Spare and accessory part supply chain full-process digital collaborative management system and method

Abstract

The invention relates to the technical field of supply chain management, in particular to a part supply chain full-process digital collaborative management system and method, and the method comprises the steps: evaluating supplier data through constructing a dynamic scoring model, and obtaining a supplier score; when the supplier score is lower than a predetermined threshold value, generating a multi-source purchase plan, and dynamically adjusting purchase distribution according to the supplier score; constructing a warehouse-in and warehouse-out part intelligent verification model to analyze warehouse-in and warehouse-out parts, generating an acceptance report and block chain record association, and triggering an alarm and freezing an order when a difference is found; a supply chain risk topological graph is constructed through the supplier fulfillment rate, the logistics data and the market demand fluctuation, and high-risk nodes are identified and early warned in real time; initiating emergency purchase to suppliers with high supplier scores based on high-risk nodes, and optimizing inventory scheduling based on geographic proximity; the inventory level is optimized in real time based on the production line state and the in-transit logistics data, and the safe inventory threshold is dynamically adjusted to cope with supply chain interruption.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain management, and in particular, to a full-process digital collaborative management system and method for spare parts supply chain. Background Art

[0002] With the rapid development of digital technologies, enterprises increasingly rely on computer information systems to achieve efficient, accurate, and collaborative operation management in supply chain management. Especially in the medical device industry, due to the particularity of products and strict regulatory requirements, the full-life cycle management of spare parts supply chain is particularly crucial. In the prior art, the supply chain management systems of most enterprises are only limited to data recording and inventory scheduling in a single link, lacking dynamic monitoring and collaborative management of the whole process from procurement, warehousing, storage, outbound, to terminal distribution and after-sales service of products, and it is difficult to form a complete product traceability system. In view of this problem, how to achieve full-life cycle management, ensure that the supply chain has high resilience in the face of emergencies, and use these capabilities to provide reliable and transparent supply chain management support services for downstream spare parts procurement customers has become an urgent technical problem in the industry.

[0003] In existing digital management systems, although some enterprises have begun to use technologies such as barcodes, RFID, and the Internet of Things to automate the transformation of warehouse management, due to the non-uniform data standards between systems and the widespread existence of information silos, it is difficult to achieve real-time data sharing and collaborative decision-making between supply chain nodes. This not only affects inventory accuracy and order fulfillment efficiency, but also increases the risks faced by the medical device supply chain to a certain extent, such as abnormal temperature and humidity, logistics delays, quality problems, etc. At the same time, the multi-link regulatory requirements involved in medical device products require the system to support full-process traceability and data record-keeping to meet the regulations of industry standards such as GMP and GSP. Therefore, how to build a data collaboration platform covering the whole process from procurement, warehousing, inventory, outbound, distribution to after-sales, so that it can not only optimize internal management, but also provide the required data visibility and service guarantee for procurement customers, has become an important direction to improve supply chain resilience and ensure product quality.

[0004] In addition, supply chain resilience, as an important factor in ensuring the continuity of medical device supply, lies in the real-time monitoring, dynamic adjustment, and risk warning of all links in the supply chain through digital means, so as to quickly respond and timely adjust in the face of emergencies (such as logistics interruptions, equipment failures, etc.), and achieve efficient resource scheduling. There is still a lack of a digital collaborative platform in the existing technology that can integrate full life cycle management and supply chain resilience guarantee. It not only requires data interconnection, intelligent warning, and decision support in terms of technology, but also has the ability to make some management functions or information transparent, providing value-added supply chain management services for spare parts procurement customers. Moreover, it is necessary to achieve collaborative operation of all links and full process transparency in the management mode to meet the high standards of the medical device industry for safe, timely, and reliable supply.

[0005] Therefore, a full-process digital collaborative management system and method for spare parts supply chain are proposed. Summary of the Invention

[0006] The purpose of the present invention is to provide a full-process digital collaborative management system and method for spare parts supply chain. By constructing a dynamic scoring model to evaluate supplier data, supplier scores are obtained; when the supplier score is lower than a predetermined threshold, a multi-source procurement plan is generated, and the procurement allocation is dynamically adjusted according to the supplier score; an intelligent verification model for inbound and outbound spare parts is constructed to analyze outbound and inbound spare parts, generate an inspection report and associate it with a blockchain record, and trigger an alarm and freeze the order when differences are found; a supply chain risk topology map is constructed based on supplier fulfillment rate, logistics data, and market demand fluctuations to identify and warn high-risk nodes in real time; based on the high-risk nodes, emergency procurement is initiated to suppliers with high supplier scores, and inventory scheduling is optimized based on geographical proximity; the inventory level is optimized in real time based on the production line status of outbound and inbound spare parts and in-transit logistics data, and the safety inventory threshold is dynamically adjusted to cope with supply chain interruptions.

