Electronic component supply chain intelligent management method and system

Abstract

The invention discloses an electronic component supply chain intelligent management method and system, and relates to the technical field of intelligent supply chain management, and the method comprises the steps: collecting the real-time market transaction data of electronic components, and the real-time market transaction data comprises historical purchase data, historical inventory data, supplier delivery cycle data and component life cycle data; constructing a demand prediction model of the electronic component, and generating a demand prediction result; inputting the demand prediction result into a risk assessment model, and outputting a stockout risk index through a fuzzy comprehensive evaluation method; and according to the stockout risk index, carrying out risk grade division on the electronic component, generating a supplier purchase quota scheme, and realizing intelligent management of the electronic component supply chain. According to the invention, the prediction accuracy and anti-risk capability of the supply chain are effectively improved, scientific decision and intelligent management of electronic component purchase are realized, and the operation cost and supply interruption risk caused by stockout are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent supply chain management, and in particular to an intelligent management method and system for an electronic component supply chain. Background Art

[0002] Electronic component supply chain management plays a vital role in modern manufacturing, involving the coordination and optimization of the entire process from raw material procurement and inventory scheduling to terminal delivery. Traditional supply chain management methods often rely on manual experience or rule-driven management systems. When dealing with scenarios such as large demand fluctuations, a wide variety of component categories, and sensitive delivery cycles, they are prone to problems such as low information processing efficiency, delayed supply and demand matching, and insufficient forecasting capabilities. Especially in the context of the massive emergence of multi-source heterogeneous data, how to achieve efficient abstract expression and intelligent analysis and processing of electronic component supply and demand data has become a key challenge to improving management efficiency and responsiveness. Therefore, there is an urgent need for a data-driven approach to improve the intelligence level of the electronic component supply chain and achieve accurate matching of supply and demand information and trend forecasting.

[0003] CN119863087A discloses a supply and demand management method for an electronic component supply chain, comprising: S1, abstractly labeling the acquired electronic data to convert the original complex data into a unified format; S2, selecting a large language model (LLM), inputting the labeled data into the model for analysis, and outputting an explanation of the supply and demand situation and a forecast of future supply and demand trends; S3, performing personalized training and optimization on the LLM model through a fine-tuning module to enhance the model's adaptability and predictive capabilities in specific supply chain scenarios; S4, after the system is built, vectorizing the stored labeled supply and demand data and material information, and utilizing vector search technology to achieve rapid matching and querying of supply and demand data. By introducing abstract label regularization technology, this method improves the uniformity of data processing and the efficiency of intelligent analysis, thereby enhancing the accuracy and response speed of overall supply and demand management. However, this method still has the following shortcomings: it mainly focuses on data processing and the forecasting model itself, without considering the upstream and downstream feedback mechanisms and dynamic adjustment mechanisms involved in actual supply chain operations. It is difficult to conduct closed-loop management of real-time changing market demand and multi-stage logistics constraints, resulting in insufficient response capabilities in emergencies or atypical demand scenarios. Summary of the Invention

[0004] In view of the problems that the existing electronic component supply chain management methods lack the ability to respond to real-time market fluctuations, the risk assessment mechanism is imperfect, the supplier quota allocation lacks intelligent optimization, and the demand forecasting model has low accuracy, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to improve the accuracy and response efficiency of risk perception in the electronic component supply chain, and realize intelligent management of the entire process from demand forecasting, risk assessment to supplier quota formulation and dynamic adjustment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides an electronic component supply chain intelligent management method, which includes:

[0008] Collecting real-time market transaction data of electronic components, wherein the real-time market transaction data includes historical procurement data, historical inventory data, supplier delivery cycle data, and component life cycle data;

[0009] Building a demand forecasting model for electronic components based on the historical procurement data and the historical inventory data, and generating demand forecasting results;

[0010] Based on the supplier delivery cycle data and the component life cycle data, the demand forecast result is input into a risk assessment model, and a stock-out risk index is output through a fuzzy comprehensive evaluation method;

[0011] The electronic components are classified into risk levels according to the out-of-stock risk index, and a supplier procurement quota plan is generated to achieve intelligent management of the electronic component supply chain.

[0012] As a preferred solution of the electronic component supply chain intelligent management method of the present invention, the electronic components are classified into risk levels according to the out-of-stock risk index, and a supplier procurement quota plan is generated, including:

[0013] Based on the out-of-stock risk index, electronic components are divided into high-risk, medium-risk and low-risk areas;

[0014] Establish supplier qualification rating standards for electronic components in high-risk areas, including financial status scores, production capacity scale scores, historical credit scores, and capacity utilization rates;

[0015] Based on the supplier qualification rating standards, suppliers are ranked and high-quality suppliers are selected;

[0016] A genetic algorithm is used to perform multi-objective optimization on the purchase quota of the high-quality supplier to generate a purchase quota plan Y;

[0017] Generate a procurement quota plan X for electronic components in medium-risk and low-risk areas according to the preset strategy;

[0018] Based on the procurement quota plan Y and the procurement quota plan X, a supplier procurement benchmark quota is set, and a dynamic adjustment mechanism is established. Through the periodic risk index assessment results, the quota allocation is corrected in real time to achieve intelligent management of the electronic component supply chain.

