWS system to achieve seamless data docking.

[0119] Onboard Weighing System: A weighing system integrated on handling equipment such as forklifts and AGVs, suitable for mobile weighing scenarios, facilitating weighing during the process of goods handling.

[0120] Intelligent Processing Flow of Weight Data:

[0121] Data Collection and Verification: The system collects the weight data uploaded by the weighing equipment in real time and conducts data verification. For example, data format verification, data range verification, and data stability verification are performed to ensure data quality.

[0122] Weight Comparison and Anomaly Detection: The system automatically conducts intelligent comparison between the actual weighed weight and the declared weight in the shipping note, and dynamically adjusts the weight difference threshold according to factors such as the type of goods, transportation mode, and destination. For example, for valuable items or air freight goods, the threshold is set smaller (e.g., ±1%), and for ordinary goods or sea freight goods, the threshold is set larger (e.g., ±5%). If the weight difference exceeds the preset threshold, the system automatically triggers the anomaly warning mechanism and initiates the manual review process.

[0123] Volume Weight Calculation (Optional): For bulky and light goods, the system supports automatically calculating the volume weight based on the dimensions of the goods, comparing it with the actual weight, and taking the larger value as the billing weight, which conforms to the billing rules of the International Air Transport Association (IATA) and major shipping companies.

[0124] Automatic Calculation of Logistics Fees: For the weight data (or volume weight) that has been confirmed to be correct, the system intelligently and automatically calculates the logistics fees according to the preset billing rules (such as segmented billing, stepped billing, billing by weight / volume / quantity) and the currently effective logistics contracts (automatically obtained from the contract management system), and updates them in real time to the shipping note and the expense management module. The system supports automatic conversion of multiple currencies and exchange rates.

[0125] The intelligence and automation of the weighing integrated unit have significantly improved the efficiency, accuracy and transparency of logistics cost calculation, reduced manual operations and human errors, and provided strong support for enterprises to reduce logistics costs.

[0126] The abnormal warning unit triggers the warning mechanism for abnormal orders and pushes warning information through mobile applications. The system defines a variety of abnormal situation detection rules:

[0127] Abnormal weight: The difference between the actual weight and the declared weight exceeds the threshold;

[0128] Abnormal address: the delivery address is incomplete or contains obvious errors;

[0129] Logistics delay: the delivery processing time exceeds the SLA agreement;

[0130] Special cargo: cargo that requires special handling (such as dangerous goods, fragile goods) is not marked;

[0131] Compliance risk: There are trade restrictions or special permits required at the shipping destination.

[0132] When an abnormality is detected, the system sends an early warning to relevant personnel through push notifications, SMS, emails, etc. according to the type and severity of the abnormality. The early warning message contains the type of abnormality, impact level, specific description and recommended handling method. The system also supports automatic escalation strategy for abnormal events. If it is not handled within a certain period of time, it will be reported to higher management level step by level.

[0133] Detailed implementation of the customs declaration agent module:

[0134] The customs declaration agent module 30 includes an invoice identification unit, a compliance verification unit and a customs declaration form generation unit.

[0135] The invoice recognition unit uses high-precision, adaptive OCR (optical character recognition) technology to recognize cargo invoice information and automatically extract customs declaration fields. The system is deeply optimized to support the import of invoices in multiple complex formats, including scanned PDF, image formats (JPEG, PNG, TIFF) and electronic invoice XML / UBL files, and has automatic layout analysis and field positioning capabilities.

[0136] The OCR engine is built based on the convolutional neural network (CNN) and recurrent neural network (RNN) deep learning models, and uses large-scale commercial invoice data sets for training and continuous optimization. The model has carried out special algorithm optimization based on the characteristics of commercial invoices, such as table structure, multi-language mixture, complex typesetting, etc., which significantly improves recognition accuracy and robustness. The system supports recognition in more than 50 languages, including English, Chinese, Japanese, Korean, French, German, Spanish, etc.

[0137] The OCR processing flow includes the following key steps:

[0138] Intelligent image preprocessing: An adaptive image enhancement algorithm is adopted to perform preprocessing operations such as denoising, skew correction, contrast enhancement, and shadow removal on invoice images of different qualities, improving the image quality and laying a foundation for subsequent recognition.

[0139] Layout analysis and region detection: Using a deep learning-based layout analysis model, automatically identify the layout structure of the invoice (e.g., table region, text region, image region), and accurately locate the regions where the key fields are located.

[0140] High-precision text recognition: A CNN-RNN hybrid model is used for text recognition, making full use of the image feature extraction ability of CNN and the sequence modeling ability of RNN to achieve high-precision text recognition. The model is specifically optimized for common invoice fonts and characters, and context semantic understanding is introduced to further improve the recognition accuracy.