[0007] To achieve the above purpose, the present invention provides the following technical solutions:

[0008] A full-process digital collaborative management system and method for spare parts supply chain, including:

[0009] A multi-dimensional dynamic scoring module that constructs a dynamic scoring model to evaluate supplier data, obtains supplier scores; when the supplier score is lower than a predetermined threshold, generates a multi-source procurement plan, and dynamically adjusts the procurement allocation according to the supplier score;

[0010] A full life cycle data chain module that constructs an intelligent verification model for inbound and outbound spare parts to analyze outbound and inbound spare parts, generates an inspection report and associates it with a blockchain record, and triggers an alarm and freezes the order when differences are found;

[0011] Supply chain risk assessment module, which constructs a supply chain risk topology map through supplier fulfillment rate, logistics data, and market demand fluctuations, and real-time identifies and warns high-risk nodes; initiates emergency procurement to suppliers with high scores based on high-risk nodes, and optimizes inventory scheduling based on geographical proximity;

[0012] Inventory optimization module, which real-time optimizes the inventory level based on the production line status of outgoing and incoming spare parts and in-transit logistics data, and dynamically adjusts the safety inventory threshold to cope with supply chain disruptions.

[0013] Preferably, the supplier data includes supplier historical performance data, real-time geographical risk data, and production capacity flexibility data; a unique digital identifier is assigned to the outgoing and incoming spare parts; the unique digital identifier includes the product batch number, production date, quality inspection information, and logistics tracking code.

[0014] Preferably, the dynamic scoring model includes a data input processing layer, a data feature extraction layer, a weight update layer, and a supplier score generation layer;

[0015] The data input processing layer preprocesses the supplier historical performance data, real-time geographical risk data, and production capacity flexibility data;

[0016] The data feature extraction layer analyzes the preprocessed supplier historical performance data, real-time geographical risk data, and production capacity flexibility data to obtain the on-time delivery rate trend, disaster probability, and redundancy index;

[0017] The weight update layer learns and generates the weights of the on-time delivery rate trend, disaster probability, and redundancy index based on the input of historical supplier data and real-time supplier data;

[0018] The supplier score generation layer generates a supplier score by performing a weighted combination calculation on the weights and features.

[0019] Preferably, the incoming and outgoing spare parts intelligent verification model includes an outgoing spare parts intelligent verification analysis layer and an incoming spare parts intelligent verification analysis layer;

[0020] The outgoing spare parts intelligent verification analysis layer identifies the identity information of the outgoing spare parts through RFID technology; verifies the physical characteristics of the outgoing spare parts through image recognition technology; compares the identified identity information and physical characteristics with the spare parts information in the sales order; when a mismatch is found during the comparison, an alarm is triggered and the relevant order is frozen;

[0021] The incoming spare parts intelligent verification analysis layer checks the appearance and quality of the incoming spare parts through AI vision technology; compares the inspection results with the spare parts specifications in the purchase order; automatically generates an inspection report based on the comparison results; associates the inspection report with the blockchain record.

[0022] Preferably, the specific steps for constructing the supply chain risk topology map are as follows:

[0023] By analyzing the supplier performance rate, logistics data, and market demand fluctuations, obtain the supplier performance rate, logistics delay data, and market demand fluctuation data;

[0024] Construct a supply chain map, where the nodes in the supply chain map represent suppliers, manufacturers, and logistics hubs, and the edges represent materials and / or information flows;

[0025] Input the supply chain map, supplier performance rate, logistics delay data, and market demand fluctuation data into a graph neural network, extract node features through the graph neural network, predict the risk probability of each node, obtain the nodes whose risk probability exceeds a predetermined threshold, and obtain high-risk nodes.

[0026] Preferably, the specific steps for obtaining the dynamically adjusted safety inventory threshold include:

[0027] Preprocess the production line status data and in-transit logistics data;

[0028] Through the input of historical inventory data, production plans, market demands, and real-time data, combine transfer learning to construct a demand forecasting model and generate a forecasting result; evaluate the current inventory level based on the forecasting result, calculate the difference between the inventory and future demands, automatically adjust the safety inventory threshold in inventory management based on the calculation result, and identify inventory risks; introduce a dynamic inventory optimization algorithm to obtain an inventory optimization result; generate a supply chain adjustment plan based on the inventory optimization result.

[0029] A full-process digital collaborative management method for spare parts supply chains includes:

[0030] Construct a dynamic scoring model to evaluate supplier data and obtain a supplier score; when the supplier score is lower than a predetermined threshold, generate a multi-source procurement plan and dynamically adjust the procurement allocation based on the supplier score;

[0031] Construct an intelligent verification model for in-and-out spare parts to analyze the out-of-warehouse spare parts and in-warehouse spare parts, generate an inspection report and associate it with a blockchain record, and trigger an alarm and freeze the order when a difference is found;

[0032] Construct a supply chain risk topology map through the supplier performance rate, logistics data, and market demand fluctuations, real-time identify and warn high-risk nodes; initiate emergency procurement from suppliers with high supplier scores based on the high-risk nodes, and optimize inventory scheduling based on geographical proximity;

[0033] Optimize the inventory level in real time based on the production line status of outgoing and incoming spare parts and in-transit logistics data, and dynamically adjust the safety inventory threshold to cope with supply chain disruptions.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] 1. The present invention adopts a multi-dimensional dynamic scoring module. By comprehensively preprocessing, feature extracting, and weight updating the historical performance of suppliers, real-time geographical risks, and production capacity elasticity data, it generates accurate supplier scores in real time. The present invention reduces the uncertainty risks caused by information lag and one-sided data by realizing the dynamic feedback of supplier risks and supply capabilities. When the score is lower than the preset threshold, a multi-source procurement plan is automatically generated, and the procurement allocation is intelligently adjusted according to the score to ensure the stable supply of critical spare parts and the optimization of procurement costs. This technology improves the transparency and response speed of the supply chain and promotes the intelligent upgrade of supply chain management.

