A machine learning-based MBOM material consumption index monitoring method, system, medium and device

Abstract

This invention discloses a method, system, medium, and equipment for monitoring MBOM material consumption indicators based on machine learning, belonging to the technical field of intelligent manufacturing in the aerospace industry. It acquires MBOM material consumption data, classifies the data by material type, and divides it based on responsible units within a time period to form a consumption dataset. Based on the consumption dataset, a linear regression model is trained to predict the estimated completion time for each responsible unit for different material types. For the original unconsumed data, the linear regression model predicts material consumption time based on the responsible unit, material type, and material quantity. Then, the material consumption completion time is evaluated in conjunction with the EBOM release time. By comparing the evaluated material consumption completion time with the required completion time, the MBOM material consumption indicators are analyzed. This invention can accurately identify the material consumption indicator rate, effectively improving the completeness of material consumption indicator monitoring and possessing good practicality.

Description

Technical Field

[0001] This invention belongs to the technical field of intelligent manufacturing in aviation, specifically relating to a method, system, medium, and equipment for monitoring MBOM material consumption indicators based on machine learning. Background Technology

[0002] In the aerospace manufacturing industry, identifying discrepancies between the MBOM and EBOM to ensure that all materials on the EBOM have been consumed by the MBOM has become a standard practice. However, identifying discrepancies and directly using them as monitoring results is too simplistic and fails to consider the dimension that "although there is a discrepancy, the relevant responsible unit can complete the consumption within the specified time." That is, if the time difference between the EBOM data release time and the estimated consumption time is within a specified range, this data should not be marked as abnormal. For a single aircraft, this method can currently support production, but if multiple different types and flights are managed simultaneously, this assessment method is too simplistic and cannot truly reflect the actual situation of MBOM material consumption. Therefore, how to more accurately monitor MBOM material consumption indicators based on digital means is of considerable importance. Summary of the Invention

[0003] The purpose of this invention is to provide a method, system, medium, and device for monitoring MBOM material consumption indicators based on machine learning, in order to solve the above-mentioned problems.

[0004] This invention is mainly achieved through the following technical solutions:

[0005] A machine learning-based method for monitoring MBOM (Material Business Components) material consumption indicators includes the following steps:

[0006] Step S1: Obtain MBOM material consumption data;

[0007] Step S2: Classify the consumption data according to material type, and divide it based on the responsible unit in combination with the time period to form a consumption dataset;

[0008] Step S3: Based on the consumption dataset, train a linear regression model to predict the expected completion time of each responsible unit for different material types;

[0009] Step S4: Conduct material consumption monitoring and calculate consumption indicators;

[0010] For the original unconsumed data, a linear regression model is used to predict the material consumption time based on the responsible unit, material type, and material quantity. Then, the material consumption completion time is evaluated in conjunction with the EBOM release time. The evaluated material consumption completion time is compared with the required completion time to analyze the MBOM material consumption index.

[0011] To better realize the present invention, in step S1, the MBOM material consumption data includes material number, responsible unit, material release time and completion consumption time.

[0012] To better realize the present invention, step S2 further includes the following steps:

[0013] Step S21: Identify the material type and classify the consumption data into three categories: standard parts, components, and finished products;

[0014] Step S22: For each material, calculate the difference between the completion consumption time and the material release time, and use it as the material consumption time; Step S23: Set a time period based on the material release time; count the quantity of materials by material type, and calculate the average material consumption time for each material type within the set time period based on the responsible unit, forming a consumption dataset.

[0015] To better realize the present invention, step S3 further includes the following steps:

[0016] Step S31: Divide the consumed dataset into a training dataset and a test dataset;

[0017] Step S32: Train a linear regression model based on the training dataset;

[0018] Step S33: Evaluate the linear regression model using the test dataset and perform model parameter tuning to obtain the optimal linear regression model.

[0019] To better realize the present invention, step S32 further includes the following steps:

[0020] Step A1: Preprocess the training dataset by converting text or string data into numeric data;

[0021] Step A2: Use the average material consumption time as the output value Y and other information as the input value X. Build a CART regression tree model based on the sklearn library in Python and complete the model training.

[0022] To better realize the present invention, in step S4, if the time for the assessed material consumption to be completed is before the required completion time, then the MBOM material consumption index is normal.

[0023] This invention is mainly achieved through the following technical solutions:

[0024] A machine learning-based MBOM material consumption index monitoring system is used to implement the aforementioned machine learning-based MBOM material consumption index monitoring method, including a data acquisition and processing module, a model training module, a consumption time prediction module, and an index analysis module.

[0025] The data acquisition and processing module is used to acquire MBOM material consumption data and process it to form a consumption dataset;

[0026] The model training module is used to form a linear regression model based on the consumed dataset;

[0027] The consumption time prediction module is used to predict the material consumption time based on the responsible unit, material type, and material quantity using a linear regression model.

[0028] The indicator analysis module is used to analyze MBOM material consumption indicators based on the predicted material consumption time and EBOM release time, compared with the required completion time.

[0029] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned machine learning-based MBOM material consumption index monitoring method.

[0030] An electronic device includes a memory and a processor; the memory stores a computer program; the processor is configured to execute the computer program in the memory to implement the above-described machine learning-based MBOM material consumption index monitoring method.

[0031] The beneficial effects of this invention are as follows:

[0032] This invention uses a linear regression model to predict the expected completion time of each responsible unit for different types of materials. By combining the data release time and the required completion time, it further accurately identifies the material consumption index rate, improves the completeness of material consumption index monitoring, and has good practicality. Attached Figure Description

[0033] Figure 1 This is a flowchart of the MBOM material consumption index monitoring method based on machine learning according to the present invention. Detailed Implementation

[0034] Example 1:

[0035] A machine learning-based method for monitoring MBOM material consumption indicators, such as Figure 1 As shown, it includes the following steps:

[0036] Step 1: Obtain MBOM material consumption data;

[0037] For example, as shown in Table 1, the MBOM material consumption data of each relevant responsible department is obtained, and it is the basic data that can be directly exported from each information system.