[0141] Structured information extraction: Based on predefined customs declaration field templates and natural language processing (NLP) technology, automatically extract the key field information required for customs declaration from the recognized text information, such as commodity name, HS code, country of origin, quantity, unit price, total value, currency, invoice number, invoice date, invoice title, invoice amount, etc. The system supports customizing field templates to adapt to invoice formats in different countries and regions.

[0142] Intelligent proofreading and manual intervention: The system provides a visual proofreading interface to compare and display the OCR recognition results with the original invoice image, and highlight the fields with low confidence, facilitating quick manual proofreading and correction. The system supports users to customize proofreading rules, such as field value range limits, format verification, etc., to improve data quality.

[0143] The system focuses on extracting more than 30 key fields for customs declaration, and supports batch invoice recognition and processing, greatly improving the customs declaration efficiency. The extraction results will be stored in a structured JSON or XML data format for convenient subsequent data processing and system integration. The system also has a learning ability, and the user's manual proofreading results will be used to continuously optimize the OCR model and continuously improve the recognition accuracy.

[0144] The compliance verification unit conducts compliance verification through the customs system API. The system integrates the customs system interfaces of major trading countries / regions and can query and verify the following information in real time:

[0145] Validity and latest changes of HS codes;

[0146] Commodity import and export restrictions and licensing requirements;

[0147] Tariff rates and preferential policies;

[0148] Rules of origin and certificate requirements;

[0149] Special regulatory requirements (such as for food, drugs, chemicals, etc.).

[0150] Compliance verification uses a multi-level rule engine. First, basic compliance checks are performed (such as HS code format, completeness of required fields), then country / region-specific rule verification is carried out, and finally product-specific rule verification is conducted. For compliance issues found, the system grades them by severity (warning, error, block) and provides specific correction suggestions.

[0151] The customs declaration form generation unit generates a standardized customs declaration form. The system has built-in customs declaration form templates for major countries / regions and automatically selects the appropriate template according to the destination. The process of generating the customs declaration form includes:

[0152] Data assembly: Integrate invoice data, product information, compliance verification results, etc.;

[0153] Format conversion: Generate customs declaration data in the format required by the customs;

[0154] Document generation: Generate a customs declaration form file that can be directly submitted;

[0155] Electronic signature: Digitally sign the customs declaration form to ensure integrity and non-repudiation.

[0156] The generated customs declaration form supports multiple output formats, including the customs standard electronic data format, PDF documents, and structured data. The system also supports version control and change tracking of the customs declaration form, facilitating the viewing of the modification history and auditing.

[0157] Detailed implementation of the in-transit order tracking module:

[0158] The in-transit order tracking module 40 includes a logistics data integration unit, a real-time push unit, and an order status update unit.

[0159] The logistics data integration unit adopts a three-mode fusion data integration strategy of "active push (Push) + intelligent polling (Pull) + auxiliary parsing (Parse)", efficiently and comprehensively integrating multi-logistics platform API interfaces to obtain global logistics status data in real time. The system has pre-integrated the standard RESTful API interfaces of more than 100 mainstream international logistics carriers (such as DHL, FedEx, UPS, Maersk, CMACGM, etc.) and domestic logistics platforms (such as Cainiao, SF Express, Zhongtong, Yuantong, Shentong, Yunda, JD Logistics, Debon Express, KuaiDi100, etc.).

[0160] "Active Push" mode: For more than 60% of the logistics providers that offer WebHook or EventAPI push services, the system actively registers the WebHook callback interface or subscribes to EventAPI events to receive the latest status update information actively pushed by the logistics providers in real time. This mode has the advantages of high real-time performance and timely data update, enabling the acquisition of changes in logistics status at the earliest time.

[0161] "Intelligent Polling (Pull)" mode: For approximately 30% of the logistics providers that only offer query APIs, the system adopts an intelligent polling strategy, dynamically adjusting the API polling frequency based on the cargo status and transportation stage. For example, at the initial stage of cargo shipment and key transportation nodes (such as customs clearance, transshipment, before delivery), the polling frequency is increased (such as polling once every 5 minutes), and during the stable stage of cargo transportation, the polling frequency is decreased (such as polling once every 30 minutes) to balance data real-time performance and system resource consumption. The system also supports event-triggered intelligent polling. For example, when it detects that the cargo status has not been updated for a long time, the polling frequency is automatically increased.

[0162] "Auxiliary Parsing (Parse)" mode: For approximately 10% of small logistics providers or special line logistics that do not offer standard APIs or have unstable API interfaces, the system adopts email parsing and web crawler technologies to obtain logistics information.

[0163] Email parsing: The system intelligently parses the email notifications sent by logistics providers and automatically extracts key data such as logistics order numbers and status update information.

[0164] Web crawler: The system simulates browser behavior, regularly crawls the logistics tracking information on the official websites of logistics providers or third-party query websites, and uses HTML parsing and data extraction technologies to convert web page data into structured data.

