Automobile part manufacturing enterprise production leveling management and control method based on data driving

Abstract

The invention relates to the field of index data, information visualization and manufacturing industry lean management, in particular to an automobile part manufacturing enterprise production leveling management and control method based on data driving, and the method comprises the steps: building a logic model entity relation graph according to a data table structure of each data source of an automobile part manufacturing enterprise and a production business process; making a data acquisition strategy; designing a data interface of each data source; forming a data-driven management system; analyzing and processing the data of each data source; various index data related to production in the key performance index system are converted into reports by utilizing a data visualization tool, so that the problems of inventory overstock, production line imbalance and unreasonable resource allocation in the production process are avoided, the yield and the variety of products are reasonably matched, and the production efficiency is improved. And the produced products are kept balanced in time and quantity as far as possible, so that the production process of the whole enterprise is more stable.

Description

Technical Field

[0001] The present invention relates to the fields of indicator data, information visualization and lean management of manufacturing industry, and in particular to a data-driven production leveling control method for an automobile parts manufacturing enterprise. Background Art

[0002] Traditional production management models often have difficulty in responding to market fluctuations flexibly, which can easily lead to problems such as inventory backlogs, unbalanced production lines, and irrational resource allocation during the production process, which in turn affects the overall operational efficiency and profitability of the company. For example, an automobile manufacturer that produces both sedans and SUVs may focus on sedan production for a period of time and then focus on SUV production if it does not carry out leveled production. This will cause peaks and troughs in parts supply, equipment use and other links in the production process. This situation is particularly prominent in the automobile parts manufacturing industry, which has a large variety of parts and small batches.

[0003] How to avoid inventory backlogs, unbalanced production lines, and unreasonable resource allocation during the production process, reasonably match product output and variety, and keep the production of products as balanced as possible in time and quantity, so that the production process of the entire enterprise is smoother, has always been an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0004] The purpose of the present invention is to provide a data-driven production leveling control method for automobile parts manufacturing enterprises to address the corresponding deficiencies in the prior art, so as to avoid inventory backlogs, unbalanced production lines, and unreasonable resource allocation problems in the production process, reasonably match the output and variety of products, and keep the production of products balanced in time and quantity as much as possible, so that the production process of the entire enterprise is smoother.

[0005] The objective of the present invention is achieved by adopting the following scheme:

[0006] A data-driven production leveling control method for an automobile parts manufacturing enterprise comprises the following steps:

[0007] 1) Based on the data table structure of each data source of the automotive parts manufacturing enterprise and the production business process, a logical model entity relationship diagram is constructed to intuitively display the relationship between each data entity and the data flow path;

[0008] 2) Formulate data collection strategies based on the entity relationship diagram of the logical model, and design data interfaces for each data source based on the formulated data collection strategies, determine the data collection method, and form a data-driven management system;

[0009] 3) Use data-driven management systems to analyze and process data from various data sources and create data views of production business processes;

[0010] 4) Build a key performance indicator system based on the data view of the production business process, and use data visualization tools to convert the various production-related indicator data in the key performance indicator system into reports.

[0011] Preferably, the data-driven management system also includes a data early warning unit, which is used to set early warning thresholds for key indicators. Through an automated early warning mechanism, it regularly checks whether these indicators exceed a preset normal range. Once abnormal data is detected, an alarm is immediately sent to relevant business personnel to ensure real-time monitoring of production conditions and timely handling of anomalies.

[0012] Preferably, in step 1), according to the data table structure of each data source of the automobile parts manufacturing enterprise and the production business process, the specific method of constructing the logical model entity relationship diagram is as follows:

[0013] 1-1) First, sort out the business process, identify and record the key business nodes in the production business process;

[0014] 1-2) According to each key business node recorded in step 1-1), determine the corresponding data source;

[0015] 1-3) Analyze the data table structure of each data source in step 1-2) to determine the logical relationship between the data tables;

[0016] 1-4) Based on the logical associations between the data tables, the logical model entity relationship diagram corresponding to the enterprise's production business process is obtained.

