Intelligent supply chain optimization management method and system based on AI

Abstract

The invention discloses an AI-based intelligent supply chain optimization management method and system, and relates to the technical field of supply chain optimization management, and the method comprises the following steps: 1, building a data collection model, carrying out the collection of data, and carrying out the all-around processing of the data after the collection, and 2, formulating a transportation scheme, carrying out the design of a transportation mode in advance, and carrying out the optimization of the transportation mode. The method comprises the following steps: step 1, formulating a demand scheme, formulating an inventory management model, and managing inventory positions in a transportation process, and step 2, formulating a scheme optimization and adjustment model to optimize a supply chain, and formulating a bidirectional problem collection scheme in an implementation process to collect problems generated during implementation of the demand scheme and the scheme optimization and adjustment model. According to the method, in the process of optimizing the supply chain, the demand is predicted, the demand scheme is formulated after prediction is completed, stage processing is carried out in the formulating process, and the most real demand can be accurately predicted.

Description

Technical Field

[0001] This invention relates to the field of supply chain optimization management technology, specifically to an AI-based intelligent supply chain optimization management method and system. Background Technology

[0002] Currently, a supply chain is a network composed of organizations, people, activities, information, and resources. Its ultimate goal is to deliver products or services from suppliers to end customers. Chinese patent CN120525255A discloses "A Smart Supply Chain Optimization Management Method and System Based on Digital AI Analysis." This method includes: collecting supply chain data to construct a resource database; using LSTM to analyze sales patterns and K-means++ to cluster customer characteristics and predict demand; calculating inventory parameters and supplier scores to form a collaborative dataset; analyzing product associations through knowledge graphs to generate product selection decisions; applying multi-objective optimization for resource allocation to achieve cross-channel inventory collaboration; and constructing points rules and merchant empowerment schemes based on transaction data to form a digital asset closed-loop mechanism. This application, by establishing a smart supply chain optimization management method based on AI analysis, achieves global visualization, dynamic scheduling, and precise allocation of inventory resources, solving the technical challenges of cross-channel inventory collaborative allocation and significantly improving the overall supply chain's operational efficiency and resource utilization. Existing technologies only address issues such as information silos in the existing supply chain, the inability to adapt to market demands, the inability to manage and coordinate inventory, and the inability to optimize costs, efficiency, and service. Furthermore, they lack tiered processing capabilities, resulting in a general approach that reduces efficiency. They also cannot combine transportation and inventory management, leading to insufficient supply chain efficiency. Additionally, they are not adaptable to various supply chains, limiting their applicability. Moreover, they cannot collect data on problems encountered during use, hindering real-time optimization and adjustments to solutions when issues arise. Summary of the Invention

[0003] The purpose of this invention is to provide an AI-based intelligent supply chain optimization management method and system to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an AI-based intelligent supply chain optimization management method, comprising the following steps: Step 1: Build a data acquisition model to collect data, and after the data is collected, process the data in all aspects. After the processing is completed, build a demand prediction model to analyze the demand in real time. After the demand analysis is completed, formulate a demand solution. Step Two: Develop a transportation plan and design the transportation methods in advance. At the same time, develop an inventory management model to manage the inventory location during transportation, adjust the transportation plan, and build a collaborative processing model to handle internal and external issues collaboratively. Step 3: Develop a solution optimization and adjustment model to optimize the supply chain. During implementation, develop a two-way problem collection plan to collect problems that arise during the implementation of the demand plan and the solution optimization and adjustment model. Once the collection is complete, optimize the solution.

[0005] Preferably, in step one, the data acquisition model collects various data from multiple supply chains, including basic data, dynamic data, and related data. The basic data includes product data, main data, facility data, and rule data. The dynamic data includes data from the procurement process, production process, inventory process, logistics process, sales and demand data, and financial and cost data. The related data includes market data, environmental data, and other data. After the data collection is completed, it is integrated and processed. A real-time integration plan is formulated to integrate and process the data collected in real time, including data format integration and unification, data correction and noise reduction, etc. A data storage repository is built to centrally collect and store the processed data, and the data is marked during the storage process.

[0006] Preferably, in step one, the demand forecasting model formulates a demand forecasting plan based on the data processed and analyzed within the data repository. When formulating the demand forecasting plan, multiple supply chains are classified, with the most complex supply chain type being Class A, the moderately complex type being Class B, and the conventional type being Class C. Specifically, the demand forecasting plan includes a first demand forecasting plan, a second demand forecasting plan, a third demand forecasting plan, and a collaborative adjustment plan. The first demand forecasting plan forecasts for Class A supply chains, the second demand forecasting plan forecasts for Class B supply chains, and the third demand forecasting plan forecasts for Class C supply chains. The collaborative adjustment plan receives data from the upstream and downstream of multiple supply chains in real time, analyzes it immediately after receiving the data, and adjusts the first, second, and third demand forecasting plans after the analysis is completed.