[0165] The data integration architecture adopts the adapter pattern (Adapter Pattern), developing independent adapters (DataAdapter) for each logistics provider API or data source. The adapter is responsible for converting the non-standard and heterogeneous logistics data formats of different logistics providers into the unified and standardized internal data format (Canonical Data Model) of the system. This mode has good scalability and maintainability, facilitating the subsequent integration of new logistics providers and interface upgrades.

[0166] The system also has built-in data cleaning and data quality monitoring mechanisms to deduplicate, correct errors, and standardize the integrated data, and real-time monitor data quality, such as data integrity, data accuracy, data timeliness, etc., to ensure the high quality and reliability of logistics data.

[0167] The real-time push unit uses WebSocket technology to push logistics status data to the client in real time. The system establishes a persistent WebSocket connection pool, supporting tens of thousands of users to query the logistics status online simultaneously. When the logistics status is detected to be updated, the server actively pushes the latest information to relevant users without the need for users to manually refresh the page.

[0168] The pushed content includes basic logistics status (such as picked up, in transit, arrived, etc.) and detailed event records (such as specific location, operation time, operator, etc.). The system supports a push priority mechanism to ensure that important status changes (such as exceptions, delays, approaching arrival, etc.) are pushed first. To improve the user experience, the push also supports mixed graphic and text display and multi-language automatic translation functions.

[0169] The order status update unit is responsible for updating the order status. The system defines a standardized logistics status model, including multiple statuses such as order creation, pending shipment, shipped, in transit, arrived, signed for, exception, etc. After the logistics tracking data is parsed and mapped by rules, the status of the corresponding order is automatically updated.

[0170] The status update triggers a series of automated workflows, such as:

[0171] After the shipment status is updated, a shipment notice is automatically generated for the customer;

[0172] When an abnormal status is detected, an exception handling process is triggered;

[0173] After the goods are signed for, the settlement process is automatically started;

[0174] When the status update meets the preset conditions, data synchronization of relevant business systems is triggered.

[0175] The system also realizes the visual display of the logistics track, intuitively presenting the real-time location and historical movement track of the goods through an interactive map. The map supports multi-level zooming and optimized route display. Combined with the timeline function, users can view the goods status at any time point.

[0176] The detailed implementation of the logistics cost management module

[0177] The logistics cost management module 50 includes a blockchain ledger unit, a multi-currency settlement unit, and a cost arbitration unit.

[0178] The blockchain ledger unit builds a highly secure and traceable distributed ledger based on blockchain technology to record the details of logistics costs. The system preferably uses the Hyperledger Fabric framework to build a consortium chain architecture, and the participating parties include relevant parties such as the shipper, logistics provider, consignee, financial institution, and regulatory agency.

[0179] Under the premise of ensuring data security and privacy, the consortium blockchain architecture enables multi-party participation and collaborative management. Hyperledger Fabric provides mature permission management, identity authentication, and transaction encryption mechanisms to ensure that only authorized participants can access and modify ledger data.

[0180] Each logistics transaction is recorded as a block on the chain, and the block data contains the following key information:

[0181] Transaction unique identifier: Ensures the traceability and uniqueness of each transaction.

[0182] Transaction participant information: Clearly records the identity information of participants such as the shipper, logistics provider, and consignee.

[0183] Transaction timestamp: Records the exact time when the transaction occurred, ensuring the immutability of the time sequence.

[0184] Fee details (structured data): Details the composition of basic freight, various surcharges, taxes, discounts, etc., using a standardized data format for easy subsequent analysis and settlement.

[0185] Service records (unstructured data): Records information such as the actual service content, service quality evaluation, and exception handling records, supporting the storage of data in various formats such as text, pictures, and videos.

[0186] Hash value of proof documents: Stores the hash values of key documents such as electronic bills of lading, receipt proofs, invoices, and service vouchers to ensure the integrity and immutability of the document content and support the quick verification of document authenticity.

[0187] Hash value of smart contract code: Records the hash value of the smart contract code that executes the transaction to ensure the traceability and consistency of contract execution.

[0188] The core consensus mechanism of the blockchain adopts the PBFT (Practical Byzantine Fault Tolerance) algorithm, which achieves low latency and high throughput while ensuring high fault tolerance (tolerating (N - 1) / 3 malicious nodes). The PBFT algorithm ensures that the system can still reach a consensus and guarantee data consistency and reliability through a three-phase commit protocol in the event of failures or malicious behaviors of some nodes.

[0189] The smart contract (Chaincode) pre-sets business logics such as fee calculation, automatic reconciliation, multi-party settlement, and dispute arbitration. It is developed in the Go language and has undergone strict security audits. The smart contract automatically executes the pre-set business processes, reducing manual intervention, improving efficiency, and reducing the risks of human errors and fraud.

