Manufacturing industry supply chain management method based on block chain
By deploying blockchain and IoT devices in the manufacturing supply chain, combining data analysis tools and machine learning algorithms, the problem of high-cost transformation of traditional supply chain management is solved, real-time monitoring and optimization of the supply chain is realized, and production is improved Efficiency and market response speed are suitable for digital management of small and medium-sized enterprises.
Patent Information
- Application Number
- CN202410787460.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-18
- Publication Date
- 2024-11-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional manufacturing supply chain management faces the problem of high-cost transformation in the process of digital transformation, especially small and medium-sized enterprises are unable to undertake technological transformation and lack effective optimization management methods, which affects market changes adaptability and production efficiency.
Adopt blockchain-based manufacturing supply chain management method, by deploying data acquisition equipment at key nodes of the supply chain, using data analysis tools to collect, process and analyze real-time data, generate optimization suggestions reports, and combine the Internet of Things and machine learning algorithms. Realize real-time monitoring and optimization of the supply chain.
Real-time monitoring and optimization of the supply chain is realized, storage and logistics costs are reduced, production efficiency and market response speed are improved, data-driven decision-making support is provided, and digital management upgrades are suitable for small and medium-sized enterprises.
Smart Images

Figure CN118886931A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of supply chain management, and specifically relates to a manufacturing supply chain management method based on blockchain. Background Art
[0002] With the development of big data, artificial intelligence and blockchain technology, the supply chain management of modern traditional manufacturing industry is facing the need for transformation. This transformation can not only reduce costs and increase efficiency through digital management of supply chain, but also optimize each link in the supply chain based on market demand and changes, thereby improving the competitiveness of enterprises. However, in the face of the trend of digital management of enterprise supply chain, many enterprises do not know how to optimize the supply chain, and even if they know, they still face the problem of high cost transformation. For example, in the field of traditional leather shoe manufacturing, traditional leather shoe manufacturing industry is often small in scale and limited in funds, so it is difficult to bear the high cost of technological transformation. Summary of the invention
[0003] In view of the shortcomings of the existing technology, the present invention provides a manufacturing supply chain management method based on blockchain. Based on blockchain technology, the method not only significantly improves the digital management of the supply chain by enterprises by using data analysis tools, but also is low-cost and easy to promote and popularize; thus meeting the needs of small and medium-sized enterprises in the digital management upgrade to adapt to market changes and improve production efficiency.
[0004] In order to achieve the above objectives, the present invention provides a manufacturing supply chain management method based on blockchain through the following technical solutions:
[0005] The first step is to deploy data collection equipment at key nodes of the supply chain to collect key data in key links of the supply chain in real time;
[0006] In the second step, the data collection module normalizes and stores the key data collected in real time in the supply chain, and uploads the normalized data to the data processing and analysis module;
[0007] In the third step, the data analysis algorithm model in the data processing and analysis module is trained with the historical data set of key data in the supply chain to obtain a trained data analysis model;
[0008] In the fourth step, the data collection module inputs the key data of the supply chain collected in real time into the forming data analysis model, and the forming data analysis model outputs the key performance indicators;
[0009] Step 5: The optimization suggestion algorithm model of the optimization suggestion and reporting module is trained using the historical data set of key performance indicators to obtain a trained optimization suggestion model.
[0010] Finally, the data collection module will collect key data of the supply chain in real time or retrieve the key data of the historical supply chain stored in it, and input it into the forming data analysis model of the data processing and analysis module. The forming data analysis model will input the key performance indicators output in real time into the forming optimization suggestion model of the optimization suggestion and reporting module, thereby generating an optimization suggestion report.
[0011] The present invention has the following advantages:
[0012] 1. Improved efficiency: Real-time monitoring of production line status through IoT devices, automatic recording of data to blockchain, ensuring the real-time and accuracy of data, and speeding up the process of products from production to market.
[0013] 2. Cost reduction: Reduce overproduction and material waste through accurate inventory data and production monitoring, optimize the supply chain structure, and thus reduce storage and logistics costs.