[0036] 2. The full-life-cycle data chain module and spare part intelligent verification model constructed by the present invention use multiple identification technologies such as RFID, image recognition, and AI vision to automatically verify the identity and quality of incoming and outgoing spare parts, and accurately compare them with sales and procurement order data. The present invention can trace the spare part information throughout the process, timely discover the situation where the incoming and outgoing data does not match the order information, automatically trigger an alarm and freeze the relevant orders, effectively preventing incorrect transfer and quality risks. At the same time, the inspection report is associated with the blockchain record to achieve data anti-tampering and real-time sharing, significantly improving the transparency and trust of the supply chain and reducing the errors and delays caused by manual checking.

[0037] 3. The inventory optimization module of the present invention collects the production line status and in-transit logistics data in real time, combines the historical inventory data, production plan, and market demand information, and uses transfer learning to construct a demand prediction model to accurately evaluate the difference between the current inventory level and future demand. Based on this evaluation result, the safety inventory threshold in inventory management is automatically adjusted, and a dynamic inventory optimization algorithm is introduced to generate inventory optimization results and supply chain adjustment plans, thereby effectively avoiding supply chain disruption risks. The present invention has the advantages of real-time and intelligence, can quickly respond to changes in all links of the supply chain, dynamically balance inventory costs and supply security, and significantly improve operational efficiency and market competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic structural diagram of a full-process digital collaborative management system for a spare part supply chain provided by the present invention;

[0039] Figure 2 It is a schematic flow diagram of a full-process digital collaborative management method for a spare part supply chain provided by the present invention;

[0040] Figure 3 Schematic diagram of the dynamic scoring model structure provided by the embodiment of the present invention;

[0041] Figure 4 Schematic diagram of the full-process structure of the supply chain provided by the embodiment of the present invention. Detailed implementation manners

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0043] Embodiment 1:

[0044] Please refer to Figures 1 to 4 , the present invention provides a full-process digital collaborative management system for spare parts supply chain, which is applied to a full-process digital collaborative management method for spare parts supply chain. A full-process digital collaborative management system for spare parts supply chain includes a multi-dimensional dynamic scoring module, a full-life cycle data chain module, a supply chain risk assessment module, and an inventory optimization module. Specifically, refer to Figure 1 ; The corresponding method steps of the modules in a full-process digital collaborative management system for spare parts supply chain are referred to Figure 2 , and the specific technical solutions are as follows:

[0045] The multi-dimensional dynamic scoring module constructs a dynamic scoring model to evaluate supplier data and obtains supplier scores; when the supplier score is lower than a predetermined threshold, a multi-source procurement plan is generated, and the procurement allocation is dynamically adjusted according to the supplier score; the supplier data includes supplier historical performance data, real-time geographical risk data, and production capacity flexibility data;

[0046] Further, the dynamic scoring model includes a data input processing layer, a data feature extraction layer, a weight update layer, and a supplier score generation layer. Refer to Figure 3 ;

[0047] The data input processing layer preprocesses the supplier historical performance data, real-time geographical risk data, and production capacity flexibility data;

[0048] The data feature extraction layer analyzes the preprocessed supplier historical performance data, real-time geographical risk data, and production capacity flexibility data to obtain the on-time delivery rate trend, disaster probability, and redundancy index;

[0049] The weight update layer learns and generates the weights of the on-time delivery rate trend, disaster probability, and redundancy index based on the input of historical supplier data and real-time supplier data;

[0050] The supplier score generation layer generates a supplier score by performing a weighted combination calculation on weights and features.

[0051] In this embodiment, by constructing a multi-dimensional dynamic scoring module, the present invention can comprehensively evaluate the historical performance, real-time geographical risk, and production capacity flexibility of suppliers, and achieve accurate calculation of supplier scores. Through the collaborative action of the data input processing layer, data feature extraction layer, weight update layer, and supplier score generation layer, the dynamics and adaptability of the scores are ensured. Among them, the data feature extraction layer can mine key features from supplier data, such as the trend of on-time delivery rate, disaster probability, and redundancy index, making the supplier evaluation more comprehensive. The weight update layer performs adaptive learning by combining historical and real-time data, enabling the scoring results to be dynamically adjusted with changes in the supply chain environment, and enhancing the system's response ability to sudden risks. When the supplier score is lower than a predetermined threshold, the system can automatically generate a multi-source procurement plan and dynamically adjust the procurement allocation plan according to changes in the supplier score, thereby reducing the risk of supply chain interruption and improving the flexibility and reliability of procurement decisions.