[0190] All transaction records are organized in a Merkle tree structure to ensure data integrity and anti-tampering. The system provides a complete API interface for easy integration with the enterprise's existing ERP, financial system, etc., to achieve automatic data synchronization and sharing. The blockchain ledger data supports multi-dimensional queries and audits, facilitating compliance reviews by regulatory and auditing agencies.

[0191] The logistics cost management module is built using blockchain technology, effectively solving the following pain points in the traditional model:

[0192] Lack of information transparency: The blockchain ledger is publicly transparent, and all participating parties can view the expense details and transaction records in real time, improving information transparency.

[0193] Low reconciliation efficiency: Smart contracts automatically reconcile accounts, reducing the workload of manual checking and improving the reconciliation efficiency and accuracy.

[0194] High trust risk: The immutability and traceability of blockchain technology enhance the trust between trading parties and reduce the risks of expense disputes and fraud.

[0195] High arbitration cost: Blockchain evidence storage technology provides credible evidence for expense dispute arbitration, reducing the arbitration cost and time.

[0196] The multi-currency settlement unit supports multi-currency automatic conversion and settlement. The system integrates real-time exchange rate data from the international foreign exchange market and supports automatic conversion between more than 50 major currencies. The exchange rate data is updated hourly to ensure conversion accuracy.

[0197] The settlement process supports the following functions:

[0198] Multi-currency bill generation: Automatically generate multi-currency bills according to the settlement currency specified in the contract;

[0199] Exchange rate risk management: Provide early warnings of exchange rate fluctuations and hedging suggestions;

[0200] Automatic reconciliation: Based on the expense data recorded on the blockchain, automatically reconcile accounts with the logistics provider's system;

[0201] Diversified settlement methods: Support multiple settlement methods such as letters of credit, wire transfers, and third-party payments;

[0202] Tax processing: Automatically calculate the payable tax amount and generate tax return data according to the tax policies of different countries / regions.

[0203] The expense arbitration unit provides an expense dispute arbitration function. When an expense dispute occurs, the system provides a structured dispute submission process:

[0204] Dispute initiation: Specify the details, amount, and reasons for the dispute in detail;

[0205] Evidence Submission: Upload evidence materials to support each party's claims;

[0206] Automatic Verification: The system verifies the authenticity and integrity of the fee data based on blockchain evidence storage technology;

[0207] Intelligent Analysis: Analyze historical similar cases and contract terms through machine learning algorithms;

[0208] Mediation Suggestion: The system gives preliminary mediation suggestions to promote the two parties to reach an agreement quickly;

[0209] Manual Arbitration: For complex disputes, the system assists in appointing a neutral third party for final arbitration.

[0210] The arbitration results and solutions are recorded on the blockchain for reference in future cooperation. The system also provides a dispute analysis report to help enterprises identify and prevent common dispute types, and optimize contract terms and business processes.

[0211] The present invention also provides an all - process collaborative production logistics management optimization method, including the following steps:

[0212] Step S1: Receive inquiry requirements, screen candidate freight forwarders through a dynamic weight algorithm, push encrypted inquiry forms, receive quotation information, and generate an optimal quotation plan.

[0213] In this step, the optimal freight forwarder is screened through a dynamic weight algorithm to achieve the intelligence and high efficiency of the inquiry process, specifically including:

[0214] Step S11: Receive the inquiry requirements submitted by the user, including cargo type, weight, volume, origin, destination, and time - limit requirements; the system automatically standardizes the requirements and checks the data integrity and compliance;

[0215] Step S12: Screen suitable candidate freight forwarders based on historical cooperation records, service quality scores, and price levels through a dynamic weight algorithm; the system comprehensively considers the freight forwarder qualifications, capacity matching degree, and professional fields to ensure that the candidate freight forwarders have corresponding service capabilities;

[0216] Step S13: Push the encrypted inquiry form to the screened freight forwarders through the API interface or email system; adopt end - to - end encryption technology to ensure the security of inquiry information, and at the same time support two modes of batch inquiry and targeted inquiry;

[0217] Step S14: Receive the quotation information submitted by the freight forwarder and perform format standardization processing; the system automatically identifies the quotation structures of each freight forwarder, uniformly converts them into a comparable standard format, and marks out the details of each item of fees;

[0218] Step S15: Generate an optimal quotation plan by comprehensively considering price, timeliness, and service quality; the system provides an interface for comparing and analyzing plans, supports multi-dimensional parameter adjustment, and assists users in making the final decision.