[0017] Preferably, in step 2), a data collection strategy is formulated according to the entity relationship diagram of the logical model, and the data interface of each data source is designed according to the formulated data collection strategy, and the data collection method is determined to form a data-driven management system in the following specific manner:

[0018] 2-1) Set the data collection frequency threshold and formulate the data collection strategy based on the logical model entity relationship diagram;

[0019] 2-2) According to the data collection strategy formulated in step 2-1), a data warehouse for the leveling control of production of automobile parts manufacturing enterprises is established, and the data interfaces of various data sources are designed so that the data of various data sources are aggregated in the data warehouse to form a data-driven management system.

[0020] Preferably, in step 3), the specific method of using the data-driven management system to analyze and process the data of each data source and create a data view of the production business process is as follows:

[0021] 3-1) Clean the collected raw data in the data warehouse of the data-driven management system;

[0022] 3-2) Integrate the actual data of production, installation and inventory with the planned data to build a data view of the production business process.

[0023] Preferably, the dimensions of the data view include major product categories (platforms), factories, markets, production lines, and minor product categories, and the time granularity of the data view includes years, months, weeks, and days to support multi-dimensional and multi-granular data analysis needs.

[0024] Preferably, the data view includes several dimensional tables established based on product codes to ensure that each product model corresponds to its major category, factory, production line, market and minor category, which facilitates subsequent data analysis.

[0025] Preferably, in step 4), a key performance indicator system is constructed based on the data view of the production business process, and the specific method of converting various production-related indicator data in the key performance indicator system into a report using a data visualization tool is as follows:

[0026] 4-1) Design corresponding key performance indicators based on the enterprise's levelized management needs and the data view of the production business process;

[0027] 4-2) Use data visualization tools to present the annual production-related data in the key performance indicator system in a tabular form;

[0028] 4-3) Use data visualization tools to present the monthly production-related data in the key performance indicator system in a line chart based on the change trend;

[0029] 4-4) Use data visualization tools to present the distribution of different types of data in the key performance indicator system in the form of a pie chart.

[0030] The beneficial effects of the present invention are as follows:

[0031] ① The present invention combines the lean production management mode of automobile parts manufacturing enterprises, builds a data analysis model, deeply integrates the internal and external data resources of the enterprise, realizes the dynamic display and multi-dimensional analysis of production, installation, inventory, and planning, accurately and real-timely displays the current production and manufacturing status of the enterprise, optimizes resource allocation, reduces waste, improves production efficiency and response speed, thereby maintaining competitive advantages in a complex and changing market environment, and promoting the transformation and upgrading of the automobile parts manufacturing industry towards intelligence and leanness;

[0032] ② The present invention sets early warning thresholds for key indicators, establishes an automated early warning mechanism, and regularly checks whether these indicators exceed preset normal ranges. Once abnormal data is detected, an alarm is immediately sent to relevant business personnel to ensure real-time monitoring of production conditions and timely handling of abnormalities.

[0033] Glossary

[0034] Data-driven: The data-driven management concept refers to the use of data as the core decision-making basis in the process of organizational management, and the collection, analysis and use of data to guide management strategies, optimize business processes, evaluate performance, and predict future trends. For example, an e-commerce company no longer relies solely on the intuition or experience of managers to determine whether the product sales strategy is effective, but analyzes website traffic data, customer purchase behavior data, product evaluation data, etc., to accurately understand customer needs and then adjust product recommendations, pricing strategies and promotional activities.

[0035] Leveling: Under the conditions of mixed production, the production sequence of various products on the assembly line is scientifically arranged through key factors such as order demand and production scheduling. The goal is to achieve a balanced distribution of product variety, output, working hours and inventory levels, ensuring that the entire production system and finished product assembly line can maintain a stable load state and avoid efficiency loss or resource waste caused by production fluctuations. Production leveling refers to the reasonable matching of product output and variety in the case of multi-variety and small-batch production, so as to keep the production of products balanced in time and quantity as much as possible.