[0052] The full-life-cycle data chain module constructs an intelligent verification model for in-out parts to analyze the out-bound parts and in-bound parts, generates an inspection report and associates it with the blockchain record, and triggers an alarm and freezes the order when differences are found; assigns a unique digital identifier to the out-bound parts and in-bound parts; the unique digital identifier includes the product batch number, production date, quality inspection information, and logistics tracking code.

[0053] Further, the intelligent verification model for in-out parts includes an intelligent verification analysis layer for out-bound parts and an intelligent verification analysis layer for in-bound parts.

[0054] The intelligent verification analysis layer for out-bound parts identifies the identity information of out-bound parts through RFID technology; verifies the physical characteristics of out-bound parts through image recognition technology; compares the identified identity information and physical characteristics with the part information in the sales order; when a mismatch is found during the comparison, triggers an alarm and freezes the relevant order.

[0055] The intelligent verification analysis layer for in-bound parts checks the appearance and quality of in-bound parts through AI vision technology; compares the inspection results with the part specifications in the purchase order; automatically generates an inspection report based on the comparison results; associates the inspection report with the blockchain record.

[0056] In this embodiment, the present invention realizes the intelligent verification of incoming and outgoing spare parts through the construction of a full-life-cycle data chain module, improving the accuracy and security of spare parts management. By assigning unique digital identifiers to outgoing and incoming spare parts and combining RFID technology, image recognition technology, and AI vision technology, the full-process traceability of spare parts information can be ensured. The intelligent verification and analysis layer for outgoing spare parts can effectively identify the identity information of outgoing spare parts and ensure its consistency with the sales order through physical feature comparison, avoiding risks of misdelivery, missing delivery, or forgery; the intelligent verification and analysis layer for incoming spare parts detects the appearance and quality of spare parts through AI vision technology and automatically generates an acceptance report, which is associated with the blockchain record to ensure the credibility and immutability of the acceptance data. When the system detects differences in incoming and outgoing spare parts, it can promptly trigger an alarm and freeze the relevant order, thereby quickly discovering and preventing abnormal circulation and reducing losses and risks in supply chain management. This method can significantly improve the intelligent level of spare parts circulation and optimize the inventory management process. By combining RFID and AI vision technology with blockchain records, the present invention reduces the incoming batch error rate from 8.2% to 0.5% and improves the efficiency of acceptance report generation and freeze response by 95.8%, effectively solving the problems of low efficiency and high error rate in traditional manual checking. Specific data can be referred to in Table 1.

[0057] Table 1 Effect of Incoming and Outgoing Intelligent Verification Technology on Reducing Error Rate

[0058] Technical indicators Traditional manual verification The intelligent verification system of the present invention Reduction rate of error rate Batch matching error rate 8.2％ 2.3% (RFID + AI vision) -72.0％ Quality missed inspection rate 12％ 2.7% (blockchain-related acceptance) -77.5％ Order freeze response time Manual processing (2 hours) Automatically trigger (<20 minutes) -83.3％

[0059] The supply chain risk assessment module constructs a supply chain risk topology map through the supplier performance rate, logistics data, and market demand fluctuations, and real-time identifies and warns high-risk nodes; based on the high-risk nodes, it initiates emergency procurement from suppliers with high scores and optimizes inventory scheduling based on geographical proximity; at the same time, it uses a combination of transfer learning and dynamic clustering to predict the demand for spare parts with intermittent demand patterns;

[0060] Further, the specific steps for constructing the supply chain risk topology map are as follows:

[0061] By analyzing the supplier performance rate, logistics data, and market demand fluctuations, obtain the supplier performance rate, logistics delay data, and market demand fluctuation data;

[0062] Construct a supply chain map, where the nodes in the supply chain map represent suppliers, manufacturers, and logistics hubs, and the edges represent materials and / or information flows;

[0063] Input the supply chain map, supplier performance rate, logistics delay data, and market demand fluctuation data into a graph neural network, extract node features through the graph neural network, predict the risk probability of each node, and obtain the nodes with a risk probability exceeding a predetermined threshold to obtain high-risk nodes.

[0064] In this embodiment, through the construction of the supply chain risk assessment module, the present invention realizes the intelligent identification and dynamic early warning of the overall supply chain risk, improving the security and response ability of supply chain management. Based on the supplier compliance rate, logistics data, and market demand fluctuations, the system can construct a supply chain risk topology map and use graph neural networks to extract features and predict risks for suppliers, manufacturers, and logistics hubs in the supply chain, thereby identifying high-risk nodes in real time. When the risk probability of a high-risk node exceeds a predetermined threshold, the system can automatically trigger an early warning mechanism and take targeted countermeasures, such as initiating emergency procurement from suppliers with higher scores, to reduce the risk of supply chain disruption. At the same time, optimize inventory scheduling based on geographical proximity to ensure the efficiency of spare parts supply. In addition, a method combining transfer learning and dynamic clustering is used to predict the demand for spare parts with intermittent demand patterns, enabling the supply chain to more accurately adapt to market fluctuations and improving the flexibility and accuracy of inventory management. This method not only enhances the stability and risk resistance of the supply chain but also optimizes the allocation of supply chain resources and improves the overall operational efficiency.