[0007] Preferably, the demand scheme in step one is formulated based on the forecast results of the demand forecasting scheme, and the demand scheme includes a first demand scheme, a second demand scheme and a third demand scheme. The first demand scheme processes the forecast results of the first demand forecasting scheme, the second demand scheme processes the forecast results of the second demand forecasting scheme, and the third demand scheme processes the forecast results of the third demand forecasting scheme.

[0008] Preferably, in step two, the transportation plan is formulated for multiple supply chain transportation modes and routes. The formulation first prioritizes the objectives and establishes core elements for transportation, including transportation mode, carrier, route planning, and packaging and protection. Transportation modes include road, rail, air, sea, and multimodal transport. Carriers are categorized based on "qualification + service quality + price" and are graded. Route planning avoids congested sections and integrates transshipment nodes. Packaging and protection are matched according to cargo characteristics, and high-time-efficiency and low-cost requirements are defined. High-time-efficiency requirements prioritize air freight and same-city express, while low-cost requirements prioritize rail, sea freight, and full truckload / less-than-truckload (LTL). A monitoring and emergency response mechanism is established, including real-time tracking and anomaly response. Real-time tracking monitors the cargo location and status through a TMS system, and anomaly response involves developing alternative plans for delays, damage, and weather impacts.

[0009] Preferably, in step two, the inventory management model first classifies and manages the inventory, with levels including Level 1, Level 2, Level 3, Level 4, Level 5, and Level 6. Level 1 represents high value and low sales volume, and strictly monitors inventory levels, adopting a lean inventory strategy to reduce capital occupation. Level 2 represents medium value and medium sales volume, balancing inventory costs and supply risks, and employing regular inventory checks and appropriate replenishment. Level 3 represents low value and high sales volume, simplifying management processes, and using an economic order quantity model for bulk purchasing to reduce operating costs. Level 4 represents stable demand, employing fixed-cycle replenishment to reduce frequent ordering costs. Level 5 represents medium volatility, combining reorder point (ROP) and safety stock management. Level 6 represents high volatility, employing agile response strategies, such as safety stock + dynamic replenishment or through supplier collaborative management.

[0010] Preferably, the specific construction steps of the collaborative processing model in step two are as follows; (1) Process standardization and flexibility are addressed, the entire process of "customer demand capture - production - logistics - after-sales" is sorted out, non-value-added links are identified and optimized, and flexible process modules are designed to cope with sudden demands; (2) Establish a collaborative business model, implement models such as VMI, JIT, and CPFR, and promote the upgrading of the supply chain from a transactional relationship to a partnership; (3) Establish cross-functional teams and rapid response mechanisms, set up cross-departmental supply chain collaboration groups, adopt OKR management model to align objectives, and establish an end-to-end responsibility mechanism; (4) Formulate organizational collaboration plan: ① Promote internal resource integration and information flow through organizational restructuring; ② Establish strategic alliances externally, sign long-term collaboration agreements with core suppliers and logistics providers, clarify the profit distribution and risk-sharing mechanism, and jointly build a joint innovation center; ③ Establish a supply chain collaboration performance management system, measure collaboration effect through quantitative indicators, conduct regular performance evaluations, and reward outstanding collaborators. (5) Deploy sensors and RFID tags in the warehousing and transportation links to collect data such as inventory, temperature and humidity and vehicle location in real time, and build a real-time storage database for storage. Use AI algorithms to optimize demand forecasting, production plan adjustment and transportation route planning to improve the intelligence level of the supply chain. (6) Build a digital twin model of the supply chain and simulate the system response under scenarios such as demand fluctuations and facility failures through the digital twin model of the supply chain to optimize the layout in advance; (7) Monitor key indicators such as OTD and inventory turnover rate in real time, hold a collaborative review meeting every quarter, optimize strategies based on data iteration, identify and assess potential risks in the supply chain, and formulate corresponding risk response measures.

[0011] Preferably, in step three, the scheme optimization and adjustment model formulates optimization and adjustment schemes, and the optimization and adjustment schemes include a first optimization and adjustment scheme, a second optimization and adjustment scheme, and a third optimization and adjustment scheme. The optimization and adjustment schemes optimize and adjust the transportation scheme, inventory management, and collaborative processing model. The first optimization and adjustment scheme has the largest range, the second optimization and adjustment scheme has a medium range, and the third optimization and adjustment scheme has the smallest range.