[0036] JDBC: Java Database Connectivity is an application programming interface (API) for the Java programming language. It provides a set of standard methods and classes that enable Java programs to easily connect and interact with various databases. You can think of JDBC as a bridge between Java programs and databases. Through it, Java programs can send SQL (Structured Query Language) statements to the database and obtain the results returned by the database.

[0037] Flink CDC: Change Data Capture is a technology based on Apache Flink that is used to capture change data in the database. It can monitor the data insertion (Insert), update (Update) and deletion (Delete) operations in the database table in real time, and output these change data in the form of streams for subsequent real-time processing, such as real-time data synchronization, incremental update of data warehouses, real-time data analysis and other application scenarios.

[0038] JPH: "Jobs Per Hour" in the production field. In industrial production and management, JPH is an important indicator for measuring production efficiency. It indicates the number of jobs that can be completed in one hour. This indicator is critical for production line planning, capacity assessment, and human resource allocation. For example, in an automobile manufacturing plant, JPH can be used to measure the efficiency of an automobile assembly line. If the JPH of a production line is 30, it means that the production line can assemble 30 cars per hour. JPH is usually calculated by dividing the total amount of work completed in a period of time by this period of time (in hours). For example, if a workshop completes the production of 240 products in 8 hours, then JPH = 240 ÷ 8 = 30.

[0039] FTT: In the production process, FTT is a key quality control indicator. It measures the proportion of products that meet quality standards in the production process, from raw materials to the first completion of all production processes. For example, on the production line of circuit boards, after a series of processes such as plug-in, welding, and testing, the percentage of the number of circuit boards that pass the first test to the number of raw material circuit boards put into production is FTT. FTT is calculated as follows: FTT = (number of products that passed the test for the first time / total number of products put into production) × 100%. This indicator helps companies to promptly discover quality bottlenecks in the production process, optimize production processes, reduce rework and waste, thereby improving production efficiency and reducing costs.

[0040] FTR: First-Time-Through Rate on Production Line. FTR usually refers to the first-time pass rate. The production line FTR is the ratio of products that complete all production processes on the production line at one time and meet the quality standards. The calculation method is to divide the number of products that pass the quality inspection for the first time by the total number of products put into production, and then multiply by 100%. For example, an electronic product production line puts into production 1,000 products in one day. After all the processes, 800 products pass the first inspection. Then the FTR of this production line on that day = (800 / 1000) × 100% = 80%. FTR is an important indicator to measure the production efficiency and quality stability of the production line. A high FTR means less waste (such as rework and scrap) in the production process, which can effectively reduce production costs. At the same time, it also reflects the maturity of the production process and the reliability of the equipment. A lower FTR indicates that there may be problems with the production process, insufficient equipment accuracy, or improper operation of personnel, and the production process needs to be analyzed and improved.

[0041] Similarly, at the workstation level, the FTR of a workstation can represent the first pass rate of the workstation. It refers to the ratio of products (or work tasks) that pass the quality inspection of this workstation at one time. For example, at an automotive parts welding workstation, workers will conduct a preliminary appearance and dimensional inspection after completing welding, and only qualified products can enter the next workstation. If the workstation processes a total of 100 parts in one day, and 80 of them pass the first inspection, then the FTR of this workstation on that day is 80%. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the process of the present invention;

[0043] Figure 2 A schematic diagram of a production business process and a data-driven management system in an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of the relationship between various business entities in a production business process in an embodiment of the present invention;

[0045] Figure 4 The figure is a flow chart of data early warning in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] like Figure 1 As shown, a data-driven production leveling control method for an automobile parts manufacturing enterprise includes the following steps:

[0047] 1) Based on the data table structure of each data source of the automotive parts manufacturing enterprise and the production business process, a logical model entity relationship diagram is constructed to intuitively display the relationship between each data entity and the data flow path. The specific method is as follows:

[0048] 1-1) First, sort out the business process, identify and record the key business nodes in the production business process;

[0049] 1-2) According to each key business node recorded in step 1-1), determine the corresponding data source;

[0050] 1-3) Analyze the data table structure of each data source in step 1-2) to determine the logical relationship between the data tables;

[0051] 1-4) Based on the logical associations between the data tables, the logical model entity relationship diagram corresponding to the enterprise's production business process is obtained.

