Supply chain service management system and method based on data analysis

Abstract

The invention relates to the technical field of supply chain service management, and discloses a supply chain service management system and method based on data analysis, and the system comprises a data collection layer, a data preprocessing layer, a data analysis layer, a prediction analysis model module, an optimization analysis model module, a risk analysis model module, a core service layer, and an application layer. According to the supply chain service management system and method based on data analysis and data value deep mining, supply chain data is converted into a decision basis through a multi-level analysis model, the problem of data application superficial layer is solved, the purchase cost is reduced, the inventory turnover rate is effectively improved, and the data value deep mining efficiency is improved. According to the model adaptation scenario, professional analysis models are designed for different scenarios such as purchase, inventory and logistics, the problem of model simplification is solved, and the requirements of multi-industry and multi-scale enterprises are met; according to the full-link collaborative optimization, the bottleneck root of a supply chain is positioned through data association analysis, so that purchase-inventory-logistics collaborative optimization is realized, and the linkage problem caused by local adjustment is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of supply chain service management, in particular to a supply chain service management system and method based on data analysis. BACKGROUND

[0002] Although the current supply chain management system has covered the basic business processes, there are still significant shortcomings in the data value mining and application level: data application is shallow: the existing system only realizes data storage and simple statistics, and does not form a closed loop of "data collection - analysis - decision - feedback", which cannot provide deep support for supply chain optimization; the analysis model is single: there is a lack of professional analysis model suitable for multiple scenarios of supply chain, which is difficult to cope with complex problems such as demand fluctuation, supplier risk and logistics cost fluctuation; the decision response is lagging: relying on manual interpretation of data reports leads to lagging decisions such as procurement adjustment, inventory optimization and risk response, which cannot match market changes in real time; the data linkage is fragmented: the analysis data of procurement, inventory and logistics are independent of each other, and the whole-link data correlation analysis has not been formed, which makes it difficult to locate the root cause of the supply chain bottleneck.

[0003] Therefore, there is an urgent need for a supply chain service management system and method based on data analysis as the core engine, which integrates multi-dimensional data, adapts to multiple scene models and realizes decision automation. SUMMARY

[0004] (I) Technical problems to be solved In view of the shortcomings of the prior art, the present application provides a supply chain service management system and method based on data analysis, which has the advantages of procurement-inventory-logistics collaborative optimization and adaptation to the needs of enterprise business growth and industry development, and solves the chain problem caused by local adjustment; procurement-inventory-logistics collaborative optimization avoids the chain problem caused by local adjustment.

[0005] (II) Technical solutions To achieve the above-mentioned purpose, the present application provides the following technical solutions: a supply chain service management system based on data analysis, comprising a data collection layer, a data preprocessing layer, a data analysis layer, a prediction analysis model module, an optimization analysis model module, a risk analysis model module, a core service layer, an application layer, A supply chain service management method based on data analysis, comprising the following steps S1, data collection: through the data collection layer, synchronously collecting internal business data, external collaborative data, environmental dynamic data and unstructured data, forming a supply chain whole-link data set; S2, data preprocessing: cleaning, fusing and standardizing the collected data, establishing whole-link data correlation and storing in the corresponding database; S3, data analysis; S4, Service Execution: Based on the data analysis results, the core service layer automatically executes operations such as generating procurement suggestions, reminding inventory to replenish stock, and planning logistics routes, while also pushing the analysis results to the application layer; S5. Decision Feedback: Users view the analysis results and optimization plans through the application layer, execute decision-making operations, and the operation data flows back to the data acquisition layer, forming a closed loop.

[0006] Preferably, the data analysis includes the following steps: S301: Basic statistical analysis, generating basic reports on procurement, inventory, and logistics; S302: Predictive Analysis, calls demand forecasting, price forecasting, and inventory demand models, and outputs forecast results; S303: Optimization analysis, based on the forecast results, calls the procurement optimization, inventory optimization, and logistics optimization models to generate optimization solutions; S304: Risk analysis, running supplier risk, bottleneck location, and anomaly early warning models to identify risk points and trigger early warnings.