[0019] As a preferred solution of the electronic component supply chain intelligent management method of the present invention, wherein:

[0020] Electronic components are classified into risk levels, including:

[0021] When the out-of-stock risk index of an electronic component exceeds a first threshold and the capacity utilization rate exceeds a second threshold, the electronic component is placed on a temporary watch list. Dynamic monitoring of key indicators of the electronic components on the temporary watch list is conducted over three consecutive monitoring cycles, where the key indicators include supplier quotation fluctuations, supplier inventory, and market demand growth rate.

[0022] If the supplier's quotation fluctuation exceeds the third threshold, or the supplier's inventory is lower than the fourth threshold and the market demand growth rate exceeds the fifth threshold, the electronic component will be classified as a high-risk area;

[0023] If the number of alternative suppliers of electronic components in the high-risk area is higher than the sixth threshold and the supplier concentration index is lower than the seventh threshold, a risk downgrade assessment is triggered;

[0024] When the market supply-demand ratio is greater than the eighth threshold during the downgrade assessment period and there are no abnormal fluctuations during the continuous monitoring period, the electronic component will be adjusted to the medium-risk zone;

[0025] If the supplier fulfillment rate of electronic components in the medium-risk zone exceeds the ninth threshold and the inventory turnover rate is lower than the tenth threshold, the electronic components will be classified as low-risk areas;

[0026] If a supplier's products enter a high-risk area, the risk level of the associated products will automatically be raised by one level.

[0027] As a preferred solution of the electronic component supply chain intelligent management method of the present invention, the method for obtaining the out-of-stock risk index is:

[0028] Establish a risk assessment model based on deep neural network architecture;

[0029] Inputting historical out-of-stock event data into the risk assessment model for training, and optimizing the model parameters through a back-propagation algorithm;

[0030] Construct an evaluation indicator matrix, and standardize the delivery on-time rate, average delay time, and delivery volatility in the supplier's delivery cycle data to form a delivery evaluation vector;

[0031] Extract life cycle stage indicators, alternative product indicators and discontinuation warning indicators from component life cycle data to generate life cycle assessment vectors;

[0032] Calculate demand growth index, fluctuation index and forecast accuracy index based on demand forecast results to form a demand assessment vector;

[0033] Inputting the delivery assessment vector, the life cycle assessment vector, and the demand assessment vector into the risk assessment model to obtain an initial risk score;

[0034] Establishing a fuzzy membership function, mapping the initial risk score to the interval [0,1], and constructing a fuzzy relationship matrix;

[0035] Calculate the evaluation index weight vector using the hierarchical analysis method, perform fuzzy compound operation on the weight vector and the fuzzy relationship matrix, and output the out-of-stock risk index;

[0036] When the evaluation index fluctuates abnormally, the risk assessment model automatically adjusts the risk score weight and recalculates the out-of-stock risk index.

[0037] As a preferred solution of the electronic component supply chain intelligent management method of the present invention, the risk assessment model is constructed as follows:

[0038] Building the deep neural network architecture, wherein the deep neural network architecture includes an input layer, a feature extraction layer, a risk quantification layer, and an output layer; the feature extraction layer is provided with a plurality of convolutional layers and pooling layers alternating in structure;

[0039] A first neuron group is set in the input layer, wherein the first neuron group corresponds to the input interface of supplier delivery cycle characteristics, component life cycle characteristics and demand forecast characteristics;

[0040] Setting a sliding convolution kernel in the convolution layer of the feature extraction layer, wherein the sliding convolution kernel performs a nonlinear transformation on the input features to extract multi-scale feature information;

[0041] A maximum pooling operation is adopted in the pooling layer of the feature extraction layer to reduce the feature dimension and retain the features;

[0042] Setting a second neuron group in the risk quantification layer, wherein the second neuron group combines and weights the extracted feature information through a fully connected layer;

[0043] A third neuron group is set in the output layer, wherein the third neuron group uses a Softmax activation function to output the probability of the risk level;

[0044] Setting a batch normalization layer to standardize the outputs of the feature extraction layer and the risk quantification layer;

[0045] A residual connection structure is adopted to establish skip connections between the convolution blocks of the feature extraction layer to alleviate the gradient vanishing problem and complete the establishment of the risk assessment model.