[0219] In this embodiment, the dynamic weight algorithm dynamically adjusts the weights of various evaluation indicators according to different cargo types and transportation routes. For example, for goods with high timeliness requirements, the weights of service quality and on-time rate will be correspondingly increased; for bulk commodities, the weight of the price factor will be correspondingly increased. The mathematical expression of the algorithm is:

[0220] Score = Σ(Wi × Pi);

[0221] Among them, Score is the comprehensive score of the freight forwarder, Wi is the weight of the i-th evaluation indicator, and Pi is the score of the freight forwarder on the i-th evaluation indicator. The weight Wi is dynamically calculated by the following formula:

[0222] Wi = Bi + α × Ti + β × Vi + γ × Ri

[0223] Among them, Bi is the basic weight, Ti is the timeliness factor, Vi is the value factor, Ri is the risk factor, and α, β, and γ are dynamic adjustment coefficients. The system will automatically optimize these coefficients according to the historical order execution situation. The introduction of the risk factor Ri enables the system to evaluate potential logistics risks, such as the impact of external factors such as weather on logistics execution.

[0224] Step S2: Automatically parse the order data in the ERP system through the API interface, generate a shipping order, integrate the data of the weighing sensor, automatically calculate the logistics cost, and trigger an early warning mechanism for abnormal orders.

[0225] This step realizes seamless connection from order to shipment, greatly improving the order processing efficiency and accuracy, specifically including:

[0226] Step S21: Automatically parse the order data in the ERP system through the API interface and extract the shipment-related information; the system supports the parsing of multiple data formats, including but not limited to XML, JSON, CSV, and proprietary formats;

[0227] Step S22: Automatically generate a standardized shipping order based on the parsed order data; the system intelligently selects the corresponding document template according to the requirements of different destinations and logistics methods to ensure that the documents comply with the regulations of customs and logistics providers in various countries;

[0228] Step S23: Integrate the data of the electronic weighing sensor, automatically record the actual weight of the goods, and compare it with the declared weight of the order; the system supports the fusion of multi-point weighing data to improve the measurement accuracy, and at the same time automatically marks the goods with a volume weight greater than the actual weight to optimize the freight calculation;

[0229] Step S24: Trigger the early warning mechanism for overweight, address anomalies, special goods, etc., and push it to the relevant responsible persons; the system adopts a hierarchical early warning strategy, which is divided into three levels: prompt, warning, and emergency according to the degree of anomaly, and corresponding notification and processing processes are adopted for different levels.

[0230] In this embodiment, the system supports the integration with mainstream ERP systems (such as SAP, Oracle, Kingdee, etc.). Through standardized API interfaces or database connections, order data can be obtained in real time. The system adopts a microservices architecture, enabling each functional module to be independently upgraded and expanded to ensure compatibility with new ERP systems.

[0231] For abnormal orders, the system not only provides an early warning function but also has an intelligent auxiliary decision-making mechanism. Based on historical processing experience, it provides recommended solutions for users. The system will continuously learn and update these solutions to continuously improve the efficiency and quality of anomaly handling.

[0232] Step S3: Use OCR technology to identify the goods invoice information, automatically extract the customs declaration fields, conduct compliance verification through the customs system API, and generate a standardized customs declaration form.

[0233] Specifically, step S3 includes the following sub-steps:

[0234] This step realizes the intelligentization and automation of the customs declaration process, improving the customs declaration efficiency and compliance. Specifically, it includes:

[0235] Step S31: Use OCR technology to automatically identify the goods invoice information and extract the fields required for customs declaration; the system supports the identification of invoices in multiple languages and formats, including PDF, pictures, and scanned documents, and has the ability of automatic correction and learning;

[0236] Step S32: Through the customs system API interface, conduct compliance verification on the extracted customs declaration information; the system is connected to the customs databases and HS code libraries of various countries in real time to ensure that the declared information meets the requirements of the latest regulations, and at the same time pre-detects potential compliance risks;

[0237] Step S33: Based on the verified information, automatically generate a standardized customs declaration form that meets the requirements of the customs; the system intelligently fills in the necessary fields according to the customs declaration requirements of different countries and regions, and provides multi-language support to ensure the accuracy and effectiveness of the declared documents;

[0238] Step S34: Monitor the customs declaration progress in real time and trigger an early warning notification when an anomaly occurs; the system tracks the status changes of each link through continuous interaction with the customs system, provides a visual progress display, and predicts the clearance time.

[0239] In this embodiment, the system adopts an OCR model trained by a deep learning algorithm, which can accurately identify invoices in various formats. The OCR model is based on a combined architecture of a convolutional neural network (CNN) and a recurrent neural network (RNN). First, the CNN extracts image features, and then the RNN performs sequence recognition, and finally outputs a text result. The model adopts a continuous learning mechanism and continuously optimizes the recognition accuracy through user feedback. Especially for invoices in enterprise-specific formats, the accuracy rate can reach more than 98%.

[0240] The system also integrates a global commodity coding knowledge base, which can intelligently recommend the most suitable HS code according to the commodity description and provide a customs duty query function for various countries, supporting users to conduct customs duty cost assessment and optimization. For frequently changing regulations and requirements, the system ensures the timely update of the compliance rule base by cooperating with professional data service providers.