[0012] Preferably, in step three, the two-way problem collection scheme collects problems encountered during the implementation of the optimization and adjustment scheme, the demand forecasting scheme, and the demand scheme. After the problem collection is completed, the problems are analyzed and processed. After the analysis and processing, the problems are classified into three levels: Level 1, Level 2, and Level 3. When the analyzed problem is at Level 1, the problem is optimized and processed immediately. When the analyzed problem is at Level 2, the problem is optimized and processed within a certain period of time. When the analyzed problem is at Level 3, the problem can be optimized and processed later.

[0013] The AI-based intelligent supply chain optimization management system includes a data acquisition and processing unit, a transportation and inventory management unit, and a solution optimization unit. The data acquisition and processing unit builds a data acquisition model to collect data, and performs comprehensive processing on the data after the acquisition is completed. After the processing is completed, it builds a demand prediction model to analyze the demand in real time, and formulates a demand solution after the demand analysis is completed. The transportation and inventory management unit formulates transportation plans and designs transportation methods in advance. At the same time, it develops an inventory management model to manage the location of inventory during transportation, adjusts the transportation plan, and builds a collaborative processing model to handle internal and external issues collaboratively. The scheme optimization unit formulates a scheme optimization and adjustment model to optimize the supply chain. During the implementation process, it formulates a two-way problem collection scheme to collect problems generated during the implementation of the demand scheme and the scheme optimization and adjustment model. Once the collection is completed, optimization processing is carried out.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This invention optimizes the supply chain by forecasting demand and formulating demand plans accordingly. The plan is then processed in a tiered manner, enabling accurate prediction of the most realistic demand. This leads to greater efficiency in subsequent transportation planning and inventory management. Furthermore, a collaborative processing model is established to comprehensively receive information from the supply chain and transmit it across all channels immediately, breaking down information silos. During implementation, problems are collected and analyzed to optimize the implementation plan, ensuring it remains at its optimal state and enhancing problem-solving capabilities. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method flow provided in an embodiment of the present invention. Detailed Implementation

[0016] 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.

[0017] Example 1 Please see Figure 1 This invention provides a technical solution: an AI-based intelligent supply chain optimization management method, comprising the following steps: Step 1: Build a data acquisition model to collect data, and after the data is collected, process the data in all aspects. After the processing is completed, build a demand prediction model to analyze the demand in real time. After the demand analysis is completed, formulate a demand solution. Step Two: Develop a transportation plan and design the transportation methods in advance. At the same time, develop an inventory management model to manage the inventory location during transportation, adjust the transportation plan, and build a collaborative processing model to handle internal and external issues collaboratively. Step 3: Develop a solution optimization and adjustment model to optimize the supply chain. During implementation, develop a two-way problem collection plan to collect problems that arise during the implementation of the demand plan and the solution optimization and adjustment model. Once the collection is complete, optimize the solution.

[0018] In step one, the data acquisition model collects various data from multiple supply chains, including basic data, dynamic data, and related data. The basic data includes product data, main data, facility data, and rule data. The dynamic data includes data from the procurement process, production process, inventory process, logistics process, sales and demand data, and financial and cost data. The related data includes market data, environmental data, and other data. After the data collection is completed, it is integrated and processed. A real-time integration plan is developed to integrate and process the data collected in real time, including data format integration and unification, data correction and noise reduction, etc. A data storage repository is built to centrally collect and store the processed data, and the data is marked during the storage process. The procurement data includes: ① Order data: purchase order number, order time, quantity, amount, delivery date and performance status; ② Execution data: supplier delivery time, arrival time, acceptance results, non-conformity rate and return / exchange records; ③ Cost data: purchase unit price, transportation cost, tax, negotiation cost and supplier service fee. The production data includes: ① Planning data: production orders, scheduling plans, capacity allocation and material requirements planning (MRP) execution status; ② Process data: production progress, equipment operating status, labor hours consumption, work-in-process inventory and quality inspection data; ③ Loss data: material loss rate, scrap rate and rework times and reasons. The inventory data includes: ① Real-time inventory: SKU quantity, batch, expiration date and inventory value of each warehouse / location; ② Change data: Inbound records (source, time and quantity), outbound records (destination, time and quantity) and transfer records; ③ Early warning data: Inventory backlog days, stockout warnings and details of near-expiry / expired products. The logistics data includes: ① Transportation data: waybill information, carrier, vehicle information, loading and unloading time, on-the-way location and delivery time; ② Cost data: transportation fee, warehousing fee, loading and unloading fee, packaging fee and compensation amount; ③ Timeliness and quality data: on-time delivery rate, cargo damage rate, loss rate and customer receipt feedback. The sales and demand data include ① order data: sales order number, order time, customer, quantity, amount and delivery address; ② fulfillment data: order confirmation time, outbound time, delivery time, receipt time and reason for return or exchange; ③ demand data: historical sales volume, order fluctuations, promotional activity effects, customer forecast orders and terminal inventory. The financial and cost data includes: ① Cost data: procurement costs, production costs, logistics costs, inventory holding costs, and stockout losses; ② Settlement data: accounts payable / receivable amounts, payment / collection times, invoice information, and reconciliation discrepancies; ③ Profitability data: gross profit per unit, regional profitability, customer profit contribution, and overall supply chain operating costs.