[0046] As a preferred solution of the electronic component supply chain intelligent management method of the present invention, the method for generating the demand forecast result is:

[0047] The sequence segmentation method is used to divide the historical purchase data and historical inventory data in the real-time market transaction data into training set, validation set and test set according to the preset ratio;

[0048] Build a long short-term memory network model to capture the long-term dependency of electronic component procurement through a gating mechanism;

[0049] Construct a gradient boosting tree model, use the decision tree ensemble method to model the nonlinear demand pattern of electronic components, and improve the prediction accuracy by iteratively fitting the residuals;

[0050] Using a stacking integration method to weightedly fuse the outputs of the long short-term memory network model and the gradient boosting tree model to form a demand forecasting model;

[0051] Automatically tune the hyperparameters of the demand forecasting model based on a Bayesian optimization algorithm, and optimize the model performance on a validation set through cross-validation;

[0052] The test set is input into the optimized demand forecasting model, and combined with the safety stock level and market fluctuation factors, the demand forecast results of electronic components are generated.

[0053] As a preferred solution of the electronic component supply chain intelligent management method of the present invention, the method for collecting real-time market transaction data is:

[0054] Connect multiple electronic component trading platforms through the market data collection interface, establish data collection channels, and collect real-time market transaction data of electronic components;

[0055] Classify and store the real-time market transaction data according to data type to form historical procurement data, historical inventory data, supplier delivery cycle data, and component life cycle data;

[0056] A distributed data processing framework is used to clean the real-time market transaction data, and a time series feature extraction module is established to perform time series decomposition on the historical procurement data and the historical inventory data to extract trend items, periodic items, and random items;

[0057] Construct a supplier profile model to quantify the delivery time, delivery accuracy, and delivery quality indicators in the supplier's delivery cycle data;

[0058] Setting a life cycle status label for the component life cycle data, wherein the life cycle status label includes a mass production period, a mature period, a decline period, and a production suspension warning period;

[0059] The processed real-time market transaction data is stored in a distributed database and an index structure is established based on timestamps.

[0060] In a second aspect, an embodiment of the present invention provides an electronic component supply chain intelligent management system, which includes:

[0061] A collection module is used to collect real-time market transaction data of electronic components, wherein the real-time market transaction data includes historical procurement data, historical inventory data, supplier delivery cycle data and component life cycle data;

[0062] A demand forecasting module, which constructs a demand forecasting model for electronic components based on the historical procurement data and the historical inventory data, and generates a demand forecasting result;

[0063] a stock-out risk assessment module, which inputs the demand forecast result into a risk assessment model based on the supplier delivery cycle data and the component life cycle data, and outputs a stock-out risk index through a fuzzy comprehensive evaluation method;

[0064] The classification management module is used to classify the risk levels of electronic components according to the out-of-stock risk index, generate supplier procurement quota plans, and realize intelligent management of the electronic component supply chain.

[0065] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the electronic component supply chain intelligent management method as described in the first aspect of the present invention are implemented.

[0066] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the electronic component supply chain intelligent management method as described in the first aspect of the present invention are implemented.

[0067] Compared with the existing technology, the beneficial effects of the present invention are as follows: by collecting and integrating real-time market transaction data from multiple dimensions such as historical procurement, inventory, supplier delivery cycle and life cycle, an accurate demand forecasting model and risk assessment model are constructed, thereby realizing quantitative analysis and dynamic early warning of the risk of shortage of electronic components; the shortage risk index is output through the fuzzy comprehensive evaluation method, and electronic components are divided into high, medium and low risk levels accordingly, which not only strengthens the key management and control capabilities of key components, but also improves the overall refinement of resource allocation; a supplier qualification rating system is introduced in high-risk areas, and a genetic algorithm is used for multi-objective optimization allocation, which realizes the optimal configuration of high-risk component procurement quotas and improves supply continuity and stability; this method effectively improves the forecast accuracy and risk resistance of the supply chain, realizes scientific decision-making and intelligent management of electronic component procurement, and significantly reduces operating costs and supply interruption risks caused by shortages. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0069] Figure 1 This is a flow chart of the electronic component supply chain intelligent management method according to Example 1.

[0070] Figure 2 This is a system framework diagram of the electronic component supply chain intelligent management system in Example 2. DETAILED DESCRIPTION

[0071] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0072] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0073] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0074] Example 1

[0075] Reference Figure 1 , which is the first embodiment of the present invention, provides an electronic component supply chain intelligent management method, including:

[0076] S1: Collect real-time market transaction data of electronic components, including historical procurement data, historical inventory data, supplier delivery cycle data, and component life cycle data.

[0077] S1.1: Connect multiple electronic component trading platforms through the market data collection interface, establish a data collection channel, and collect real-time market transaction data of electronic components.