[0007] Preferably, the data acquisition layer is used to collect multi-dimensional data across the entire supply chain, including: Internal business data module: Connects to ERP, production management, and financial systems to collect structured data such as purchase orders, inventory ledgers, production plans, and settlement records; External collaborative data module: Connects to supplier systems, customer systems, and logistics service provider systems via API; Environmental dynamics data module: Real-time capture of market price data, policy and regulatory data, and external impact data; Unstructured data module: Collects unstructured data such as supplier qualification documents, scanned copies of logistics documents, and customer feedback texts, and converts them into structured data through OCR and NLP technologies.

[0008] Preferably, the data preprocessing layer provides a high-quality data foundation for data analysis, including: Data cleaning module: Removes redundant data, corrects outliers, and completes missing data; Data fusion module: Establishes data association rules to achieve unique identification association between purchase orders, inventory batches, logistics tracking numbers, and customer orders, forming a full-link data chain; Data standardization module: unifies data formats and quantifies unstructured data; Data storage module: It adopts a hybrid storage architecture of relational database + time series database + data lake. Structured data is stored in relational database, time series data is stored in time series database, and raw data is stored in data lake for backtracking analysis.

[0009] Preferably, the data analysis layer includes a basic statistical analysis module, a predictive analysis model module, an optimization analysis model module, and a supplier risk sub-model. The data analysis layer is designed to build a multi-level data analysis model system to provide decision support for core services. Basic statistical analysis module: Enables basic analysis such as procurement cost statistics, inventory turnover rate calculation, and logistics timeliness analysis, and generates standardized reports; Predictive analysis model module: Demand forecasting sub-model: Based on LSTM neural network, it integrates historical sales data, market trends and holiday factors to predict customer demand in the next 1-3 months with high prediction accuracy; Price forecasting sub-model: Using the ARIMA time series model, combined with the supply and demand relationship of raw materials and the impact of policies, it predicts the future price trend of raw materials and provides a basis for procurement pricing; Inventory demand sub-model: Based on the safety stock formula and demand forecast data, calculate the optimal safety stock and replenishment point for different materials; Optimization analysis model module: Procurement optimization sub-model: With the goal of "lowest cost + controllable risk", it generates the optimal procurement plan by combining supplier fulfillment rate, price and delivery cycle data; Inventory optimization sub-model: Optimize inventory structure and reduce backlog costs and material shortage risks by using ABC classification and EOQ economic order quantity model; Logistics optimization sub-model: Based on genetic algorithm, combined with transportation distance, time requirements, and logistics provider quotations, it plans the optimal delivery route and logistics provider allocation scheme to reduce logistics costs; Supplier Risk Sub-model: Construct a risk assessment indicator system, use the analytic hierarchy process (AHP) to quantify supplier risk levels, and trigger high-risk warnings; Supply chain bottleneck sub-model: Identify bottleneck links by correlating and analyzing procurement, inventory, and logistics data; Anomaly warning sub-model: Set thresholds for key indicators, monitor in real time and trigger anomaly warnings.

[0010] Preferably, the core service layer Based on the output of the data analysis layer, intelligent operation of supply chain business is achieved, including: Intelligent Procurement Service Module: Based on demand forecasting and procurement optimization model results, it automatically generates procurement suggestion orders and supports one-click approval; it also pushes supplier risk warnings in real time to assist in adjusting procurement strategies. Precise Inventory Service Module: Automatically triggers replenishment reminders based on inventory optimization models and safety stock data; generates inventory adjustment plans through inventory health analysis; Dynamic Logistics Service Module: Based on the output of the logistics optimization model, it automatically allocates logistics providers and plans routes; tracks logistics trajectories in real time, and combines an anomaly warning model to proactively address transportation delays; Collaborative Decision-Making Service Module: Pushes demand forecast data to suppliers to guide their capacity planning; synchronizes order fulfillment forecasts with customers to improve customer experience; Data Dashboard Module: Integrates the results of various analytical models, visualizes key supply chain indicators, and supports drill-down analysis.