[0019] In step one, the demand forecasting model formulates a demand forecasting plan based on the data processed and analyzed within the data repository. When formulating the demand forecasting plan, multiple supply chains are classified. The most complex supply chain is classified as Class A, the moderately complex as Class B, and the conventional as Class C. Specifically, the demand forecasting plan includes a first demand forecasting plan, a second demand forecasting plan, a third demand forecasting plan, and a collaborative adjustment plan. The first demand forecasting plan forecasts for Class A supply chains, the second for Class B supply chains, and the third for Class C supply chains. The collaborative adjustment plan receives data from the upstream and downstream of multiple supply chains in real time, analyzes it immediately after receiving the data, and adjusts the first, second, and third demand forecasting plans accordingly.

[0020] The demand plan in step one is formulated based on the forecast results of the demand forecasting plan, and the demand plan includes a first demand plan, a second demand plan and a third demand plan. The first demand plan processes the forecast results of the first demand forecasting plan, the second demand plan processes the forecast results of the second demand forecasting plan, and the third demand plan processes the forecast results of the third demand forecasting plan. Furthermore, the demand solution integrates demand data during implementation and uses intelligent forecasting tools for real-time demand monitoring. This integration connects CRM, ERP, e-commerce platforms, and POS systems to build a unified demand data platform, integrating historical sales, customer feedback, promotional records, and external market data (such as competitor activities and weather). When using intelligent forecasting tools, the system categorizes businesses into SMEs and large enterprises. SMEs use Excel and SPSS for basic statistical forecasting (exponential smoothing, moving average), while large enterprises deploy AI forecasting systems (LSTM and random forest). During implementation, multi-dimensional data is integrated to improve forecast accuracy. Real-time demand monitoring establishes demand fluctuation warning thresholds (e.g., actual sales deviation ±15%), monitors demand changes in real time, and automatically adjusts the solution when warnings are triggered (e.g., convening an emergency collaborative meeting when the deviation is too large).

[0021] Step two involves developing transportation plans for multiple supply chains, prioritizing objectives (e.g., FMCG prioritizes timeliness and high-frequency delivery, industrial manufacturing prioritizes cost and stability, and fresh produce prioritizes temperature control and loss control). Core elements for transportation are then established, including transportation methods, carriers, route planning, and packaging and protection. Transportation methods include road (full truckload / less-than-truckload / express), rail, air, sea, and multimodal transport (e.g., road-rail combined, sea-air combined). Carriers are categorized based on "qualifications + service quality + price" and are graded (core carriers + alternative carriers). The route planning involves avoiding congested roads and integrating transit nodes. Packaging and protection are tailored to the characteristics of the goods (e.g., insulated boxes for fresh produce and shockproof packaging for large items). High-speed-rate and low-cost requirements are also defined, with priority given to air freight and same-city express delivery for high-speed-rate requirements, and priority given to rail freight, sea freight, and full truckload / less-than-truckload (LTL) freight for low-cost requirements. Monitoring and emergency response mechanisms are established, including real-time tracking and anomaly response. Real-time tracking monitors the location and status of goods (e.g., cold chain temperature) through the TMS system. Anomaly response involves developing alternative plans for delays, damage, and weather impacts (e.g., alternative routes and emergency carriers). Its transportation solutions include fast-moving consumer goods transportation solutions, industrial manufacturing transportation solutions, fresh food cold chain transportation solutions, cross-border e-commerce transportation solutions, and project-based transportation solutions; The FMCG transportation solution includes high frequency, small batch and wide coverage, and the transportation mode is mainly road less-than-truckload (LTL) and express. In core cities, "trunk line + branch line" direct delivery is opened. When laying out the network, distribution centers are set up in core areas to realize two-level distribution of "central warehouse → distribution warehouse → terminal store". During transportation, scattered orders are integrated and long-term cooperation with chain logistics companies is carried out to lock in prices and timeliness. The industrial manufacturing transportation plan includes large items, low frequency, and long distances. The transportation methods prioritize full truckloads by road and dedicated railway trains. Special vehicles are used for large items, and the route planning selects "dedicated line transportation". Risk control is carried out, oversized transport permits are obtained in advance, vehicle load-bearing capacity and fixing devices are checked before transportation, and high-value cargo insurance is purchased. The fresh food cold chain transportation solution includes strict temperature control, high loss and short delivery time. The transportation methods include dedicated cold chain vehicles and refrigerated containers (sea / rail). Refrigerated vans are used for short distances and refrigerated trailers are used for long distances. Temperature control management is implemented, and temperature-zone transportation is adopted (such as differentiating between frozen and refrigerated layers in the same vehicle). IoT sensors monitor the temperature in real time, and abnormalities are automatically alarmed. Optimization is carried out during implementation, prioritizing "door-to-door" direct cold chain to reduce temperature fluctuations during transit. Cooperation with professional cold chain logistics providers ensures compliance with qualifications. The cross-border e-commerce transportation solution includes long distances, multiple customs checkpoints, and high compliance requirements. The transportation methods are international express or postal parcels for small and light items, and sea-air combined transport (such as sea freight to the port + local express delivery) for large items. It also includes customs clearance and warehousing, setting up overseas warehouses in the target country for advance stocking, shortening the delivery time through local delivery, entrusting customs clearance agents to ensure complete documentation, establishing risk response methods, reserving customs clearance buffer time, and purchasing cross-border transportation insurance to cover the risks of lost and detained items. Project-based transportation solutions include customized, large-item, one-off, and highly complex solutions. The transportation mode adopts "specialized transportation + multimodal transport" (such as sea freight to port → special road vehicle → on-site hoisting). Before transportation, preliminary planning is carried out, the transportation route is surveyed in advance (such as bridge load-bearing capacity and tunnel height), a special transportation plan (including loading and unloading procedures and emergency plans) is developed, and collaborative management is carried out. A special team is formed by the carrier, loading and unloading team and the site manager to follow up on the transportation and delivery throughout the process. During the transportation process, diagnose the current transportation status, analyze the characteristics of goods in the supply chain, time requirements, cost budgets, and existing transportation pain points. Conduct pilot verification before implementing the transportation plan, selecting 1-2 core routes for pilot testing, and tracking indicators such as cost, timeliness, and damage rate. Optimize the transportation plan based on the pilot results, sign long-term agreements with carriers, build a TMS system to achieve full-chain visualization, and review transportation data monthly (such as empty load rate, unit transportation cost, and on-time rate) to dynamically adjust routes and carriers (such as replacing suppliers with poor service).