[0038] Table 1 MBOM Material Consumption Data

[0039]

[0040] Step 2: Classify the consumption data by material type and divide it based on the responsible unit according to the time period to form a consumption dataset;

[0041] 1) As shown in Table 2, identify the material type, which can be divided into three categories: standard parts, components, and finished products.

[0042] Table 2 Consumption data after identifying material types

[0043]

[0044] 2) As shown in Table 3, calculate the consumption time for each material, i.e., calculate the value of "completion consumption time - material release time".

[0045] Table 3 Consumption data after identifying material consumption time.

[0046]

[0047] 3) Based on the material release time, take 3 days as a time period, count the quantity of materials by material type, and calculate the average time of material consumption for each responsible unit during that time period. Each row of data in the table represents the average time of consumption of a certain material type by a certain responsible unit over 3 days.

[0048] Table 4 Average Time of Material Consumption

[0049]

[0050] Step 3: For the consumed dataset, divide 80% of the data into the training dataset and use the remaining 20% ​​as the test dataset;

[0051] Step 4: Based on the training dataset, train a linear regression model to predict the expected completion time of each responsible unit under different material types;

[0052] 1) As shown in Table 5, the dataset is preprocessed to convert text or string data into numeric data.

[0053] Table 5 Average consumption time of pretreated materials

[0054]

[0055] 2) Using material consumption time as the output value Y and other information as the input value X, construct a CART regression tree model based on the sklearn library in Python and complete the model training.

[0056] Step 5: Evaluate the linear regression model for the test set and tune the model parameters to improve the model's prediction accuracy.

[0057] Step 6: When conducting material consumption monitoring, for the original unconsumed data, use the model to predict the material consumption time based on the responsible unit, material type, and material quantity. Then, combine the EBOM release time to evaluate the material consumption completion time. If the time is before the expected completion time, the data can be considered normal.

[0058] The timeframe for completion after the EBOM data is published (i.e., the required completion time) is determined by each manufacturing unit. The EBOM data publication time can be directly obtained from the statistics of each unit's information platform, and the process of calculating the estimated completion time can be represented as a linear regression process.

[0059] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications or equivalent changes made to the above embodiments based on the technical essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for monitoring MBOM material consumption indicators based on machine learning, characterized in that, Includes the following steps: Step S1: Obtain MBOM material consumption data; Step S2: Classify the consumption data according to material type, and divide it based on the responsible unit in combination with the time period to form a consumption dataset; Step S3: Based on the consumption dataset, train a linear regression model to predict the expected completion time of each responsible unit for different material types; Step S4: Conduct material consumption monitoring and calculate consumption indicators; For the original unconsumed data, a linear regression model is used to predict the material consumption time based on the responsible unit, material type, and material quantity; then, the material consumption completion time is evaluated in conjunction with the EBOM release time; the evaluated material consumption completion time is compared with the required completion time to analyze the MBOM material consumption index. In step S4, if the assessed material consumption completion time is before the required completion time, then the MBOM material consumption index is normal.

2. The method for monitoring MBOM material consumption indicators based on machine learning according to claim 1, characterized in that, In step S1, the MBOM material consumption data includes the material number, responsible unit, material release time, and completion consumption time.

3. The method for monitoring MBOM material consumption indicators based on machine learning according to claim 2, characterized in that, Step S2 includes the following steps: Step S21: Identify the material type and classify the consumption data into three categories: standard parts, components, and finished products; Step S22: For each material, calculate the difference between the completion consumption time and the material release time, and use it as the material consumption time; Step S23: Set a time period based on the material release time; count the quantity of materials by material type, and calculate the average material consumption time for each material type within the set time period based on the responsible unit, forming a consumption dataset.

4. The method for monitoring MBOM material consumption indicators based on machine learning according to claim 2, characterized in that, Step S3 includes the following steps: Step S31: Divide the consumed dataset into a training dataset and a test dataset; Step S32: Train a linear regression model based on the training dataset; Step S33: Evaluate the linear regression model using the test dataset and perform model parameter tuning to obtain the optimal linear regression model.

5. The method for monitoring MBOM material consumption indicators based on machine learning according to claim 4, characterized in that, Step S32 includes the following steps: Step A1: Preprocess the training dataset by converting text or string data into numeric data; Step A2: Use the average material consumption time as the output value Y and other information as the input value X. Build a CART regression tree model based on the sklearn library in Python and complete the model training.

6. A machine learning-based MBOM material consumption index monitoring system, used to implement the machine learning-based MBOM material consumption index monitoring method according to any one of claims 1-5, characterized in that, It includes a data acquisition and processing module, a model training module, a time consumption prediction module, and an indicator analysis module; The data acquisition and processing module is used to acquire MBOM material consumption data and process it to form a consumption dataset; The model training module is used to form a linear regression model based on the consumed dataset; The consumption time prediction module is used to predict the material consumption time based on the responsible unit, material type, and material quantity using a linear regression model. The indicator analysis module is used to analyze MBOM material consumption indicators based on the predicted material consumption time and EBOM release time, compared with the required completion time.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the MBOM material consumption index monitoring method based on machine learning as described in any one of claims 1-5.

8. An electronic device, characterized in that, It includes a memory and a processor; the memory stores a computer program; the processor is used to execute the computer program in the memory to implement the MBOM material consumption index monitoring method based on machine learning according to any one of claims 1-5.

Images