[0011] Compared with existing technologies, the present invention provides a supply chain service management system and method based on data analysis, which has the following beneficial effects: 1. This data-driven supply chain service management system and methodology deeply mines the value of data through multi-level analytical models, transforming supply chain data into decision-making support, solving the problem of superficial data application, reducing procurement costs, and effectively improving inventory turnover. Automated analysis and early warning replace manual interpretation, shortening demand forecasting and risk identification response time from days to hours, enhancing supply chain flexibility. Scenario-adaptive models are designed for different scenarios such as procurement, inventory, and logistics, solving the problem of model uniformity and adapting to the needs of enterprises of various industries and sizes. End-to-end collaborative optimization uses data correlation analysis to pinpoint the root causes of supply chain bottlenecks, achieving collaborative optimization of "procurement-inventory-logistics," avoiding chain reactions caused by partial adjustments. It is highly scalable, supporting the addition of new analytical models and data sources to adapt to enterprise business growth and industry development needs. Attached Figure Description

[0012] Fig. 1 This is a schematic diagram of the service management method of the present invention; Fig. 2 This is a schematic diagram of the data analysis steps of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Please see Figs. 1-2 A data-driven supply chain service management system includes a data acquisition layer, a data preprocessing layer, a data analysis layer, a predictive analysis model module, an optimization analysis model module, a risk analysis model module, a core service layer, and an application layer. A data-driven supply chain service management method includes the following steps: S1. Data Acquisition: Through the data acquisition layer, internal business data, external collaborative data, environmental dynamic data, and unstructured data are collected synchronously to form a full-chain dataset of the supply chain; S2. Data preprocessing: Cleaning, merging, and standardizing the collected data, establishing full-link data association, and storing it in the corresponding database; S3. Conduct data analysis; S4, Service Execution: Based on the data analysis results, the core service layer automatically executes operations such as generating procurement suggestions, reminding inventory to replenish stock, and planning logistics routes, while also pushing the analysis results to the application layer; S5. Decision Feedback: Users view the analysis results and optimization plans through the application layer, execute decision-making operations, and the operation data flows back to the data acquisition layer, forming a closed loop.

[0015] Furthermore, the data analysis includes the following steps: S301: Basic statistical analysis, generating basic reports on procurement, inventory, and logistics; S302: Predictive Analysis, calls demand forecasting, price forecasting, and inventory demand models, and outputs forecast results; S303: Optimization analysis, based on the forecast results, calls the procurement optimization, inventory optimization, and logistics optimization models to generate optimization solutions; S304: Risk analysis, running supplier risk, bottleneck location, and anomaly early warning models to identify risk points and trigger early warnings.

[0016] Furthermore, the data acquisition layer is used to collect multi-dimensional data across the entire supply chain, including: Internal business data module: Connects to ERP, production management, and financial systems to collect structured data such as purchase orders, inventory ledgers, production plans, and settlement records; External collaborative data module: Connects to supplier systems, customer systems, and logistics service provider systems via API; Environmental dynamics data module: Real-time capture of market price data, policy and regulatory data, and external impact data; Unstructured data module: Collects unstructured data such as supplier qualification documents, scanned copies of logistics documents, and customer feedback texts, and converts them into structured data through OCR and NLP technologies.

[0017] Furthermore, the data preprocessing layer provides a high-quality data foundation for data analysis, including: Data cleaning module: Removes redundant data, corrects outliers, and completes missing data; Data fusion module: Establishes data association rules to achieve unique identification association between purchase orders, inventory batches, logistics tracking numbers, and customer orders, forming a full-link data chain; Data standardization module: unifies data formats and quantifies unstructured data; Data storage module: It adopts a hybrid storage architecture of relational database + time series database + data lake. Structured data is stored in relational database, time series data is stored in time series database, and raw data is stored in data lake for backtracking analysis.