[0022] In step two, the inventory management model first classifies and manages the inventory, with levels including Level 1, Level 2, Level 3, Level 4, Level 5, and Level 6. Level 1 represents high value and low sales volume, and strictly monitors inventory levels, adopting a lean inventory strategy to reduce capital occupation. Level 2 represents medium value and medium sales volume, balancing inventory costs and supply risks, and adopting regular inventory checks and appropriate replenishment. Level 3 represents low value and high sales volume, simplifying management processes, adopting an economic order quantity model for bulk purchasing to reduce operating costs. Level 4 represents stable demand, adopting fixed-cycle replenishment to reduce frequent ordering costs. Level 5 represents medium volatility, combining reorder point (ROP) and safety stock management. Level 6 represents high volatility, adopting an agile response strategy, such as safety stock + dynamic replenishment or through supplier collaborative management. Collaborative processing with suppliers enables them to manage inventory and proactively replenish stock based on real-time sales data, reducing inventory pressure on enterprises. This involves developing collaborative plans, forecasts, and replenishment strategies, specifically sharing demand forecasts, inventory status, and replenishment plans with supply chain partners to achieve end-to-end collaboration. It also integrates procurement, production, and sales data to enable inventory visualization and automated replenishment, real-time tracking of inventory location and status, and reduction of inventory errors and stockout risks. Develop a dynamic replenishment plan to dynamically replenish inventory and manage risks. The specific steps are as follows: (1) Replenishment is triggered when inventory drops to a preset level, and safety stock is used to deal with demand fluctuations; (2) Combining quantitative ordering and periodic ordering, with quantitative ordering replenishing a fixed quantity each time, is suitable for high-value or critical materials, while periodic ordering replenishes at a fixed cycle, which is suitable for low-value or bulk-purchased goods. (3) Perform multi-level inventory optimization processing. Based on the inventory coordination of multi-level nodes in the supply chain (such as factories, warehouses and retail stores), optimize the overall inventory level through mathematical models; (4) Set up safety stock, calculate the safety stock amount based on demand fluctuations, supply cycle and service level requirements, formulate emergency plans in conjunction with scenario analysis (such as epidemics and natural disasters), and use the emergency plan when encountering force majeure during the implementation of the emergency plan; (5) Monitor inventory turnover rate, analyze inventory turnover rate regularly, identify slow-moving products and take promotion, return or scrapping measures.