[0014] 3. Enhanced decision support: Use real and reliable data guaranteed by blockchain for in-depth analysis to help companies make data-driven decisions on procurement, production scheduling and sales strategies. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a schematic diagram of the three major module architectures of the present invention;
[0016] Figure 2 It is a schematic diagram of an embodiment of the supply chain optimization for small and medium-sized enterprises manufacturing traditional leather shoes according to the present invention. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0018] Reference Figure 1 and Figure 2 The present invention provides a manufacturing supply chain management method based on blockchain, which mainly includes the following steps:
[0019] The first step is to deploy data collection equipment at key nodes of the supply chain to collect key data in key links of the supply chain in real time;
[0020] In the second step, the data collection module normalizes and stores the key data collected in real time in the supply chain, and uploads the normalized data to the data processing and analysis module;
[0021] In the third step, the data analysis algorithm model in the data processing and analysis module is trained with the historical data set of key data in the supply chain to obtain a trained data analysis model;
[0022] In the fourth step, the data collection module inputs the key data of the supply chain collected in real time into the forming data analysis model, and the forming data analysis model outputs the key performance indicators;
[0023] Step 5: The optimization suggestion algorithm model of the optimization suggestion and reporting module is trained using the historical data set of key performance indicators to obtain a trained optimization suggestion model.
[0024] Finally, the data collection module inputs the key data of the supply chain collected in real time into the forming data analysis model of the data processing and analysis module. The forming data analysis model inputs the key performance indicators output in real time into the forming optimization suggestion model of the optimization suggestion and reporting module to generate an optimization suggestion report.
[0025] Reference Figure 1 ,The overall architecture and working mode of the tool of the present invention are as follows.
[0026] 1. Data Collection Module
[0027] Specific data collection content: IoT devices (such as RFID tags, sensors, cameras) are deployed at key nodes of the supply chain (warehouses, production lines, sales points) to collect real-time image data including leather quality, thickness, color, inventory quantity, machine data on the operating status of production equipment, environmental data (such as temperature, humidity), etc.
[0028] Collection equipment and methods: Use high-resolution cameras and environmental sensors to automatically collect images and environmental data. RFID readers are used to track material flow.
[0029] Data processing: The collected data needs to be preprocessed, including data cleaning (removing outliers and filling missing values), normalization, and necessary format conversion to ensure data quality and consistency in preparation for subsequent analysis.
[0030] 2. Data processing and analysis module
[0031] Analytical algorithms and models: Use machine learning algorithms such as decision trees, random forests, and neural networks to conduct in-depth analysis of data. These models can help predict leather material demand, optimize production processes, and predict market trends.
[0032] Model training and parameters: The model is trained using historical datasets and the parameters are adjusted to achieve the best prediction accuracy. For example, cross-validation is used to optimize the number and depth of trees in a random forest.
[0033] Data input and output: Input includes real-time and historical data collected from IoT devices, and output is key performance indicators (KPIs) such as production efficiency, material utilization, and market demand forecasts.
[0034] 3. Optimization suggestions and reporting module
[0035] Report generation algorithm: Based on the output of the above analysis model, use a report generation algorithm (such as a rule-based reasoning system) to automatically generate optimization recommendations and detailed network reports.
[0036] Optimization process: The system proposes specific improvement measures based on the data analysis results, such as adjusting the raw material procurement plan, optimizing the production rhythm, or redesigning the logistics path. The report will include charts and visualizations to help decision makers intuitively understand supply chain performance and market dynamics.
[0037] Report content: The report details the reasons for the optimization recommendations and the expected results, while assessing possible risks and costs to provide a scientific basis for corporate decision-making.
[0038] Regarding data analysis, the specific process is as follows:
[0039] 1. Real-time monitoring
[0040] process:
[0041] IoT device deployment: Deploy sensors and other IoT devices in leather shoe production lines and material warehouses to capture key data such as equipment operating status, consumption of leather and other materials, and finished product inventory in real time.
[0042] Data upload and recording: These devices automatically upload the captured data to the blockchain network to ensure the security and immutability of the data and improve the credibility of the data.
[0043] Monitoring and Alerts: Real-time monitoring systems analyze data and send alerts when they detect equipment failures or abnormal inventory levels, facilitating quick responses to avoid production delays.
[0044] 2. Predictive Analytics
[0045] process:
[0046] Data collection: Use historical data collected by real-time monitoring systems as the basis for analysis.
[0047] Model building: Use machine learning methods such as time series analysis and regression models to build prediction models.
[0048] Forecast execution: Run these models periodically or on demand to predict future demand for, for example, leather and other raw materials, as well as changes in market demand.
[0049] 3. Cost Analysis
[0050] process:
[0051] Cost data summary: Collect cost data from procurement, production, logistics to storage and other links.
[0052] Cost model application: Use cost analysis models (such as activity-based costing) to deeply analyze data and identify links with excessively high costs.