[0018] Furthermore, the data analysis layer includes a basic statistical analysis module, a predictive analysis model module, an optimization analysis model module, and a supplier risk sub-model. The data analysis layer is designed to build a multi-level data analysis model system to provide decision support for core services. Basic statistical analysis module: Enables basic analysis such as procurement cost statistics, inventory turnover rate calculation, and logistics timeliness analysis, and generates standardized reports; Predictive analysis model module: Demand forecasting sub-model: Based on LSTM neural network, it integrates historical sales data, market trends and holiday factors to predict customer demand in the next 1-3 months with high prediction accuracy; Price forecasting sub-model: Using the ARIMA time series model, combined with the supply and demand relationship of raw materials and the impact of policies, it predicts the future price trend of raw materials and provides a basis for procurement pricing; Inventory demand sub-model: Based on the safety stock formula and demand forecast data, calculate the optimal safety stock and replenishment point for different materials; Optimization analysis model module: Procurement optimization sub-model: With the goal of "lowest cost + controllable risk", it generates the optimal procurement plan by combining supplier fulfillment rate, price and delivery cycle data; Inventory optimization sub-model: Optimize inventory structure and reduce backlog costs and material shortage risks by using ABC classification and EOQ economic order quantity model; Logistics optimization sub-model: Based on genetic algorithm, combined with transportation distance, time requirements, and logistics provider quotations, it plans the optimal delivery route and logistics provider allocation scheme to reduce logistics costs; Supplier Risk Sub-model: Construct a risk assessment indicator system, use the analytic hierarchy process (AHP) to quantify supplier risk levels, and trigger high-risk warnings; Supply chain bottleneck sub-model: Identify bottleneck links by correlating and analyzing procurement, inventory, and logistics data; Anomaly warning sub-model: Set thresholds for key indicators, monitor in real time and trigger anomaly warnings.

[0019] Furthermore, the core service layer Based on the output of the data analysis layer, intelligent operation of supply chain business is achieved, including: Intelligent Procurement Service Module: Based on demand forecasting and procurement optimization model results, it automatically generates procurement suggestion orders and supports one-click approval; it also pushes supplier risk warnings in real time to assist in adjusting procurement strategies. Precise Inventory Service Module: Automatically triggers replenishment reminders based on inventory optimization models and safety stock data; generates inventory adjustment plans through inventory health analysis; Dynamic Logistics Service Module: Based on the output of the logistics optimization model, it automatically allocates logistics providers and plans routes; tracks logistics trajectories in real time, and combines an anomaly warning model to proactively address transportation delays; Collaborative Decision-Making Service Module: Pushes demand forecast data to suppliers to guide their capacity planning; synchronizes order fulfillment forecasts with customers to improve customer experience; Data Dashboard Module: Integrates the results of various analytical models, visualizes key supply chain indicators, and supports drill-down analysis.

[0020] Example 1: Applications of small and medium-sized electronic manufacturing enterprises Data Acquisition: Collect purchase orders and inventory data from the ERP system, connect with the systems of 3 core chip suppliers (delivery cycle, pass rate), and capture chip market price and logistics timeliness data; Data preprocessing: Cleaning abnormal purchase prices and linking them to the data chain of "chip purchase orders - inventory batches - production work orders - customer orders"; Data Analysis: Demand Forecasting: Based on customer order and new product launch data from the past 6 months, the LSTM model predicts chip demand growth for the next month; Procurement optimization: Combining supplier delivery cycles (2 weeks for supplier A, 3 weeks for supplier B) and prices, a solution is generated that allocates 60% of orders to supplier A (with guaranteed delivery period) and 40% of orders to supplier B. Risk warning: Identify the threshold of supplier B's contract fulfillment rate decline over the past 3 months, triggering a risk warning; Service Execution: The system automatically generates a procurement plan and pushes risk alerts for Supplier B to the procurement end; Decision feedback: The procurement staff approves the plan, communicates with supplier B to make improvements, and the operational data is fed back into the system.