[0023] The specific construction steps of the collaborative processing model in step two are as follows: (1) To address the standardization and flexibility of processes, we will streamline the entire process of "customer demand capture - production - logistics - after-sales service", identify and optimize non-value-added links (such as merging duplicate inspections and simplifying approval nodes), and design flexible process modules (such as green channels for emergency orders and multi-source procurement switching mechanisms) to cope with sudden demands. (2) Establish a collaborative business model, implement models such as VMI, JIT, and CPFR, and promote the upgrading of the supply chain from a transactional relationship to a partnership; (3) Establish cross-functional teams and rapid response mechanisms, set up cross-departmental supply chain collaboration groups (including procurement, production, logistics and marketing), adopt OKR management model to align objectives, and establish an end-to-end responsibility mechanism; (4) Formulate organizational collaboration plan: ① Promote internal resource integration and information flow through organizational restructuring; ② Establish strategic alliances externally, sign long-term collaboration agreements with core suppliers and logistics providers, clarify the profit distribution and risk-sharing mechanism, and jointly build a joint innovation center; ③ Establish a supply chain collaboration performance management system, measure collaboration effect through quantitative indicators, conduct regular performance evaluations, and reward outstanding collaborators. (5) Deploy sensors and RFID tags in the warehousing and transportation links to collect data such as inventory, temperature and humidity and vehicle location in real time, and build a real-time storage database for storage. Use AI algorithms to optimize demand forecasting, production plan adjustment and transportation route planning to improve the intelligence level of the supply chain. (6) Build a digital twin model of the supply chain and simulate the system response under scenarios such as demand fluctuations and facility failures through the digital twin model of the supply chain to optimize the layout in advance; (7) Monitor key indicators such as OTD and inventory turnover rate in real time, hold a collaborative review meeting every quarter, optimize strategies based on data iteration, identify and assess potential risks in the supply chain (such as supply disruptions, price fluctuations and quality issues), and formulate corresponding risk response measures (such as establishing a diversified supplier system and signing long-term cooperation agreements).

[0024] In step three, the scheme optimization and adjustment model formulates optimization and adjustment schemes, and the optimization and adjustment schemes include a first optimization and adjustment scheme, a second optimization and adjustment scheme, and a third optimization and adjustment scheme. The optimization and adjustment schemes optimize and adjust the transportation scheme, inventory management, and collaborative processing model. The first optimization and adjustment scheme has the largest range, the second optimization and adjustment scheme has a medium range, and the third optimization and adjustment scheme has the smallest range. The optimization and adjustment plan is refined and tailored to different scenarios during implementation. The plan is adjusted for different SKUs, regions and customer groups, and the parameters are dynamically calibrated. Key parameters in the optimization and adjustment plan, such as safety stock coefficient (adjusted according to demand fluctuations, increased from 1.2 to 1.5 when fluctuations are large), replenishment lead time (adjusted according to supplier fulfillment rate, when the fulfillment rate drops from 95% to 85%, the lead time is extended from 7 days to 10 days), and transportation route weight (adjusted delivery priority according to real-time road conditions). During the optimization and adjustment process, we optimized the collaboration with upstream and downstream partners, shared execution data with suppliers / logistics providers, made joint adjustments, and held regular collaborative meetings. For example, we reviewed execution issues with core partners every month (such as the reasons for supplier delivery delays) and jointly developed improvement plans (such as optimizing order placement time to avoid the pressure of concentrated inventory preparation for suppliers). A supply chain command module is built to monitor the execution status of transportation plans, inventory management and collaborative processing models in real time (such as inventory levels, order fulfillment progress, logistics nodes and supplier fulfillment rates), and set up abnormal warnings (such as setting inventory below the safety line and delivery delays exceeding 2 hours). When an abnormal situation occurs, the problem is solved first and the problem is reported for processing.

[0025] In step three, the two-way problem collection scheme collects problems encountered during the implementation of the optimization and adjustment scheme, the demand forecasting scheme, and the demand scheme. After the problem collection is completed, the problems are analyzed and processed. After the analysis and processing, the problems are classified into three levels: Level 1, Level 2, and Level 3. When the analyzed problem is at Level 1, the problem is optimized and processed immediately. When the analyzed problem is at Level 2, the problem is optimized and processed within a certain period of time. When the analyzed problem is at Level 3, the problem can be optimized and processed later. When optimizing, we prioritize implementation, addressing low-cost, high-return issues first, followed by medium-investment, medium-return issues, and finally high-investment, high-return issues. Low-cost, high-return issues involve: current situation diagnosis → process simplification → establishment of collaborative mechanisms. Medium-investment, medium-return issues involve: adjustment of solution parameters → implementation of automation tools. High-investment, high-return issues involve: system integration → construction of a data platform → iteration of AI tools.