[0053] Optimization measures: Based on the analysis results, specific cost reduction measures such as adjusting suppliers and improving production processes are proposed.
[0054] 4. Auxiliary tools
[0055] process:
[0056] Cloud computing platform deployment: Cooperate with third-party cloud computing platforms to carry out large-scale data storage and processing, ensuring the flexibility and scalability of data processing.
[0057] Advanced analytics software usage: Combined use of advanced analytics software such as SAP HANA or Oracle Analytics for complex data processing and analysis tasks.
[0058] Reference Figure 2 , aiming at optimizing the supply chain of traditional leather shoe manufacturing small and medium-sized enterprises.
[0059] IoT device deployment, temperature and humidity sensors in leather warehouses. The deployment method is to deploy multiple sensors in different areas of the warehouse (entrance, middle, deep) to obtain more accurate environmental data. Monitoring purpose: Monitor the storage environment of leather, such as temperature and humidity, to prevent material damage and ensure leather quality.
[0060] Automated robot sensors on production lines. Function: These robots integrate multiple sensors to not only monitor the operating status of the production line, but also detect product quality in real time, such as the accuracy of leather cutting and the uniformity of stitching. Benefit: Through real-time feedback, problems in the production process can be quickly discovered and corrected, improving production efficiency and product quality.
[0061] Trackers at the point of sale. Tracking content: Deploy visual recognition systems and mobile device tracking technology to monitor customer traffic and behavior, and analyze purchasing patterns and product preferences. Application results: Provide data support to optimize product layout, adjust marketing strategies, and improve sales efficiency.
[0062] Real-time data collection, with clear data collection points. Raw materials: From the time of storage, the quality report of each batch of leather is automatically uploaded to the system. Production line: Data such as machine running time, production quantity, failure rate, etc. are automatically recorded and uploaded by sensors. Inventory and sales: Inventory levels, product sales speed and customer feedback are synchronized in real time through the sales system.
[0063] Data upload system. Implementation method: Encrypt data and upload it to the blockchain through a secure IoT network to ensure data integrity and tamper-proof. Data circulation: Establish an internal data sharing platform for the enterprise, which is updated in real time for query and use by various departments to support quick decision-making.
[0064] Data-driven decision making, data analysis and application. Algorithmic models: Use time series analysis to predict market demand, use association rule learning to identify production bottlenecks, and optimize inventory classification through cluster analysis. Model parameters: For example, seasonal adjustment parameters in time series analysis, support and confidence thresholds for association rules, and selection of the number of clusters in clustering algorithms.
[0065] In view of the data-driven decision-making mentioned in the supply chain optimization mentioned above, the instantiation code is presented in the Jupyter Notebook environment using the Python programming language. We will focus on showing how to use time series analysis to forecast market demand, learn association rules to identify production bottlenecks, and use the K-means clustering algorithm to optimize inventory classification.
[0066] 1. Time series analysis (market demand forecast)
[0067] We will use Python’s pandas library to process the data and the statsmodels library to build the ARIMA model for demand forecasting.
[0068] Python
[0069] #Python code, environment: Jupyter Notebook
[0070] import pandas as pd
[0071] from statsmodels.tsa.arima.model import ARIMA
[0072] import matplotlib.pyplot as plt
[0073] # Sample data loading, assuming that the date and sales quantity are already available
[0074] data=pd.read_csv('sales_data.csv',parse_dates=['date'],index_col='date')
[0075] #Display data
[0076] plt.figure(figsize=(10,6))
[0077] plt.plot(data.index,data['sales'])
[0078] plt.title('Sales Over Time')
[0079] plt.xlabel('Date')
[0080] plt.ylabel('Sales')
[0081] plt.show()
[0082] #Build and fit the ARIMA model
[0083] model=ARIMA(data['sales'],order=(1,1,1))#ARIMA model parameters (p=1,d=1,q=1)
[0084] fitted_model = model.fit()
[0085] #Forecast sales for the next 12 months
[0086] forecast=fitted_model.forecast(steps=12)
[0087] # Display the prediction results
[0088] plt.figure(figsize=(10,6))
[0089] plt.plot(data.index,data['sales'],label='Historical Sales')
[0090] plt.plot(pd.date_range(data.index[-1],periods=13,freq='M')[1:],forecast,label='Forecasted Sales')
[0091] plt.title('Sales Forecast')
[0092] plt.xlabel('Date')
[0093] plt.ylabel('Sales')
[0094] plt.legend()
[0095] plt.show()
[0096] 2. Association rule learning (identifying production bottlenecks)
[0097] We will use the mlxtend library to mine association rules.