[0078] S1.2: Real-time market transaction data is classified and stored according to data type to form historical procurement data, historical inventory data, supplier delivery cycle data, and component life cycle data.

[0079] S1.3: Use a distributed data processing framework to clean real-time market transaction data and establish a time series feature extraction module to perform time series decomposition on historical procurement data and historical inventory data to extract trend items, cycle items, and random items.

[0080] S1.4: Build a supplier profiling model to quantify the delivery time, delivery accuracy, and delivery quality indicators in the supplier delivery cycle data.

[0081] In an optional implementation, the method for constructing a supplier portrait model is to establish a basic feature layer for suppliers and collect static information of suppliers, where the static information includes registered capital, establishment time, employee size and production qualifications; construct a supplier behavioral feature layer and convert the monthly delivery volume, product quality pass rate and price fluctuation range in the supplier delivery cycle data into dynamic feature vectors; use a time decay function to process the supplier delivery cycle data and assign differentiated weights to delivery performance in different periods; set up a supplier credit evaluation matrix and map the supplier's performance, quality complaint rate and payment period compliance into credit scores; introduce a collaborative filtering algorithm and cluster suppliers based on similarity calculations between suppliers; use the principal component analysis method to reduce the dimensionality of the basic feature layer, behavioral feature layer and credit evaluation matrix and extract key features; generate a supplier portrait label system and form a supplier portrait model.

[0082] S1.5: Set lifecycle status labels for component lifecycle data, where the lifecycle status labels include mass production period, maturity period, decline period, and production suspension warning period.

[0083] S1.6: Store the processed real-time market transaction data into a distributed database and establish an index structure based on timestamps.

[0084] S2: Based on historical procurement data and historical inventory data, a demand forecasting model for electronic components is constructed to generate demand forecast results.

[0085] S2.1: Use the sequence segmentation method to divide the historical purchase data and historical inventory data in the real-time market transaction data into training set, validation set and test set according to the preset ratio.

[0086] S2.2: Construct a long short-term memory network model to capture the long-term dependency of electronic component procurement through a gating mechanism.

[0087] It should be noted that the long short-term memory network model includes an input layer, two LSTM hidden layers and an output layer.

[0088] In an optional embodiment, 128 neurons are set in the first LSTM hidden layer of the long short-term memory network model, and 64 neurons are set in the second LSTM hidden layer; the effective features are used as the input of the long short-term memory network model, and the long-term dependency of electronic component procurement is captured through a gating mechanism.

[0089] S2.3: Construct a gradient boosting tree model, use the decision tree ensemble method to model the nonlinear demand pattern of electronic components, and improve the prediction accuracy by iteratively fitting the residuals.

[0090] In an optional embodiment, the tree depth of the gradient boosting tree model is set to 6, and the number of trees is set to 100; valid features are input into the gradient boosting tree model, and residuals are fitted in an iterative manner to improve prediction accuracy.

[0091] S2.4: Use the stacking integration method to perform weighted fusion on the outputs of the long short-term memory network model and the gradient boosting tree model to form a demand forecasting model.

[0092] S2.5: Automatically tune the hyperparameters of the demand forecasting model based on the Bayesian optimization algorithm, and optimize the model performance on the validation set through cross-validation.

[0093] S2.6: Input the test set into the optimized demand forecasting model, and combine it with the safety stock level and market volatility factors to generate demand forecast results for electronic components.

[0094] S3: Based on supplier delivery cycle data and component life cycle data, the demand forecast results are input into the risk assessment model, and the out-of-stock risk index is output through the fuzzy comprehensive evaluation method.

[0095] S3.1: Establish a risk assessment model based on a deep neural network architecture.

[0096] In an optional embodiment, the method for constructing a risk assessment model is to build a deep neural network architecture, wherein the deep neural network architecture includes an input layer, a feature extraction layer, a risk quantification layer and an output layer; the feature extraction layer is provided with a plurality of convolutional layers and pooling layers alternating in structure; a first neuron group is provided in the input layer, wherein the first neuron group corresponds to the input interface of the supplier delivery cycle characteristics, the component life cycle characteristics and the demand forecast characteristics; a sliding convolution kernel is provided in the convolution layer of the feature extraction layer, wherein the sliding convolution kernel performs a nonlinear transformation on the input features to extract multi-scale feature information; a maximum pooling operation is used in the pooling layer of the feature extraction layer to reduce the feature dimension and retain the features; a second neuron group is provided in the risk quantification layer, wherein the second neuron group combines and weights the extracted feature information through a fully connected layer; a third neuron group is provided in the output layer, wherein the third neuron group uses a Softmax activation function to output the probability of the risk level; a batch normalization layer is provided to standardize the outputs of the feature extraction layer and the risk quantification layer; a residual connection structure is used to establish jump connections between the convolution blocks of the feature extraction layer to alleviate the gradient disappearance problem and complete the establishment of the risk assessment model.