[0065] The inventory optimization module optimizes the inventory level in real time based on the production line status of outgoing and incoming spare parts and in-transit logistics data, and dynamically adjusts the safety inventory threshold to cope with supply chain disruptions.

[0066] The specific steps for obtaining the dynamically adjusted safety inventory threshold include:

[0067] Collect production line status data and in-transit logistics data and perform preprocessing;

[0068] Through historical inventory data, production plans, market demand, and real-time data input, combine transfer learning to construct a demand prediction model and generate prediction results; evaluate the current inventory level based on the prediction results, calculate the difference between inventory and future demand, automatically adjust the safety inventory threshold in inventory management based on the calculation results, and identify inventory risks; introduce a dynamic inventory optimization algorithm to obtain inventory optimization results; generate a supply chain adjustment plan based on the inventory optimization results.

[0069] In this embodiment, the present invention realizes the intelligentization and dynamic optimization of inventory management through the construction of an inventory optimization module, improving the stability and resilience of the supply chain. Based on the real-time monitoring of production line status and in-transit logistics data, the system can dynamically adjust the safety inventory threshold to effectively cope with the risk of supply chain interruption. By comprehensively analyzing production line status data, in-transit logistics data, and historical inventory data, and combining transfer learning to construct a demand prediction model, it can accurately predict future inventory demand and ensure that the inventory level matches the production demand. The system can evaluate the inventory level based on the prediction results, calculate the difference between the inventory and future demand, and automatically adjust the safety inventory threshold accordingly to identify potential inventory risks in advance. At the same time, a dynamic inventory optimization algorithm is introduced to optimize the inventory management strategy and ensure the rationality of inventory allocation. Finally, the system generates a supply chain adjustment plan based on the optimization results, realizes the precise allocation of inventory resources, reduces the risks of inventory backlog and shortage, and improves the overall operation efficiency of the supply chain.

[0070] The present invention realizes the intelligentization and refinement of supply chain management by constructing a multi-dimensional dynamic scoring, full-life cycle data chain, supply chain risk assessment, and inventory optimization module, improving the stability, flexibility, and risk resistance of the supply chain. Among them, the multi-dimensional dynamic scoring module comprehensively evaluates the historical performance, geographical risk, and production capacity elasticity of suppliers to ensure the accuracy and adaptability of supplier scoring, and dynamically adjusts the procurement allocation plan based on the scoring, thereby reducing the risk of supply chain interruption. The full-life cycle data chain module uses technologies such as unique digital identification, RFID, and AI vision to intelligently verify the incoming and outgoing spare parts, ensuring the authenticity and immutability of the data, and improving the safety and transparency of spare parts management. The supply chain risk assessment module constructs a supply chain risk topology map based on graph neural networks, real-time identifies high-risk nodes, and enhances the resilience of the supply chain through emergency procurement and optimized inventory scheduling. The inventory optimization module uses production line status and in-transit logistics data, combines transfer learning to predict demand, and dynamically adjusts the safety inventory threshold to optimize inventory management and reduce the risk of supply chain interruption. For details, refer to Figure 4 .

[0071] Embodiment 2:

[0072] Multi-dimensional dynamic scoring module, which constructs a dynamic scoring model to evaluate supplier data and obtain supplier scores; when the supplier score is lower than a predetermined threshold, a multi-source procurement plan is generated, and the procurement allocation is dynamically adjusted according to the supplier score; the supplier data includes supplier historical performance data, real-time geographical risk data, and production capacity flexibility data; the historical performance data includes the past on-time delivery rate and product quality qualification rate of the supplier, which are obtained through the enterprise resource planning system or historical order records; the real-time geographical risk data includes the natural disaster risk and political instability index of the supplier's location, which are collected through a third-party geographic information system (GIS) or real-time news data interface; the production capacity flexibility data includes the production capacity utilization rate and spare production capacity of the supplier, which are obtained through the production reports provided by the supplier or real-time monitoring of Internet of Things devices.

[0073] Further, the dynamic scoring model includes a data input processing layer, a data feature extraction layer, a weight update layer, and a supplier score generation layer;

[0074] The data input processing layer preprocesses the supplier historical performance data, real-time geographical risk data, and production capacity flexibility data. The preprocessing includes data cleaning, outlier removal, and normalization;

[0075] The data feature extraction layer analyzes the preprocessed supplier historical performance data, real-time geographical risk data, and production capacity flexibility data to obtain the on-time delivery rate trend, disaster probability, and redundancy index; the on-time delivery rate trend calculates the change trend of the on-time delivery rate in the recent 6 months by using time series analysis; the disaster probability estimates the disaster probability in the next 30 days based on historical disaster data through Poisson distribution; the redundancy index calculates the redundancy according to the production capacity utilization rate and spare production capacity, and the redundancy = spare production capacity / (total production capacity × utilization rate).