[0021] Example 2: Application by large FMCG retail enterprises Data collection: Collect sales data from 100+ stores, inventory data from the central warehouse, fulfillment data from 20+ suppliers, and delivery data from 5 logistics providers; capture data on raw material (such as packaging and raw materials) prices and holiday consumption trends. Data preprocessing: Integrating data from "store sales - warehouse inventory - logistics and distribution" to quantify customer feedback text; Data Analysis: Inventory optimization: The ABC classification method divides goods into categories A, B, and C. The EOQ model calculates that safety stock for category A goods increases by 20%, while inventory for category C goods decreases by 30%. Logistics optimization: Genetic algorithms plan "warehouse-store" delivery routes, and combined with holiday traffic congestion data, delivery time is shortened and logistics costs are reduced by 12%; Demand Forecast: Incorporating holiday and weather data (such as increased demand for beverages on hot days), we forecast a 30% increase in demand for Category A products next week. Service execution: The system automatically triggers replenishment of Category A products, pushes optimized routes to logistics providers, and synchronizes demand forecasts with stores; Decision feedback: The warehouse prepares goods according to the replenishment plan, the logistics provider executes the optimized route, the store adjusts the display, and the system continuously optimizes the model parameters after the data is fed back.

[0022] Example 3: A data-driven supply chain service management method, implemented based on the aforementioned system, includes the following steps: Data collection steps: The data acquisition layer modules are activated. The internal business data module connects to the enterprise's internal system and collects procurement, inventory, production, and financial data every 5 minutes. The external collaboration data module connects to the supplier, customer, and logistics provider systems and collects fulfillment, demand, and logistics data in real time. The environmental dynamic data module captures market, policy, and external impact data every hour. The unstructured data module processes qualification scans and text feedback data in real time to form a complete supply chain dataset. Data preprocessing: The data cleaning module uses the Z-score algorithm to identify outlier data, fills in missing data by using the mean, and removes duplicate data; the data fusion module establishes data associations for "purchase orders - inventory batches - logistics tracking numbers - customer orders" to generate a unique data chain; the data standardization module unifies data formats and quantifies unstructured data; the data storage module categorizes and stores the processed data into corresponding databases, while the original data is stored in a data lake. Data Analysis: The basic statistical analysis module performs statistical calculations on the preprocessed data, generates basic reports on procurement, inventory, and logistics, and pushes them to the corresponding user terminals at a preset frequency. The predictive analysis model module starts the LSTM demand forecasting sub-model, the ARIMA price forecasting sub-model, and the inventory demand sub-model, inputs preprocessed data, and outputs demand forecast, price forecast results, and the optimal safety stock threshold. Based on the prediction results, the optimization analysis model module activates the procurement optimization sub-model, inventory optimization sub-model, and logistics optimization sub-model, and combines supplier, inventory, and logistics data to generate procurement plans, inventory strategies, and logistics plans. The risk analysis model module runs the supplier risk sub-model, supply chain bottleneck sub-model, and anomaly warning sub-model to calculate supplier risk scores, locate bottleneck links, monitor indicator thresholds, and trigger risk and anomaly warnings. Service Implementation: The core service layer receives data analysis results; the intelligent procurement service module generates procurement suggestion orders and pushes them for approval; the precise inventory service module triggers replenishment reminders and generates inventory health reports; the dynamic logistics service module allocates logistics tasks and tracks trajectories; the collaborative decision-making service module pushes data and collaborative information to upstream and downstream partners; and the data dashboard module visualizes and displays core indicators. Decision feedback: Users can view analysis results and service information through various terminals in the application layer, and perform operations such as approval, inventory adjustment, and handling logistics anomalies; the operation data flows back to the data collection layer in real time as input for the next round of data collection and analysis, forming a closed-loop process of "collection-processing-analysis-application-feedback".

[0023] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A supply chain service management system based on data analysis, characterized in that: It includes a data acquisition layer, a data preprocessing layer, a data analysis layer, a predictive analysis model module, an optimization analysis model module, a risk analysis model module, a core service layer, and an application layer.