[0097] S3.2: Input historical out-of-stock event data into the risk assessment model for training, and optimize the model parameters through the back-propagation algorithm.

[0098] S3.3: Construct an evaluation indicator matrix and standardize the delivery on-time rate, average delay time and delivery volatility in the supplier delivery cycle data to form a delivery evaluation vector.

[0099] S3.4: Extract life cycle stage indicators, alternative product indicators, and discontinuation warning indicators from component life cycle data to generate a life cycle assessment vector.

[0100] S3.5: Calculate the demand growth index, fluctuation index, and forecast accuracy index based on the demand forecast results to form a demand assessment vector.

[0101] S3.6: Input the delivery assessment vector, life cycle assessment vector, and demand assessment vector into the risk assessment model to obtain an initial risk score.

[0102] S3.7: Establish a fuzzy membership function, map the initial risk score to the interval [0,1], and construct a fuzzy relationship matrix.

[0103] S3.8: Use the hierarchical analysis method to calculate the evaluation index weight vector, perform fuzzy compound operation on the weight vector and the fuzzy relationship matrix, and output the out-of-stock risk index.

[0104] S3.9: When the evaluation indicators fluctuate abnormally, the risk assessment model automatically adjusts the risk score weight and recalculates the out-of-stock risk index.

[0105] S4: Classify electronic components into risk levels based on the out-of-stock risk index, generate supplier procurement quota plans, and realize intelligent management of the electronic component supply chain.

[0106] S4.1: Based on the out-of-stock risk index, electronic components are divided into high-risk, medium-risk, and low-risk areas.

[0107] In an optional embodiment, when the out-of-stock risk index of an electronic component is greater than a first threshold and the capacity utilization rate exceeds a second threshold, the electronic component is included in a temporary observation list, and key indicators of the electronic components on the temporary observation list are dynamically monitored for three consecutive monitoring cycles, where the key indicators include supplier quotation fluctuations, supplier inventory and market demand growth rate; if the supplier quotation fluctuation range exceeds the third threshold, or the supplier inventory is lower than the fourth threshold and the market demand growth rate exceeds the fifth threshold, the electronic component is classified into a high-risk area; if the number of alternative suppliers of electronic components in the high-risk area is higher than the sixth threshold and the supplier concentration index is lower than the seventh threshold, a risk downgrade assessment is triggered; when the market supply-demand ratio during the downgrade assessment period is greater than the eighth threshold and there is no abnormal fluctuation in the continuous monitoring period, the electronic component is adjusted to the medium-risk area; when the supplier fulfillment rate of the electronic component in the medium-risk area exceeds the ninth threshold and the inventory turnover rate is lower than the tenth threshold, the electronic component is classified into a low-risk area; if the supplier's product enters the high-risk area, the risk level of the associated product is automatically raised by one level.

[0108] S4.2: For electronic components in high-risk areas, establish supplier qualification rating standards, which include financial status score, production capacity scale score, historical credit score and capacity utilization rate.

[0109] S4.3: Based on the supplier qualification rating standards, sort the suppliers and select high-quality suppliers.

[0110] In an optional implementation, a comprehensive supplier scoring matrix is ​​constructed based on the financial status score, production capacity scale score, historical credit score and capacity utilization rate in the supplier qualification rating standards; when the supplier's financial status score is greater than 85 points and the revenue growth rate has been continuously positive in the last four quarters, the supplier will be included in the financially sound list.

[0111] For example, the supplier's production capacity is classified into different levels: when the average monthly production capacity is greater than 1 million pieces and the capacity utilization rate is less than 80%, it is classified as Class A capacity; when the average monthly production capacity is between 500,000 and 1 million pieces and the capacity utilization rate is between 80% and 90%, it is classified as Class B capacity; when the average monthly production capacity is less than 500,000 pieces or the capacity utilization rate is greater than 90%, it is classified as Class C capacity.

[0112] For example, a dynamic evaluation mechanism for suppliers' historical credit history is established: when the on-time delivery rate is greater than 95% for 12 consecutive months, the credit rating is upgraded by one level; if there are major quality problems or delayed delivery for more than 15 days in a single month, the credit rating is downgraded by two levels; when a supplier is in the credit observation period, if there is no bad record for three consecutive months, the observation status will be lifted.

[0113] For example, set the supplier competition and cooperation relationship evaluation indicators: when a supplier also supplies competitors, the proportion of exclusive supply agreements must be greater than 60%; if a supplier establishes a strategic alliance with other high-quality suppliers, the cooperation period exceeds 2 years and the annual cooperation projects are no less than 3, the score weighting coefficient will be increased by 0.1.