[0076] The weight update layer, based on the input of historical supplier data and real-time supplier data, uses machine learning algorithms to learn and generate the weights of the on-time delivery rate trend, disaster probability, and redundancy index;

[0077] The supplier score generation layer generates a supplier score by calculating the weighted combination of weights and features;

[0078] The specific calculation formula for the supplier score is:

[0079] S = ω1T + ω2(1 - P) + ω3R;

[0080] where S is the supplier score, ω1 is the weight of the on-time delivery rate trend, T is the on-time delivery rate trend, ω2 is the weight of the geographical safety factor, P is the disaster probability, ω3 is the weight of the redundancy index, and R is the redundancy index.

[0081] The full life cycle data chain module builds an intelligent verification model for inbound and outbound spare parts to analyze outbound spare parts and inbound spare parts, generate acceptance reports and link them with blockchain records, and trigger alarms and freeze orders when differences are found; assign unique digital identifiers to the outbound spare parts and inbound spare parts; the unique digital identifiers include product batch number, production date, quality inspection information and logistics tracking code;

[0082] Furthermore, the in-and-out parts intelligent verification model includes an out-and-out parts intelligent verification analysis layer and an in-and-out parts intelligent verification analysis layer;

[0083] The intelligent verification and analysis layer for outbound spare parts uses RFID technology to identify the identity information of outbound spare parts; uses image recognition technology to verify the physical characteristics of outbound spare parts, which include size characteristics, color characteristics and quantity; compares the identified identity information and physical characteristics with the spare parts information in the sales order; and triggers an alarm and freezes the relevant order when the comparison finds a discrepancy.

[0084] The outbound management process includes sales orders, outbound pre-entry, shelf management, warehouse packaging, waybill generation and receipt, ensuring that spare parts are delivered to customers from the warehouse correctly according to the requirements of the sales order. Through digital and intelligent technologies, such as RFID reading identity information and image recognition to verify physical characteristics, the system can accurately match sales orders with outbound spare parts to avoid delivery errors; at the same time, it records the outbound status in real time, such as packaging information and waybill number, for easy tracking; and when differences are found (such as inconsistent spare parts size or batch), alarms are triggered in time to prevent the flow of problematic spare parts.

[0085] The intelligent verification and analysis layer for incoming spare parts uses AI visual technology to check the appearance and quality of incoming spare parts; compares the inspection results with the spare parts specifications in the purchase order; automatically generates an acceptance report based on the comparison results; and associates the acceptance report with the blockchain record.

[0086] The warehousing management process includes purchase orders, inventory, inventory pre-shelf, inventory verification, pre-arrival notification, warehousing pre-entry, acceptance, warehouse shelving and coordination to ensure that the purchased parts meet the quality standards and are put on the shelves in time to prepare for subsequent production or sales. Through AI visual technology and data comparison, the system can check the appearance and quality of incoming parts, such as surface scratches and specification deviations, to ensure compliance; check the quantity and specifications with the purchase order to reduce errors; automatically generate acceptance reports and record them through blockchain to ensure data transparency and immutability.

[0087] Supply chain risk assessment module, which constructs a supply chain risk topology map through supplier fulfillment rate, logistics data, and market demand fluctuations, and real-time identifies and warns high-risk nodes; initiates emergency procurement to suppliers with high scores based on high-risk nodes, and optimizes inventory scheduling based on geographical proximity; at the same time, uses transfer learning and dynamic clustering to combine and predict the demand for spare parts with intermittent demand patterns;

[0088] Further, the specific steps for constructing the supply chain risk topology map are as follows:

[0089] By analyzing the supplier fulfillment rate, logistics data, and market demand fluctuations, obtain the supplier fulfillment rate, logistics delay data, and market demand fluctuation data;

[0090] Construct a supply chain graph, where the nodes in the supply chain graph represent suppliers, manufacturers, and logistics hubs, and the edges represent material and / or information flows; use nodes to represent suppliers (such as S1, S2), manufacturers (such as M1), and logistics hubs (such as L1), and edges to represent material flows (such as S1→M1) or information flows (such as M1→L1);

[0091] Input the supply chain graph, supplier fulfillment rate, logistics delay data, and market demand fluctuation data into a graph neural network, extract node features through the graph neural network, predict the risk probability of each node, obtain the nodes whose risk probability exceeds a predetermined threshold, and obtain high-risk nodes; input the supply chain graph and data into a graph neural network (GNN), extract node features (such as low fulfillment rate, high delay), and predict the risk probability; if the risk probability of node S1 is 0.85 (the threshold is set to 0.7 in this embodiment), then it is identified as a high-risk node.

[0092] The calculation formula for the node risk probability is:

[0093] P risk (i) = σ(β1(1 - F i ) + β2L i + β3M i + b);

[0094] Where, P risk (i) is the risk probability of node i, β1 is the weight of the supplier fulfillment rate, F i is the supplier fulfillment rate corresponding to node i (the lower the fulfillment rate, the higher the risk), β2 is the weight of the logistics delay risk factor, L i is the logistics delay risk factor corresponding to node i (the higher the delay, the greater the risk), β3 is the weight of the market demand fluctuation risk factor, M i is the market demand fluctuation risk factor corresponding to node i (the greater the fluctuation, the higher the risk), b is the bias term, and σ() is the Sigmoid function, which is used to map the weighted sum to the 0-1 interval.