[0052] 2) Formulate data collection strategies based on the entity relationship diagram of the logical model, and design data interfaces for each data source based on the formulated data collection strategies, determine the data collection method, and form a data-driven management system. The specific methods are as follows:

[0053] 2-1) Set the data collection frequency threshold and formulate the data collection strategy based on the logical model entity relationship diagram;

[0054] 2-2) According to the data collection strategy formulated in step 2-1), a data warehouse for the leveling control of production of automobile parts manufacturing enterprises is established, and the data interfaces of various data sources are designed so that the data of various data sources are aggregated in the data warehouse to form a data-driven management system.

[0055] 3) Use the data-driven management system to analyze and process the data from various data sources and create a data view of the production business process. The specific methods are as follows:

[0056] 3-1) Clean the collected raw data in the data warehouse of the data-driven management system;

[0057] 3-2) Integrate the actual data of production, installation and inventory with the planned data to build a data view of the production business process;

[0058] The dimensions of the data view include product categories (platforms), factories, markets, production lines, and product subcategories. The time granularity of the data view includes years, months, weeks, and days to support multi-dimensional and multi-granular data analysis needs.

[0059] The data view includes several dimension tables established based on product codes to ensure that each product model corresponds to its major category, factory, production line, market and minor category, which facilitates subsequent data analysis.

[0060] 4) Build a key performance indicator system based on the data view of the production business process, and use data visualization tools to convert the various indicators related to production in the key performance indicator system into reports. The specific methods are as follows:

[0061] 4-1) Design corresponding key performance indicators based on the enterprise's levelized management needs and the data view of the production business process;

[0062] 4-2) Use data visualization tools to present the annual production-related data in the key performance indicator system in a table form in the report;

[0063] 4-3) Use data visualization tools to present the monthly production-related data in the key performance indicator system in a line chart through the change trend in the report;

[0064] 4-4) Use data visualization tools to present the distribution of different types of data in the key performance indicator system in the form of pie charts in the report.

[0065] It is worth noting that the data-driven management system also includes a data early warning unit, which is used to set early warning thresholds for key indicators. Through an automated early warning mechanism, it regularly checks whether these indicators exceed the preset normal range. Once abnormal data is detected, an alarm is immediately sent to relevant business personnel to ensure real-time monitoring of production conditions and timely handling of anomalies.

[0066] like Figures 2 to 4 As shown, according to the above method, an embodiment is made as follows:

[0067] The production business process of an automobile parts manufacturing company includes five major contents: demand, order, production, inventory, and quality, as follows:

[0068] ① Demand forecasting: Implement accurate demand forecasting models based on historical sales data, market trends and seasonal factors to improve forecast accuracy and complete production planning based on forecast results and actual order data;

[0069] ② Order management optimization: optimize the order management process, shorten the order-to-delivery cycle, and enhance the ability to respond to changes in customer demand;

[0070] ③ Flexible production: Focus on order changes and capacity utilization, ensure that the production rhythm is consistent with market demand, avoid overproduction, and quickly adapt to changes in product lines;

[0071] ④ Inventory management: The overall warehousing and outbound inventory adopts the "first in, first out" control principle, focusing on long-term inventory products and parts to reduce inventory costs and the risk of logistics aging;

[0072] ⑤Quality control: Establish a comprehensive quality management system, implement quality monitoring methods at the factory level, production line level, workstation level, and product level, and combine indicators such as FTT and FTR to reflect the overall quality situation.