[0241] Step S4: Integrate the API interfaces of multiple logistics platforms, obtain the cargo location information in real time, construct a visualized logistics track map, and trigger a warning notice in case of logistics anomalies.

[0242] Specifically, step S4 includes the following sub-steps:

[0243] This step realizes the real-time tracking and monitoring of the whole process of the cargo, improving the logistics transparency and response speed, specifically including:

[0244] Step S41: Integrate the cargo tracking information of multiple logistics platforms through the API interface; the system supports the data integration of more than 40 global mainstream logistics providers, realizing "one-stop" logistics information acquisition and eliminating information silos;

[0245] Step S42: Based on the integrated logistics data, construct a visualized logistics track map to realize the real-time display of the cargo location; the system supports various visualization methods such as 2D / 3D map display, timeline playback, and expansion of logistics node details, meeting the monitoring needs of different scenarios;

[0246] Step S43: Set intelligent logistics anomaly detection rules, and automatically trigger a hierarchical warning notice when a logistics anomaly is detected; the system supports custom rule configuration, including multiple monitoring dimensions such as geofence monitoring, delay monitoring, and temperature and humidity anomalies (for special goods);

[0247] Step S44: Establish a logistics delay prediction model to predict possible delay risks based on historical data analysis; the system comprehensively considers multi-dimensional data such as seasonal factors, weather conditions, and historical performance to early warn of potential risks.

[0248] In this embodiment, the system integrates API interfaces of mainstream domestic and foreign logistics platforms including DHL, FedEx, UPS, and SF Express. Through a unified data format standard, seamless integration of data from different logistics platforms is achieved. The system uses WebSocket technology to achieve real-time push of logistics information and combines edge computing technology to preprocess data at the IoT device end, reducing transmission load and improving update efficiency.

[0249] The visual logistics track map adopts a hierarchical rendering technology to achieve seamless zooming from the global view to the specific internal view of the warehouse, supporting real-time tracking of up to 10,000 goods in transit simultaneously. For high-value or sensitive goods, the system provides special tracking services, increasing the monitoring frequency and accuracy.

[0250] The logistics anomaly detection system uses a multi-modal fusion algorithm to comprehensively analyze multi-dimensional data such as location, time, and environment, reducing the false alarm rate and improving the accuracy of anomaly detection. The delay prediction accuracy of the system can reach over 90% on standard routes, providing strong support for production plan adjustment and customer expectation management.

[0251] Step S5: Analyze the logistics cost data, generate multi-dimensional cost reports, identify cost optimization opportunities, and provide intelligent cost reduction suggestions.

[0252] In this step, through the refined analysis of logistics costs, the potential for cost reduction is explored, and the logistics economy is improved. Specifically, it includes:

[0253] Step S51: Collect and integrate the logistics cost data of each link; the system supports multi-channel data collection, including automatic import, manual entry, and docking with third-party systems, ensuring the comprehensiveness of cost data;

[0254] Step S52: Generate a logistics cost analysis report based on a multi-dimensional analysis model; the system provides standardized and customized report templates, supporting cross-analysis in multiple dimensions such as product, region, time period, and logistics method;

[0255] Step S53: Identify cost anomaly points and optimization opportunities through advanced data mining techniques; the system uses methods such as clustering analysis, association rule mining, and time series analysis to automatically mark business scenarios where costs deviate from the normal range;

[0256] Step S54: Combine historical data and industry benchmarks to provide intelligent cost reduction suggestions; the system generates targeted optimization plans based on the enterprise's historical best practices and industry benchmark data and evaluates the implementation effects.

[0257] In this embodiment, the system adopts data warehouse and big data analysis technologies to support the storage and processing of PB-level logistics cost data. The cost analysis model contains more than 200 preset indicators, covering all aspects such as transportation, warehousing, packaging, manpower, and management. Users can flexibly select and combine them according to their needs.

[0258] The identification of abnormal costs adopts a multi-level detection mechanism. First, possible abnormal points are screened through statistical methods, then secondary verification is carried out through machine learning models, and finally, it is confirmed in combination with business rules, effectively reducing false alarms. The abnormal type knowledge base maintained by the system contains more than 50 common logistics cost abnormal patterns, supporting quick positioning of the root cause of problems.

[0259] The cost reduction suggestion engine is based on case-based reasoning and knowledge graph technologies. It can provide suggestions from multiple dimensions such as packaging optimization, route adjustment, transportation capacity integration, and inventory strategy optimization according to the specific situation of the enterprise. The system also supports simulation analysis functions, allowing users to evaluate the potential benefits and risks of different optimization measures before implementation.

[0260] Step S6: Establish a collaborative mechanism among suppliers, producers, and logistics providers to achieve information sharing and process collaboration, and improve the overall supply chain efficiency.