[0114] For example, we construct supplier preference criteria determination rules:

[0115] Basic conditions: The financial status score is in the financially sound list, and the production capacity grade is A or B;

[0116] Bonus points: Cooperation with strategic customers for more than 3 years, or possession of no less than 5 core patented technologies;

[0117] Veto items: Major quality accidents or intellectual property disputes in the past two years;

[0118] For example, when a supplier meets the following combination of conditions, it will be included in the list of high-quality suppliers:

[0119] Condition combination 1: Financially sound list + A-level production capacity + credit rating A or above + no veto items;

[0120] Condition combination 2: Financially sound list + B-level production capacity + credit rating of A or above + two or more bonus items + no veto items;

[0121] For example, a dynamic exit mechanism is set up for suppliers selected into the quality supplier list:

[0122] The assessment score for the current quarter is lower than 80 points for two consecutive quarters;

[0123] Or a single major quality problem causes customer losses exceeding RMB 1 million;

[0124] Or the capacity utilization rate exceeds 95% for three consecutive months and there are no plans to expand production.

[0125] S4.4: Use genetic algorithm to perform multi-objective optimization on the procurement quota of high-quality suppliers and generate procurement quota plan Y.

[0126] S4.5: Generate procurement quota plan X for electronic components in medium-risk areas and low-risk areas according to the preset strategy.

[0127] S4.6: Based on procurement quota plan Y and procurement quota plan X, set supplier procurement benchmark quotas and establish a dynamic adjustment mechanism. Through periodic risk index assessment results, the quota allocation is revised in real time to achieve intelligent management of the electronic component supply chain.

[0128] For example, when implementing procurement quota plan X, if the supplier has a good credit rating, a centralized procurement quota allocation strategy is adopted. Supplier qualifications are dynamically monitored and evaluated through the central procurement decision-making system, and a three-dimensional monitoring mechanism is established for supplier price commitments, delivery capabilities, and quality assurance. This three-dimensional monitoring mechanism includes real-time price monitoring, delivery cycle monitoring, and quality traceability monitoring. Real-time price monitoring dynamically tracks supplier quotations, delivery cycle monitoring provides real-time evaluation of supplier capacity utilization, and quality traceability monitoring analyzes supplier product quality data. If a supplier credit risk warning is issued, a distributed procurement quota allocation strategy is adopted. Supplier performance is decentralizedly managed through regional procurement centers, and a two-dimensional monitoring mechanism is established for supplier price commitments and delivery capabilities. Supplier product quality data is regularly evaluated through a combination of offline spot checks and online feedback. When the evaluation index falls below the threshold, the system automatically adjusts the procurement quota share of that supplier.

[0129] For example, when purchasing quota plan Y is implemented, the risk of supplier supply interruption occurs, and the emergency purchasing quota adjustment mechanism is activated. The supplier's performance is monitored throughout the process through the intelligent risk warning system. A two-dimensional monitoring mechanism is set up for the supplier's price commitment and delivery capability. The intelligent risk warning system accurately locates the supply interruption risk. When the risk is confirmed, the purchasing management system pushes a real-time warning, notifies the purchasing specialist to activate the alternative supplier resource pool, and disperses the purchase orders. The purchasing management system automatically receives the supplier status information fed back by the intelligent risk warning system, dynamically reallocates the purchasing quota, and automatically freezes the purchasing quota of the problem supplier. When the supplier risk is resolved, the purchasing management system conducts a comprehensive assessment of the supplier. After the assessment is passed, its purchasing quota is gradually restored until the risk repair process is completed.

[0130] In summary, the present invention constructs an accurate demand forecasting model and risk assessment model by collecting and integrating real-time market transaction data from multiple dimensions such as historical procurement, inventory, supplier delivery cycle and life cycle, thereby realizing quantitative analysis and dynamic early warning of the risk of shortage of electronic components; outputs the shortage risk index through the fuzzy comprehensive evaluation method, and divides the electronic components into high, medium and low risk levels accordingly, which not only strengthens the key management and control capabilities of key components, but also improves the degree of refinement of overall resource allocation; introduces a supplier qualification rating system in high-risk areas, and adopts genetic algorithms for multi-objective optimization allocation, realizes the optimal configuration of high-risk component procurement quotas, and improves supply continuity and stability; this method effectively improves the forecast accuracy and risk resistance of the supply chain, realizes scientific decision-making and intelligent management of electronic component procurement, and significantly reduces operating costs and supply interruption risks caused by shortages.