[0026] Example 2 The AI-based intelligent supply chain optimization management system includes a data acquisition and processing unit, a transportation and inventory management unit, and a solution optimization unit. The data acquisition and processing unit builds a data acquisition model to collect data, and performs comprehensive processing on the data after the acquisition is completed. After the processing is completed, it builds a demand prediction model to analyze the demand in real time, and formulates a demand solution after the demand analysis is completed. The transportation and inventory management unit formulates transportation plans and designs transportation methods in advance. At the same time, it develops an inventory management model to manage the location of inventory during transportation, adjusts the transportation plan, and builds a collaborative processing model to handle internal and external issues collaboratively. The scheme optimization unit formulates a scheme optimization and adjustment model to optimize the supply chain. During the implementation process, it formulates a two-way problem collection scheme to collect problems generated during the implementation of the demand scheme and the scheme optimization and adjustment model. Once the collection is completed, optimization processing is carried out.

[0027] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0028] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An AI-based intelligent supply chain optimization management method, characterized by: Includes the following steps: Step 1: Build a data acquisition model to collect data, and after the data is collected, process the data in all aspects. After the processing is completed, build a demand prediction model to analyze the demand in real time. After the demand analysis is completed, formulate a demand solution. Step Two: Develop a transportation plan and design the transportation methods in advance. At the same time, develop an inventory management model to manage the inventory location during transportation, adjust the transportation plan, and build a collaborative processing model to handle internal and external issues collaboratively. Step 3: Develop a solution optimization and adjustment model to optimize the supply chain. During implementation, develop a two-way problem collection plan to collect problems that arise during the implementation of the demand plan and the solution optimization and adjustment model. Once the collection is complete, optimize the solution.

2. The AI-based intelligent supply chain optimization management method according to claim 1, characterized in that: In step one, the data acquisition model collects various data from multiple supply chains, including basic data, dynamic data, and related data. The basic data includes product data, main data, facility data, and rule data. The dynamic data includes data from the procurement process, production process, inventory process, logistics process, sales and demand data, and financial and cost data. The related data includes market data, environmental data, and other data. After the data collection is completed, it is integrated and processed. A real-time integration plan is developed to integrate and process the data collected in real time, including data format integration and unification, data correction and noise reduction, etc. A data storage repository is built to centrally collect and store the processed data, and the data is marked during the storage process.

3. The AI-based intelligent supply chain optimization management method according to claim 2, characterized in that: In step one, the demand forecasting model formulates a demand forecasting plan based on the data processed and analyzed within the data repository. When formulating the demand forecasting plan, multiple supply chains are classified. The most complex supply chain is classified as Class A, the moderately complex as Class B, and the conventional as Class C. Specifically, the demand forecasting plan includes a first demand forecasting plan, a second demand forecasting plan, a third demand forecasting plan, and a collaborative adjustment plan. The first demand forecasting plan forecasts for Class A supply chains, the second for Class B supply chains, and the third for Class C supply chains. The collaborative adjustment plan receives data from the upstream and downstream of multiple supply chains in real time, analyzes it immediately after receiving the data, and adjusts the first, second, and third demand forecasting plans accordingly.

4. The AI-based intelligent supply chain optimization management method according to claim 3, characterized in that: The demand plan in step one is formulated based on the forecast results of the demand forecasting plan. The demand plan includes a first demand plan, a second demand plan, and a third demand plan. The first demand plan processes the forecast results of the first demand forecasting plan, the second demand plan processes the forecast results of the second demand forecasting plan, and the third demand plan processes the forecast results of the third demand forecasting plan.

5. The AI-based intelligent supply chain optimization management method according to claim 4, characterized in that: In step two, the transportation plan is formulated for multiple supply chain transportation modes and routes. The plan prioritizes objectives and establishes core elements for transportation, including transportation mode, carrier, route planning, and packaging and protection. Transportation modes include road, rail, air, sea, and multimodal transport. Carriers are categorized based on "qualification + service quality + price" and are graded. Route planning avoids congested sections and integrates transshipment nodes. Packaging and protection are matched to cargo characteristics, and high-time-efficiency and low-cost requirements are defined. High-time-efficiency requirements prioritize air freight and same-city express, while low-cost requirements prioritize rail, sea freight, and full truckload / less-than-truckload (LTL) shipments. Monitoring and emergency mechanisms are established, including real-time tracking and anomaly response. Real-time tracking monitors cargo location and status through a TMS system, and anomaly response involves developing alternative plans for delays, damage, and weather impacts.