[0073] In view of the specific contents of the above production business process, a set of production leveling control methods for automobile parts manufacturing enterprises is formed as follows:

[0074] S1) Leveling data scope confirmation stage, in this stage, the complete business process of auto parts manufacturers from receiving orders to final delivery of products to customers is deeply analyzed and systematically sorted out, as follows:

[0075] S1-1) Business process analysis: systematically identify and record a series of key business nodes from order receipt, production planning, production line operation, inventory changes, logistics distribution to customer delivery, to ensure comprehensive coverage and in-depth understanding of the main line process;

[0076] S1-2) Data source definition: clearly indicate all data sources involved in the main process, including but not limited to ERP system, MES system, and any other system or database that may affect the business process, to ensure the comprehensiveness and pertinence of data collection;

[0077] S1-3) Detailed analysis of table structure and relationship analysis: Detailed analysis of the data table structure in each data source related to the production process, including field definition, data type, index setting, etc., while in-depth understanding of the logical association and business relationship between each data table, laying a solid foundation for subsequent data integration and analysis;

[0078] S1-4) Core entity identification and logical model construction: Based on a deep understanding of business processes and data sources, identify and define core entities such as orders, products, inventory, production lines, etc., and build an accurate logical model entity relationship diagram to intuitively display the relationship between entities and data flow paths, providing clear guidance for subsequent data processing and analysis.

[0079] In this way, the data scope related to the production business process of the entire automotive parts manufacturing enterprise can be accurately defined, and the data source, table structure, and relationship of the four major sections of production, planning, installation (logistics or sales), and inventory can be sorted out in combination with the main line process of the automotive parts manufacturing enterprise from order to product delivery, providing a solid foundation for subsequent data collection, processing, analysis, and application. The work at this stage is not only the prerequisite for the smooth implementation of subsequent stages, but also the key to ensuring the efficient operation of the entire leveling control system.

[0080] S2) Data collection and input stage: This stage focuses on efficient data collection and input, ensuring that all key data can be accurately captured and imported into a unified data-driven management system (including a data warehouse), as follows:

[0081] S2-1) Data collection strategy formulation: Based on business needs and data characteristics, design a reasonable data collection strategy, including but not limited to a combination of real-time data stream collection and periodic data extraction to ensure the timeliness and integrity of the data. For example, data collection includes daily collection and real-time collection, through JDBC database direct connection and FlinkCDC collection respectively;

[0082] S2-2) Data interface design and implementation: Design and implement corresponding data interfaces for different data sources, such as ERP system, MES system, etc., and use technical means such as DataX and FlinkCDC to ensure that data can be collected and transmitted to the data-driven management system in the best way.

[0083] S2-3) Data quality assurance mechanism: During the data collection process, strict data quality control measures are implemented, including data cleaning, deduplication, anomaly detection, etc., to ensure the accuracy and consistency of input data, laying a solid foundation for subsequent data processing and analysis.

[0084] In this way, detailed data of production, planning, installation and inventory can be collected from OracleERP system and MES system into data-driven management system through dataX data collection technology;

[0085] S3) Data processing stage: This stage focuses on the deep processing and intelligent processing of data, aiming to transform raw data into high-quality information that can be used for analysis and decision-making. The details are as follows:

[0086] S3-1) Data cleaning and integration: Clean the collected raw data on the data-driven management system, remove duplicates, fill in missing values, and integrate the actual data of production, installation, and inventory with the planned data to build a unified data view.

[0087] S3-2) Dimension table construction: Dimension tables are established based on product codes to ensure that each product model corresponds to its major category, factory, production line, market, and subcategory. That is, the dimension table is formed through the unique product code of each product, and each product model corresponds to a unique product major category, factory, production line, market, and product subcategory.

[0088] S3-3) Data aggregation and modeling: Use technologies such as HiveSQL and FlinkSQL to aggregate and model data at different time granularities such as year, month, week, and day, as well as multiple dimensions such as product category (platform), factory, market, production line, and product subcategory, to build a unified data view and support multi-dimensional and multi-granular data analysis needs.

