Optimization of Connecting Rod Production
Using Data Analysis for Process Improvement
Back to project overview GWiEIT
Project
This project focuses on optimizing the production of connecting rods at BMW Motorrad using advanced data analysis techniques. The goal is to enhance production efficiency and product quality by systematically analyzing process variables and outcomes.
We aim to identify correlations to reduce tool wear, prevent failures, and decrease the scrap rate.
Project Team
Deepa Selva - Intern @ Melbourne Space Program
Jared Brun - Intern @ Australian Government
Anna Labchir - Assistant to a project manager @ SAP
Elisa Sedlak - Assistant to the CIO @ Verti Versicherung
Project Results
The project achieved significant improvements in production efficiency and quality standards. The results include a comprehensive analysis of process variables, development of predictive models, and recommendations for process optimization. Key outcomes are summarized below:
- Exploratory Data Analysis: Detailed examination of historical data to identify patterns and anomalies.
- Machine Learning Models: Development of predictive models to optimize process parameters.
- Data Visualization: Implementation of advanced data visualization techniques for real-time monitoring.
- Optimization Recommendations: Practical recommendations for enhancing production efficiency and reducing variability.
Visuals
Figure 1: Force Curve Progression in real time, recorded on 07.12.2023 under the production code "W03-MF-PL-KAxX-0030-21”.
Figure 2: Feature Correlation Matrix visualizing interdependencies between each misalignment parameter.
Figure 3: Correlation Analysis Tool interface for selecting and analyzing CSV files.
Video Demonstration