[0095] As described in the supply chain risk assessment module, through the topological graph analysis based on the Graph Neural Network (GNN), the risks of global supply chain nodes can be identified in real time. Compared with the expert system, the accuracy of risk prediction has increased by 35.3%, and the response speed has been leapfrogged from 48 hours to real-time. For specific data, please refer to Table 2.

[0096] Table 2 Comparison of Response Times between the Supply Chain Risk Topological Graph and Traditional Risk Assessment Methods

[0097] Evaluation dimension Expert risk assessment system Graph neural network model Efficiency improvement Risk identification delay 48 hours Real-time dynamic warning +100％ Depth of node association analysis Local node analysis Global topological relationship mining +80％ Accuracy of high-risk prediction 68％ 92% (GNN feature extraction) +35.3％

[0098] The inventory optimization module can optimize the inventory level in real time based on the production line status of outgoing and incoming spare parts and the in-transit logistics data, and dynamically adjust the safety inventory threshold to cope with supply chain disruptions.

[0099] The specific steps for obtaining the dynamically adjusted safety inventory threshold include:

[0100] Collect the production line status data and in-transit logistics data and perform preprocessing; the production line status data is collected by gathering the operating status, production efficiency, and downtime of production equipment, and the preprocessing steps include data cleaning, outlier detection, and standardization; the in-transit logistics data is obtained by acquiring the location of the goods in transit, the estimated arrival time, and the transportation delay;

[0101] Through the historical inventory data, production plan, market demand, and real-time data input, combined with transfer learning to build a demand prediction model and generate prediction results; evaluate the current inventory level based on the prediction results, calculate the difference between the inventory and future demand, automatically adjust the safety inventory threshold in inventory management based on the calculation results, and identify inventory risks; introduce a dynamic inventory optimization algorithm to obtain inventory optimization results; generate a supply chain adjustment plan based on the inventory optimization results;

[0102] Model construction idea: Use the historical inventory data, production plan, market demand data, and real-time updated production line and logistics data as inputs; adopt transfer learning technology to quickly adapt to the data characteristics in the current environment based on the existing prediction model in the source domain (such as under a similar production environment or market background), and improve the prediction accuracy.

[0103] Generation of prediction results: Through the trained demand prediction model, predict the demand volume, demand fluctuation trend, etc. within a future period of time to form a quantitative demand prediction result.

[0104] Inventory assessment: Compare the real-time inventory level with the predicted demand, calculate the difference between the inventory and future demand (i.e., inventory gap or overstock situation); conduct a preliminary assessment of inventory risks and identify potential inventory shortage or overstock risks.

[0105] Dynamic adjustment of safety stock threshold: Automatically update the safety stock threshold based on the calculated inventory gap.

[0106] Optimization objective: The goal is to balance inventory costs and inventory risks, ensuring that production interruptions do not occur due to insufficient inventory, nor will warehousing costs increase due to inventory backlogs.

[0107] Algorithm implementation: Introduce a dynamic inventory optimization algorithm that comprehensively considers the current inventory status, safety stock threshold, in-transit logistics situation, and market demand to generate an inventory optimization plan;

[0108] The algorithm output includes key indicators such as the optimal inventory level, reorder point, and replenishment quantity recommendations, guiding the adjustment of actual inventory management and procurement plans.

[0109] Plan formulation: Based on the inventory optimization results, formulate a specific supply chain adjustment plan, which may cover aspects such as procurement strategies, production scheduling, and logistics distribution optimization. The goal of the plan is to quickly respond to demand forecasting results and inventory assessments, and timely adjust procurement and production plans to ensure the stability and efficiency of the overall supply chain operation.

[0110] Risk monitoring and feedback: After the plan is formulated, a real-time monitoring mechanism needs to be established to track inventory changes and market dynamics, and timely feedback and optimize the dynamic adjustment model of the inventory safety threshold.

[0111] As described in the inventory optimization module, transfer learning combined with dynamic clustering significantly improves the intermittent demand forecasting accuracy to 85%, and the training cycle is shortened by 78.6%, solving the limitations of the traditional ARIMA model in complex demand patterns. For details, refer to Table 3.