2. A supply chain service management method based on data analysis, characterized in that: Includes the following steps S1. Data Acquisition: Through the data acquisition layer, internal business data, external collaborative data, environmental dynamic data, and unstructured data are collected synchronously to form a full-chain dataset of the supply chain; S2. Data preprocessing: Cleaning, merging, and standardizing the collected data, establishing full-link data association, and storing it in the corresponding database; S3. Conduct data analysis; S4, Service Execution: Based on the data analysis results, the core service layer automatically executes operations such as generating procurement suggestions, reminding inventory to replenish stock, and planning logistics routes, while also pushing the analysis results to the application layer; S5. Decision Feedback: Users view the analysis results and optimization plans through the application layer, execute decision-making operations, and the operation data flows back to the data acquisition layer, forming a closed loop.

3. The supply chain service management method based on data analysis according to claim 2, characterized in that: The data analysis includes the following steps: S301: Basic statistical analysis, generating basic reports on procurement, inventory, and logistics; S302: Predictive Analysis, calls demand forecasting, price forecasting, and inventory demand models, and outputs forecast results; S303: Optimization analysis, based on the forecast results, calls the procurement optimization, inventory optimization, and logistics optimization models to generate optimization solutions; S304: Risk analysis, running supplier risk, bottleneck location, and anomaly early warning models to identify risk points and trigger early warnings.

4. The supply chain service management system based on data analysis according to claim 1, characterized in that: The data acquisition layer is used to collect multi-dimensional data across the entire supply chain, including: Internal business data module: Connects to ERP, production management, and financial systems to collect structured data such as purchase orders, inventory ledgers, production plans, and settlement records; External collaborative data module: Connects to supplier systems, customer systems, and logistics service provider systems via API; Environmental dynamics data module: Real-time capture of market price data, policy and regulatory data, and external impact data; Unstructured data module: Collects unstructured data such as supplier qualification documents, scanned copies of logistics documents, and customer feedback texts, and converts them into structured data through OCR and NLP technologies.

5. A supply chain service management system based on data analysis according to claim 1, characterized in that: The data preprocessing layer provides a high-quality data foundation for data analysis, including: Data cleaning module: Removes redundant data, corrects outliers, and completes missing data; Data fusion module: Establishes data association rules to achieve unique identification association between purchase orders, inventory batches, logistics tracking numbers, and customer orders, forming a full-link data chain; Data standardization module: unifies data formats and quantifies unstructured data; Data storage module: It adopts a hybrid storage architecture of relational database + time series database + data lake. Structured data is stored in relational database, time series data is stored in time series database, and raw data is stored in data lake for backtracking analysis.

6. The supply chain service management system based on data analysis according to claim 1, characterized in that: The data analysis layer includes a basic statistical analysis module, a predictive analysis model module, an optimization analysis model module, and a supplier risk sub-model. The data analysis layer is used to construct a multi-level data analysis model system.

7. A supply chain service management system based on data analysis according to claim 1, characterized in that: The core service layer Based on the output of the data analysis layer, intelligent operation of supply chain business is achieved, including: Intelligent Procurement Service Module: Based on demand forecasting and procurement optimization model results, it automatically generates procurement suggestion orders and supports one-click approval; it also pushes supplier risk warnings in real time to assist in adjusting procurement strategies. Precise Inventory Service Module: Automatically triggers replenishment reminders based on inventory optimization models and safety stock data; generates inventory adjustment plans through inventory health analysis; Dynamic Logistics Service Module: Based on the output of the logistics optimization model, it automatically allocates logistics providers and plans routes; tracks logistics trajectories in real time, and combines an anomaly warning model to proactively address transportation delays; Collaborative Decision-Making Service Module: Pushes demand forecast data to suppliers to guide their capacity planning; synchronizes order fulfillment forecasts with customers to improve customer experience; Data Dashboard Module: Integrates the results of various analytical models, visualizes key supply chain indicators, and supports drill-down analysis.

Images