6. The AI-based intelligent supply chain optimization management method according to claim 5, characterized in that: In step two, the inventory management model first classifies and manages the inventory, with levels including Level 1, Level 2, Level 3, Level 4, Level 5, and Level 6. Level 1 represents high value and low sales volume, and strictly monitors inventory levels, adopting a lean inventory strategy to reduce capital occupation. Level 2 represents medium value and medium sales volume, balancing inventory costs and supply risks, and employing regular inventory checks and appropriate replenishment. Level 3 represents low value and high sales volume, simplifying management processes and using an economic order quantity model for bulk purchasing to reduce operating costs. Level 4 represents stable demand, employing fixed-cycle replenishment to reduce frequent ordering costs. Level 5 represents medium volatility, combining reorder point (ROP) and safety stock management. Level 6 represents high volatility, employing agile response strategies, such as safety stock + dynamic replenishment or through supplier collaborative management.

7. The AI-based intelligent supply chain optimization management method according to claim 6, characterized in that: The specific construction steps of the collaborative processing model in step two are as follows: (1) Process standardization and flexibility are addressed, the entire process of "customer demand capture - production - logistics - after-sales" is sorted out, non-value-added links are identified and optimized, and flexible process modules are designed to cope with sudden demands; (2) Establish a collaborative business model, implement models such as VMI, JIT, and CPFR, and promote the upgrading of the supply chain from a transactional relationship to a partnership; (3) Establish cross-functional teams and rapid response mechanisms, set up cross-departmental supply chain collaboration groups, adopt OKR management model to align objectives, and establish an end-to-end responsibility mechanism; (4) Formulate organizational collaboration plan: ① Promote internal resource integration and information flow through organizational restructuring; ② Establish strategic alliances externally, sign long-term collaboration agreements with core suppliers and logistics providers, clarify the profit distribution and risk-sharing mechanism, and jointly build a joint innovation center; ③ Establish a supply chain collaboration performance management system, measure collaboration effect through quantitative indicators, conduct regular performance evaluations, and reward outstanding collaborators. (5) Deploy sensors and RFID tags in the warehousing and transportation links to collect data such as inventory, temperature and humidity and vehicle location in real time, and build a real-time storage database for storage. Use AI algorithms to optimize demand forecasting, production plan adjustment and transportation route planning to improve the intelligence level of the supply chain. (6) Build a digital twin model of the supply chain and simulate the system response under scenarios such as demand fluctuations and facility failures through the digital twin model of the supply chain to optimize the layout in advance; (7) Monitor key indicators such as OTD and inventory turnover rate in real time, hold a collaborative review meeting every quarter, optimize strategies based on data iteration, identify and assess potential risks in the supply chain, and formulate corresponding risk response measures.

8. The AI-based intelligent supply chain optimization management method according to claim 7, characterized in that: In step three, the scheme optimization and adjustment model formulates optimization and adjustment schemes, which include a first optimization and adjustment scheme, a second optimization and adjustment scheme, and a third optimization and adjustment scheme. The optimization and adjustment schemes optimize and adjust the transportation scheme, inventory management, and collaborative processing model. The first optimization and adjustment scheme has the largest scope, the second optimization and adjustment scheme has a medium scope, and the third optimization and adjustment scheme has the smallest scope.

9. The AI-based intelligent supply chain optimization management method and system according to claim 8, characterized in that: In step three, the two-way problem collection scheme collects problems encountered during the implementation of the optimization and adjustment scheme, the demand forecasting scheme, and the demand scheme. After the problem collection is completed, the problems are analyzed and processed. After the analysis and processing, the problems are classified into three levels: Level 1, Level 2, and Level 3. When the analyzed problem is at Level 1, the problem is optimized and processed immediately. When the analyzed problem is at Level 2, the problem is optimized and processed within a certain period of time. When the analyzed problem is at Level 3, the problem can be optimized and processed later.

10. An AI-based intelligent supply chain optimization management system, characterized by: The supply chain optimization management system is applicable to the AI-based smart supply chain optimization management method described in claims 1-9, and includes a data acquisition and processing unit, a transportation and inventory management unit, and a solution optimization unit; The data acquisition and processing unit builds a data acquisition model to collect data, and performs comprehensive processing on the data after the acquisition is completed. After the processing is completed, it builds a demand prediction model to analyze the demand in real time, and formulates a demand solution after the demand analysis is completed. The transportation and inventory management unit formulates transportation plans and designs transportation methods in advance. At the same time, it develops an inventory management model to manage the location of inventory during transportation, adjusts the transportation plan, and builds a collaborative processing model to handle internal and external issues collaboratively. The scheme optimization unit formulates a scheme optimization and adjustment model to optimize the supply chain. During the implementation process, it formulates a two-way problem collection scheme to collect problems generated during the implementation of the demand scheme and the scheme optimization and adjustment model. Once the collection is completed, optimization processing is carried out.

Images