[0089] In this way, according to the lean management requirements and data usage scope of the automotive parts manufacturing industry, big data technologies such as HiveSql and Flinksql can be used to process and aggregate the data in the data-driven management system according to the year, month, week, day, real-time granularity, and product categories (platforms), factories, markets, production lines, product categories, etc. When processing the overall data, the actual installation and production conditions and the planned data should be integrated into the same table;

[0090] S4) Data extraction and visualization stage: The processed and analyzed data is presented to decision makers and business personnel in an intuitive and easy-to-understand form to ensure the maximum release of data value. The details are as follows:

[0091] S4-1) Construction of the indicator system: Based on the idea of ​​enterprise level management, a series of key performance indicators (KPIs) are designed. The specific indicators and their related meanings are as follows:

[0092] Installation (sales) quantity = granularity + dimension + quantity of products delivered to vehicle manufacturers;

[0093] Production quantity = granularity + dimension + quantity of products completed in the last process;

[0094] Planned quantity = granularity + dimension + planned production quantity;

[0095] Execution difference quantity = installed (sales) quantity - planned quantity;

[0096] Actual inventory number = the number of assembly warehouses in the manufacturing department in the ERP system, including qualified product warehouses, intermediate storage warehouses, in-transit warehouses, finished product temporary storage warehouses, and assembly return warehouses;

[0097] Actual inventory number = real inventory number + actual strategic inventory number;

[0098] Inventory-to-sales ratio = actual inventory ÷ monthly estimated installed capacity;

[0099] Workstation FTT = number of qualified first-time pass / total number of first-time pass;

[0100] Production line FTT = product of FTT of each station in the production line * 100%;

[0101] Station FTR = the number of qualified products at the first pass (excluding rework) / the total number of qualified products at the first pass (excluding rework);

[0102] Production line FTR = product of FTT of each station in the production line * 100%;

[0103] Capacity utilization rate = actual production / standard capacity*100%;

[0104] Monthly standard capacity = shift output * number of shifts * number of production days per month;

[0105] Shift output = effective JPH* daily working hours;

[0106] Effective JPH = 3600 ÷ production cycle * capacity OEE.

[0107] S4-2) Visualization screen generation: Use visualization tools to convert processed data into intuitive tables, pie charts and line graphs to achieve high visualization of data. For example, annual data, installed capacity (sales), plans, inventory data, and daily data are presented in table form, monthly data trends are shown in line graphs, and the distribution of different types of data (parts production distribution type) is clearly expressed through pie charts.

[0108] In this way, according to the needs of enterprise leveling management and the concerns of middle and senior leaders, the information on the execution of installation (sales) and production can be displayed in the form of indicators from top to bottom, in accordance with the management ideas of time granularity from large to small and dimension from overall to individual. The execution of the enterprise's current production, installation (sales), plan, and inventory status can be used as important indicator data and processed through visualization tools to form highly visual indicator display forms such as tables, pie charts, and line charts.

[0109] S5) In the data early warning stage, by setting reasonable target values ​​and early warning thresholds, an automated early warning mechanism is built, such as Linux crontab-e, which polls and calls python code every hour, compares abnormal data results and pushes them to business personnel to ensure that middle and senior management and business personnel have an accurate grasp of production conditions. If abnormal data or data that does not meet the execution standards appears, an abnormal message is sent to the corresponding business personnel for analysis and processing. For example, the ideal value of the inventory-to-sales ratio is set to 0.3, and an early warning is triggered when the inventory-to-sales ratio exceeds 0.4. The daily production target quantity is set to 5,000, and a forecast is triggered when the target value is not achieved.

[0110] S5-1) Business personnel sort out the change range and threshold values ​​of each indicator according to company requirements and actual business conditions, and upload the data to the MySQL database through the front-end page.

[0111] S5-2) Based on the warning line data uploaded by the business personnel, the python code is called by using Linux's crontab-e polling method every hour, and the comparison results are recorded in the database for subsequent analysis and message push.

[0112] S5-3) Push the abnormal information in the comparison result record to the business personnel through the mobile office software, so that the middle and senior management and business can have an accurate grasp of the production situation.

[0113] In this way, the automotive parts manufacturer will be able to establish a data-driven leveled management and control system to effectively improve production efficiency, optimize resource allocation, enhance market response speed, and ultimately maximize corporate value.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification made to the present invention by those skilled in the art without departing from the spirit of the present invention shall fall within the protection scope of the present invention.