[0261] This step realizes the close collaboration of all parties participating in the supply chain and improves the overall operation efficiency, specifically including:

[0262] Step S61: Build a supply chain collaboration platform based on blockchain and microservices architecture; the system adopts the design concept of business centralization and technology decentralization to ensure the balance between collaboration efficiency and data security;

[0263] Step S62: Implement a data security sharing mechanism among all parties; the system supports fine-grained data access control to ensure that each party can only access data within the authorized scope. At the same time, through data watermarking and access log tracking, data abuse is prevented;

[0264] Step S63: Establish a multi-party collaborative decision-making model to support the decision-making process jointly participated by supply chain members; the system provides a decision-making framework based on game theory to balance the interests of all parties and achieve overall optimization;

[0265] Step S64: Design smart contracts to automatically execute key nodes in the collaborative process; the system pre-sets smart contract templates for various scenarios, supports visual customization, and complex business logics can be configured without programming.

[0266] In this embodiment, the collaborative platform is based on the consortium blockchain architecture, and each participating party acts as a node on the chain to jointly maintain the ledger. The system adopts a hierarchical design, separating the frequently changing business data from the core transaction data for storage, improving the system performance while ensuring the immutability of key data. The platform supports two-way integration with mainstream enterprise systems to achieve seamless data flow.

[0267] The data security sharing mechanism combines homomorphic encryption, zero-knowledge proof, and federated learning technologies, enabling all parties to collaborate while protecting sensitive data. For example, suppliers can learn about the overall demand trends of manufacturers without knowing specific customer information; logistics providers can optimize the overall transportation capacity without obtaining the specific profit data of each business.

[0268] The collaborative decision-making model supports various collaboration scenarios, including joint demand forecasting, collaborative replenishment planning, transportation capacity sharing, and joint inventory management, etc. The system guides all parties to actively participate in collaborative decision-making by setting reasonable incentive mechanisms and distributes corresponding benefits according to the contribution degree.

[0269] The smart contract realizes the automated execution of business processes, significantly reducing manual intervention and communication costs. The system supports three contract execution methods: condition trigger, time trigger, and event trigger, and has built-in exception handling and rollback mechanisms to ensure the system reliability in case of unstable network or abnormal parameters.

[0270] Through the collaborative implementation of the above six steps, the method of the present invention realizes the full-process management from inquiry to cost optimization, breaks the information silos in traditional logistics management, improves logistics efficiency and transparency, reduces logistics costs, and has a significant optimization effect on the production logistics management of manufacturing enterprises.

[0271] Although some specific embodiments of the present invention have been described in detail through examples, those skilled in the art should understand that the above examples are only for illustration, not for limiting the scope of the present invention. Those skilled in the art should understand that the above embodiments can be modified without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. An optimized system for the whole-process collaborative production logistics management, characterized in that, Including: An inquiry management module, which is used to receive inquiry requirements, screen candidate freight forwarders by using a dynamic weight algorithm combined with a machine learning model, push an encrypted inquiry form generated by using the AES encryption algorithm, receive quotation information, and generate an optimal quotation plan based on a multi-objective optimization algorithm; A shipping management module, which is used to automatically parse order data in the ERP system through the API interface of the OData protocol, generate a shipping order, integrate load cell data, automatically calculate logistics costs, and trigger an early warning mechanism for abnormal orders; A customs brokerage module, which is used to identify cargo invoice information by using OCR technology, automatically extract customs declaration fields, conduct compliance verification through the customs system API, and generate a standardized customs declaration form; An in-transit order tracking module, which is used to integrate the API interfaces of multiple logistics platforms, pull logistics status data in real time, and push it to the client in real time through WebSocket technology to update the order status; And A logistics cost management module, which is used to build a distributed ledger based on blockchain technology, record the details of logistics costs, support automatic conversion and settlement of multiple currencies, and provide a function for arbitration of cost disputes, wherein the distributed ledger adopts the PBFT consensus mechanism.

2. The full-process collaborative production logistics management optimization system according to claim 1, wherein The inquiry management module includes: An inquiry requirement receiving unit, which is used to receive the inquiry requirements input by the salesman; A freight forwarder screening unit, which is used to screen candidate freight forwarders from the freight forwarder database based on the dynamic weight algorithm of the machine learning model, and the weight factors of the dynamic weight algorithm include the historical service quality of the freight forwarder, price competitiveness, service matching degree, capacity guarantee, and compliance performance; An inquiry form pushing unit, which is used to generate an encrypted inquiry form by using the AES encryption algorithm and push it to the candidate freight forwarders; A quotation receiving unit, which is used to receive the quotation information fed back by the freight forwarder system; and A price comparison and analysis unit, which is used to automatically conduct price comparison and analysis. The price comparison and analysis is based on a multi-objective optimization algorithm, comprehensively considers the total cost, transportation timeliness, service quality, and risk factors, and generates an optimal quotation plan.