[0131] Example 2

[0132] Reference Figure 2 , which is the second embodiment of the present invention, this embodiment further provides an electronic component supply chain intelligent management system, including:

[0133] A collection module is used to collect real-time market transaction data of electronic components, wherein the real-time market transaction data includes historical procurement data, historical inventory data, supplier delivery cycle data and component life cycle data;

[0134] A demand forecasting module, which constructs a demand forecasting model for electronic components based on the historical procurement data and the historical inventory data, and generates a demand forecasting result;

[0135] a stock-out risk assessment module, which inputs the demand forecast result into a risk assessment model based on the supplier delivery cycle data and the component life cycle data, and outputs a stock-out risk index through a fuzzy comprehensive evaluation method;

[0136] The classification management module is used to classify the risk levels of electronic components according to the out-of-stock risk index, generate supplier procurement quota plans, and realize intelligent management of the electronic component supply chain.

[0137] This embodiment also provides an electronic device, which includes a processor, memory, a communication interface, a display screen, and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a multi-task edge computing resource scheduling method is implemented. The display screen of the computer device can be a liquid crystal display or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0138] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method proposed in the above embodiment is implemented.

[0139] The storage medium proposed in this embodiment and the method proposed in the above embodiment belong to the same inventive concept. For technical details not fully described in this embodiment, please refer to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0140] Through the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the method of the embodiment of the present invention.

[0141] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0142] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiment of the present application can be implemented in various computer languages.

[0143] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0144] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0145] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0146] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0147] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. An intelligent management method for an electronic component supply chain, characterized by: include, Collecting real-time market transaction data of electronic components, wherein the real-time market transaction data includes historical procurement data, historical inventory data, supplier delivery cycle data, and component life cycle data; Building a demand forecasting model for electronic components based on the historical procurement data and the historical inventory data, and generating demand forecasting results; Based on the supplier delivery cycle data and the component life cycle data, the demand forecast result is input into a risk assessment model, and a stock-out risk index is output through a fuzzy comprehensive evaluation method; The electronic components are classified into risk levels according to the out-of-stock risk index, and a supplier procurement quota plan is generated to achieve intelligent management of the electronic component supply chain.

2. The electronic component supply chain intelligent management method according to claim 1, characterized in that: The electronic components are classified into risk levels according to the out-of-stock risk index, and a supplier procurement quota plan is generated, including: Based on the out-of-stock risk index, electronic components are divided into high-risk, medium-risk and low-risk areas; Establish supplier qualification rating standards for electronic components in high-risk areas, including financial status scores, production capacity scale scores, historical credit scores, and capacity utilization rates; Based on the supplier qualification rating standards, suppliers are ranked and high-quality suppliers are selected; A genetic algorithm is used to perform multi-objective optimization on the purchase quota of the high-quality supplier to generate a purchase quota plan Y; Generate a procurement quota plan X for electronic components in medium-risk and low-risk areas according to the preset strategy; Based on the procurement quota plan Y and the procurement quota plan X, a supplier procurement benchmark quota is set, and a dynamic adjustment mechanism is established. Through the periodic risk index assessment results, the quota allocation is corrected in real time to achieve intelligent management of the electronic component supply chain.

3. The electronic component supply chain intelligent management method according to claim 2, characterized in that: Electronic components are classified into risk levels, including: When the out-of-stock risk index of an electronic component exceeds a first threshold and the capacity utilization rate exceeds a second threshold, the electronic component is placed on a temporary watch list. Dynamic monitoring of key indicators of the electronic components on the temporary watch list is conducted over three consecutive monitoring cycles, where the key indicators include supplier quotation fluctuations, supplier inventory, and market demand growth rate. If the supplier's quotation fluctuation exceeds the third threshold, or the supplier's inventory is lower than the fourth threshold and the market demand growth rate exceeds the fifth threshold, the electronic component will be classified as a high-risk area; If the number of alternative suppliers of electronic components in the high-risk area is higher than the sixth threshold and the supplier concentration index is lower than the seventh threshold, a risk downgrade assessment is triggered; When the market supply-demand ratio is greater than the eighth threshold during the downgrade assessment period and there are no abnormal fluctuations during the continuous monitoring period, the electronic component will be adjusted to the medium-risk zone; If the supplier fulfillment rate of electronic components in the medium-risk zone exceeds the ninth threshold and the inventory turnover rate is lower than the tenth threshold, the electronic components will be classified as low-risk areas; If a supplier's products enter a high-risk area, the risk level of the associated products will automatically be raised by one level.