[0098] Python
[0099] #Python code, environment: Jupyter Notebook
[0100] import pandas as pd
[0101] from mlxtend.frequent_patterns import apriori,association_rules
[0102] # Loading sample data, assuming it is purchase data
[0103] data=pd.read_csv('production_data.csv')
[0104] #Convert the data into a format suitable for association rule mining
[0105] basket=(data.groupby(['order_id','product'])['quantity']
[0106] .sum().unstack().reset_index().fillna(0)
[0107] .set_index('order_id'))
[0108] #Convert quantity to 0 and 1 (purchased or not purchased)
[0109] basket_sets=basket.applymap(lambda x:1if x>0else 0)
[0110] #Use apriori algorithm to find frequent itemsets
[0111] frequent_itemsets=apriori(basket_sets,min_support=0.07,use_colnames=True)
[0112] #Generate association rules
[0113] rules=association_rules(frequent_itemsets,metric="lift",min_threshold=1)
[0114] # Display association rules
[0115] rules[['antecedents','consequents','support','confidence',
[0116] 'lift']]
[0117] 3. K-means clustering (inventory classification optimization)
[0118] We use the KMeans class of the sklearn library to perform clustering.
[0119] Python
[0120] #Python code, environment: Jupyter Notebook
[0121] from sklearn.cluster import KMeans
[0122] import pandas as pd
[0123] import matplotlib.pyplot as plt
[0124] # Loading sample data
[0125] data=pd.read_csv('inventory_data.csv')
[0126] #Select features for clustering
[0127] features=data[['stock_level','sales_rate']]
[0128] #Create KMeans model
[0129] kmeans = KMeans(n_clusters = 3) # Assume we are classified into 3 categories
[0130] clusters=kmeans.fit_predict(features)
[0131] #Add the classification results back to the original data
[0132] data['cluster'] = clusters
[0133] #Visualize the results
[0134] plt.figure(figsize=(10,6))
[0135] plt.scatter(data['stock_level'],data['sales_rate'],c=data['cluster'],cmap='viridis')
[0136] plt.title('Inventory Clustering')
[0137] plt.xlabel('Stock Level')
[0138] plt.ylabel('Sales Rate')
[0139] plt.show()
[0140] The above code snippet provides a basic framework for implementing time series analysis, association rule learning, and clustering algorithms in supply chain data analysis.
[0141] Strategy implementation and feedback mechanism. Implementation details: Adjust production plans based on analysis results, such as increasing production of hot-selling products and reducing inventory of slow-moving products. At the same time, adjust sales strategies based on market feedback, such as promotions and price adjustments. Feedback loop: Establish a continuous feedback mechanism to re-input post-sales data into the model and continuously adjust and optimize the algorithm.
[0142] This embodiment integrates blockchain and IoT technologies to help enterprises improve operational efficiency and market response speed. By deploying IoT devices at key nodes to collect data and using blockchain to ensure data security and transparency, this tool can monitor the status of the production line in real time and automatically record key data, thereby speeding up product launch, reducing resource waste, and supporting enterprises to make more accurate business decisions based on reliable data. Overall, this system is designed to help leather shoe manufacturers optimize production processes and inventory control through precise data management, ultimately reducing costs and improving overall efficiency.
[0143] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.
Claims
1. A manufacturing supply chain management method based on blockchain, characterized in that: The main steps include: The first step is to deploy data collection equipment at key nodes of the supply chain to collect key data in key links of the supply chain in real time; In the second step, the data collection module normalizes and stores the key data collected in real time in the supply chain, and uploads the normalized data to the data processing and analysis module; In the third step, the data analysis algorithm model in the data processing and analysis module is trained with the historical data set of key data in the supply chain to obtain a trained data analysis model; In the fourth step, the data collection module inputs the key data of the supply chain collected in real time into the forming data analysis model, and the forming data analysis model outputs the key performance indicators; Step 5: The optimization suggestion algorithm model of the optimization suggestion and reporting module is trained using the historical data set of key performance indicators to obtain a trained optimization suggestion model. Finally, the data collection module inputs the key data of the supply chain collected in real time into the forming data analysis model of the data processing and analysis module. The forming data analysis model inputs the key performance indicators output in real time into the forming optimization suggestion model of the optimization suggestion and reporting module to generate an optimization suggestion report.
Citation Information
Cited By
Shoe industry ERP data management and control method and system based on big data
CN120873525A