Claims

1. A data-driven production leveling control method for automobile parts manufacturing enterprises, characterized in that: The following steps are involved: 1) Based on the data table structure of each data source of the automotive parts manufacturing enterprise and the production business process, a logical model entity relationship diagram is constructed to intuitively display the relationship between each data entity and the data flow path; 2) Formulate data collection strategies based on the entity relationship diagram of the logical model, and design data interfaces for each data source based on the formulated data collection strategies, determine the data collection method, and form a data-driven management system; 3) Use data-driven management systems to analyze and process data from various data sources and create data views of production business processes; 4) Build a key performance indicator system based on the data view of the production business process, and use data visualization tools to convert the various production-related indicator data in the key performance indicator system into reports.

2. The data-driven production leveling control method for automobile parts manufacturing enterprises according to claim 1 is characterized in that: The data-driven management system also includes a data early warning unit, which is used to set early warning thresholds for key indicators. Through an automated early warning mechanism, it regularly checks whether these indicators exceed a preset normal range. Once abnormal data is detected, an alarm is immediately sent to relevant business personnel to ensure real-time monitoring of production conditions and timely handling of abnormalities.

3. The data-driven production leveling control method for automobile parts manufacturing enterprises according to claim 1 is characterized in that: In step 1), according to the data table structure of each data source of the automobile parts manufacturing enterprise and the production business process, the specific method of constructing the logical model entity relationship diagram is as follows: 1-1) First, sort out the business process, identify and record the key business nodes in the production business process; 1-2) According to each key business node recorded in step 1-1), determine the corresponding data source; 1-3) Analyze the data table structure of each data source in step 1-2) to determine the logical relationship between the data tables; 1-4) Based on the logical associations between the data tables, the logical model entity relationship diagram corresponding to the enterprise's production business process is obtained.

4. The data-driven production leveling control method for automobile parts manufacturing enterprises according to claim 1 is characterized in that: In step 2), a data collection strategy is formulated according to the entity relationship diagram of the logical model, and the data interface of each data source is designed according to the formulated data collection strategy, and the data collection method is determined. The specific method of forming a data-driven management system is as follows: 2-1) Set the data collection frequency threshold and formulate the data collection strategy based on the logical model entity relationship diagram; 2-2) According to the data collection strategy formulated in step 2-1), a data warehouse for the leveling control of production of automobile parts manufacturing enterprises is established, and the data interfaces of various data sources are designed so that the data of various data sources are aggregated in the data warehouse to form a data-driven management system.

5. The data-driven production leveling control method for automobile parts manufacturing enterprises according to claim 1 is characterized in that: In step 3), the specific method of using the data-driven management system to analyze and process the data from each data source and create a data view of the production business process is as follows: 3-1) Clean the collected raw data in the data warehouse of the data-driven management system; 3-2) Integrate the actual data of production, installation and inventory with the planned data to build a data view of the production business process.

6. The data-driven production leveling control method for automobile parts manufacturing enterprises according to claim 5 is characterized in that: The dimensions of the data view include product categories (platforms), factories, markets, production lines, and product subcategories. The time granularity of the data view includes years, months, weeks, and days to support multi-dimensional and multi-granular data analysis needs.

7. The data-driven production leveling control method for automobile parts manufacturing enterprises according to claim 5 is characterized in that: The data view includes several dimension tables established based on product codes to ensure that each product model corresponds to its major category, factory, production line, market and minor category, which facilitates subsequent data analysis.

8. The data-driven production leveling control method for automobile parts manufacturing enterprises according to claim 1 is characterized in that: In step 4), a key performance indicator system is constructed based on the data view of the production business process, and the specific method of using data visualization tools to convert various production-related indicator data in the key performance indicator system into reports is as follows: 4-1) Design corresponding key performance indicators based on the enterprise's levelized management needs and the data view of the production business process; 4-2) Use data visualization tools to present the annual production-related data in the key performance indicator system in a tabular form; 4-3) Use data visualization tools to present the monthly production-related data in the key performance indicator system in a line chart based on the change trend; 4-4) Use data visualization tools to present the distribution of different types of data in the key performance indicator system in the form of a pie chart.

Images