[0112] Table 3 Comparison of accuracy improvement of transfer learning in demand forecasting

[0113] Prediction scenario Traditional ARIMA model Transfer learning + dynamic clustering Accuracy improvement Steady demand prediction 89％ 93％ +4.5％ Intermittent demand prediction 52％ 85% (dynamic clustering adaptation) +63.5％

[0114] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A digital collaborative management system for the entire process of a spare parts supply chain, characterized in that, Including: A multi-dimensional dynamic scoring module that constructs a dynamic scoring model to evaluate supplier data and obtains supplier scores; When the supplier score is lower than a predetermined threshold, a multi-source procurement plan is generated, and the procurement allocation is dynamically adjusted based on the supplier score; A full-life-cycle data chain module that constructs an intelligent verification model for outgoing and incoming spare parts to analyze outgoing and incoming spare parts, generates an inspection report and associates it with a blockchain record, and triggers an alarm and freezes the order when a difference is found; A supply chain risk assessment module that constructs a supply chain risk topology map through supplier performance rates, logistics data, and market demand fluctuations, and real-time identifies and warns high-risk nodes; Initiate emergency procurement from suppliers with high supplier scores based on high-risk nodes, and optimize inventory scheduling based on geographical proximity; An inventory optimization module that real-time optimizes the inventory level based on the production line status of outgoing and incoming spare parts and in-transit logistics data, and dynamically adjusts the safety inventory threshold to cope with supply chain disruptions.

2. The full-process digital collaborative management system for a spare parts supply chain according to claim 1, wherein: The supplier data includes supplier historical performance data, real-time geographical risk data, and production capacity flexibility data; a unique digital identifier is assigned to the outgoing and incoming spare parts; the unique digital identifier includes a product batch number, a production date, quality inspection information, and a logistics tracking code.

3. The full-process digital collaborative management system for a spare parts supply chain according to claim 1, wherein: The dynamic scoring model includes a data input processing layer, a data feature extraction layer, a weight update layer, and a supplier score generation layer; The data input processing layer preprocesses the supplier historical performance data, real-time geographical risk data, and production capacity flexibility data; The data feature extraction layer obtains the on-time delivery rate trend, disaster probability, and redundancy index through the analysis of the preprocessed supplier historical performance data, real-time geographical risk data, and production capacity flexibility data; The weight update layer learns and generates the weights of the on-time delivery rate trend, disaster probability, and redundancy index based on the input of historical supplier data and real-time supplier data; The supplier score generation layer generates a supplier score through weighted combination calculation of the weights and features.

4. The full-process digital collaborative management system for a spare parts supply chain according to claim 1, wherein: The intelligent verification model for outgoing and incoming spare parts includes an intelligent verification analysis layer for outgoing spare parts and an intelligent verification analysis layer for incoming spare parts; The intelligent verification analysis layer for outgoing spare parts identifies the identity information of outgoing spare parts through RFID technology; verifies the physical characteristics of outgoing spare parts through image recognition technology; Compares the identified identity information and physical characteristics with the spare parts information in the sales order; when a mismatch is found during the comparison, triggers an alarm and freezes the relevant order; The intelligent verification analysis layer for incoming spare parts checks the appearance and quality of incoming spare parts through AI vision technology; compares the inspection results with the spare parts specifications in the purchase order; automatically generates an inspection report based on the comparison results; associates the inspection report with the blockchain record.

5. The full-process digital collaborative management system for a spare parts supply chain according to claim 1, wherein: The specific steps for constructing the supply chain risk topology map are as follows: By analyzing the supplier performance rate, logistics data, and market demand fluctuations, obtain the supplier performance rate, logistics delay data, and market demand fluctuation data; Construct a supply chain map, where the nodes in the supply chain map represent suppliers, manufacturers, and logistics hubs, and the edges represent material and / or information flows; Input the supply chain map, supplier performance rate, logistics delay data, and market demand fluctuation data into a graph neural network. Extract node features through the graph neural network, predict the risk probability of each node, obtain the nodes whose risk probability exceeds a predetermined threshold, and obtain high-risk nodes.

6. The full-process digital collaborative management system for a spare parts supply chain according to claim 1, wherein: The specific steps for obtaining the dynamically adjusted safety inventory threshold include: Preprocess the production line status data and in-transit logistics data; Input historical inventory data, production plans, market demands, and real-time data, combine transfer learning to construct a demand prediction model, and generate a prediction result; evaluate the current inventory level based on the prediction result, calculate the difference between the inventory and future demands, automatically adjust the safety inventory threshold in inventory management based on the calculation result, and identify inventory risks; introduce a dynamic inventory optimization algorithm to obtain an inventory optimization result; generate a supply chain adjustment plan based on the inventory optimization result.

7. A digital collaborative management method for the entire process of a spare parts supply chain, characterized in that, Including: Construct a dynamic scoring model to evaluate supplier data and obtain a supplier score; When the supplier score is lower than a predetermined threshold, generate a multi-source procurement plan and dynamically adjust the procurement allocation according to the supplier score; Construct an intelligent verification model for in-out spare parts to analyze the out-bound spare parts and in-bound spare parts, generate an inspection report and associate it with a blockchain record, and trigger an alarm and freeze the order when differences are found; Construct a supply chain risk topology map through the supplier performance rate, logistics data, and market demand fluctuations, and real-time identify and warn high-risk nodes; Initiate an emergency procurement to suppliers with high scores based on the high-risk nodes, and optimize inventory scheduling based on geographical proximity; Based on the production line status and in-transit logistics data of the out-bound spare parts and in-bound spare parts, real-time optimize the inventory level, and dynamically adjust the safety inventory threshold to cope with supply chain disruptions.

Images