4. The electronic component supply chain intelligent management method according to claim 3, characterized in that: The method for obtaining the out-of-stock risk index is: Establish a risk assessment model based on deep neural network architecture; Inputting historical out-of-stock event data into the risk assessment model for training, and optimizing the model parameters through a back-propagation algorithm; Construct an evaluation indicator matrix, and standardize the delivery on-time rate, average delay time, and delivery volatility in the supplier's delivery cycle data to form a delivery evaluation vector; Extract life cycle stage indicators, alternative product indicators and discontinuation warning indicators from component life cycle data to generate life cycle assessment vectors; Calculate demand growth index, fluctuation index and forecast accuracy index based on demand forecast results to form a demand assessment vector; Inputting the delivery assessment vector, the life cycle assessment vector, and the demand assessment vector into the risk assessment model to obtain an initial risk score; Establishing a fuzzy membership function, mapping the initial risk score to the interval [0,1], and constructing a fuzzy relationship matrix; Calculate the evaluation index weight vector using the hierarchical analysis method, perform fuzzy compound operation on the weight vector and the fuzzy relationship matrix, and output the out-of-stock risk index; When the evaluation index fluctuates abnormally, the risk assessment model automatically adjusts the risk score weight and recalculates the out-of-stock risk index.

5. The electronic component supply chain intelligent management method according to claim 4, characterized in that: The risk assessment model is constructed as follows: Building the deep neural network architecture, wherein the deep neural network architecture includes an input layer, a feature extraction layer, a risk quantification layer, and an output layer; the feature extraction layer is provided with a plurality of convolutional layers and pooling layers alternating in structure; A first neuron group is set in the input layer, wherein the first neuron group corresponds to the input interface of supplier delivery cycle characteristics, component life cycle characteristics and demand forecast characteristics; Setting a sliding convolution kernel in the convolution layer of the feature extraction layer, wherein the sliding convolution kernel performs a nonlinear transformation on the input features to extract multi-scale feature information; A maximum pooling operation is adopted in the pooling layer of the feature extraction layer to reduce the feature dimension and retain the features; Setting a second neuron group in the risk quantification layer, wherein the second neuron group combines and weights the extracted feature information through a fully connected layer; A third neuron group is set in the output layer, wherein the third neuron group uses a Softmax activation function to output the probability of the risk level; Setting a batch normalization layer to standardize the outputs of the feature extraction layer and the risk quantification layer; A residual connection structure is adopted to establish skip connections between the convolution blocks of the feature extraction layer to alleviate the gradient vanishing problem and complete the establishment of the risk assessment model.

6. The electronic component supply chain intelligent management method according to claim 4, characterized in that: The method for generating the demand forecast result is: The sequence segmentation method is used to divide the historical purchase data and historical inventory data in the real-time market transaction data into training set, validation set and test set according to the preset ratio; Build a long short-term memory network model to capture the long-term dependency of electronic component procurement through a gating mechanism; Construct a gradient boosting tree model, use the decision tree ensemble method to model the nonlinear demand pattern of electronic components, and improve the prediction accuracy by iteratively fitting the residuals; Using a stacking integration method to weightedly fuse the outputs of the long short-term memory network model and the gradient boosting tree model to form a demand forecasting model; Automatically tune the hyperparameters of the demand forecasting model based on a Bayesian optimization algorithm, and optimize the model performance on a validation set through cross-validation; The test set is input into the optimized demand forecasting model, and combined with the safety stock level and market fluctuation factors, the demand forecast results of electronic components are generated.

7. The electronic component supply chain intelligent management method according to claim 6, characterized in that: The method for collecting real-time market transaction data is: Connect multiple electronic component trading platforms through the market data collection interface, establish data collection channels, and collect real-time market transaction data of electronic components; Classify and store the real-time market transaction data according to data type to form historical procurement data, historical inventory data, supplier delivery cycle data, and component life cycle data; A distributed data processing framework is used to clean the real-time market transaction data, and a time series feature extraction module is established to perform time series decomposition on the historical procurement data and the historical inventory data to extract trend items, periodic items, and random items; Construct a supplier profile model to quantify the delivery time, delivery accuracy, and delivery quality indicators in the supplier's delivery cycle data; Setting a life cycle status label for the component life cycle data, wherein the life cycle status label includes a mass production period, a mature period, a decline period, and a production suspension warning period; The processed real-time market transaction data is stored in a distributed database and an index structure is established based on timestamps.

8. An electronic component supply chain intelligent management system, based on the electronic component supply chain intelligent management method according to any one of claims 1 to 7, characterized in that: include, A collection module is used to collect real-time market transaction data of electronic components, wherein the real-time market transaction data includes historical procurement data, historical inventory data, supplier delivery cycle data and component life cycle data; A demand forecasting module, which constructs a demand forecasting model for electronic components based on the historical procurement data and the historical inventory data, and generates a demand forecasting result; a stock-out risk assessment module, which inputs the demand forecast result into a risk assessment model based on the supplier delivery cycle data and the component life cycle data, and outputs a stock-out risk index through a fuzzy comprehensive evaluation method; The classification management module is used to classify the risk levels of electronic components according to the out-of-stock risk index, generate supplier procurement quota plans, and realize intelligent management of the electronic component supply chain.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the electronic component supply chain intelligent management method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the electronic component supply chain intelligent management method according to any one of claims 1 to 7 are implemented.

Images