3. The full-process collaborative production logistics management optimization system according to claim 1, wherein The shipping management module includes: An order parsing unit, which is used to automatically parse the order data in the ERP system through the API interface of the OData protocol; A shipping order generating unit, which is used to generate a shipping order according to the parsed order data; A weighing integration unit, which is used to integrate load cell data and automatically calculate logistics costs. The weighing integration unit supports integration with platform scales, truck scales, and DWS systems; and An abnormal early warning unit, which is used to trigger an early warning mechanism for abnormal orders and push early warning information through a mobile application. The abnormal situations detected by the early warning mechanism include weight abnormalities, address abnormalities, logistics delays, special goods, and compliance risks.

4. The full-process collaborative production logistics management optimization system according to claim 1, characterized in that The customs brokerage module includes: An invoice recognition unit, which is used to identify cargo invoice information by using OCR technology and automatically extract customs declaration fields. The OCR technology is based on a deep learning model and is trained and optimized for the commercial invoice format; A compliance verification unit, which is used to conduct compliance verification through the customs system API; and A customs declaration form generating unit, which is used to generate a standardized customs declaration form.

5. The full-process collaborative production logistics management optimization system according to claim 1, characterized in that, The in-transit order tracking module includes: A logistics data integration unit, which is used to integrate API interfaces of multiple logistics platforms and pull logistics status data in real time. The logistics data integration unit integrates data in a hybrid mode of active pulling + passive receiving; A real-time push unit, which is used to push logistics status data to the client in real time through WebSocket technology; and An order status update unit, which is used to update the order status.

6. The full-process collaborative production logistics management optimization system according to claim 1, characterized in that, The logistics cost management module includes: A blockchain ledger unit, which is used to build a distributed ledger based on blockchain technology and record logistics cost details. The blockchain ledger adopts a consortium chain architecture; A multi-currency settlement unit, which is used to support automatic conversion and settlement of multiple currencies; and A cost arbitration unit, which is used to provide a cost dispute arbitration function and verify the authenticity and accuracy of cost details by using blockchain evidence storage technology.

7. An optimization method for the whole-process collaborative production logistics management, characterized in that, It includes the following steps: Step S1: Receive an inquiry demand, screen candidate freight forwarders by using a dynamic weight algorithm in combination with a machine learning model, push an encrypted inquiry form generated by using the AES encryption algorithm, receive quotation information, and generate an optimal quotation plan based on a multi-objective optimization algorithm; Step S2: Automatically parse order data in the ERP system through the API interface of the OData protocol, generate a shipping note, integrate load cell data, automatically calculate logistics costs, and trigger an early warning mechanism for abnormal orders; Step S3: Use OCR technology to identify goods invoice information, automatically extract customs declaration fields, perform compliance verification through the customs system API, and generate a standardized customs declaration form; Step S4: Integrate API interfaces of multiple logistics platforms, pull logistics status data in real time, push it to the client in real time through WebSocket technology, and update the order status; And Step S5: Build a distributed ledger based on blockchain technology, record logistics cost details, support automatic conversion and settlement of multiple currencies, and provide a cost dispute arbitration function, where the distributed ledger adopts the PBFT consensus mechanism.

8. The full-process collaborative production logistics management optimization method according to claim 7, characterized in that The step S1 includes: Step S11: Receive the inquiry demand input by the salesman; Step S12: Screen candidate freight forwarders from the freight forwarder database based on the dynamic weight algorithm of the machine learning model. The weight factors of the dynamic weight algorithm include the historical service quality of the freight forwarder, price competitiveness, service matching degree, transport capacity guarantee, and compliance performance; Step S13: Generate an encrypted inquiry form by using the AES encryption algorithm and push it to the candidate freight forwarders; Step S14: Receive the quotation information fed back by the freight forwarder system; and Step S15: Automatically conduct price comparison analysis. The price comparison analysis is based on a multi-objective optimization algorithm, comprehensively considers the total cost, transport timeliness, service quality, and risk factors, and generates an optimal quotation plan.

9. The full-process collaborative production logistics management optimization method according to claim 7, wherein The step S2 includes: Step S21: Automatically parse order data in the ERP system through the API interface of the OData protocol; Step S22: Generate a shipping note according to the parsed order data; Step S23: Integrate load cell data, automatically calculate logistics costs. The integrated load cell data supports integration with platform scales, truck scales, and DWS systems; and Step S24: Trigger an early warning mechanism for abnormal orders, and push early warning information through the mobile application. The abnormal situations detected by the early warning mechanism include weight abnormality, address abnormality, logistics delay, special goods, and compliance risks.

10. The full-process collaborative production logistics management optimization method according to claim 7, wherein The said step S3 includes: Step S31: Use OCR technology to identify the goods invoice information and automatically extract the customs declaration fields. The OCR technology is based on a deep learning model and is trained and optimized for the commercial invoice format; Step S32: Conduct compliance verification through the customs system API; and Step S33: Generate a standardized customs declaration form.

Images