Bottle Product Guidance Using Preform-to-Bottle Quality Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing stretch blow molding and filling systems face challenges in efficiently evaluating preform defects and machine errors, leading to difficulties in optimizing production quality and minimizing downtimes, as current methods are time-consuming and do not effectively correlate preform properties with bottle outcomes.
Innovation Solution
A method for product guidance in stretch blow molding and filling systems that involves automated measurement and data analysis of preform and bottle parameters, using machine learning to calculate lead-out criteria for faulty preforms and bottles, allowing for real-time optimization of production processes and machine performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If automated measurement and data analysis of preform parameters is implemented, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The system divides the complex evaluation process into distinct modules: initial data acquisition from multiple sensors, intermediate data processing through machine learning models, and final lead-out decisions. This segmentation allows each module to be optimized independently while maintaining overall system precision.
Solution Approach 2:
A control unit acts as an intermediary between the measurement systems and the stretch blow molding machine. This intermediary processes the raw sensor data, applies machine learning algorithms, and translates complex data patterns into actionable lead-out decisions, simplifying the overall system architecture.
2Productivity
If lead-out criteria are continuously updated during production, then productivity is improved, but loss of time increases due to data processing
Solution Approach 1:
The system pre-processes and stores measurement data in structured formats during production, preparing it for rapid analysis. Machine learning models are trained in advance on historical data, enabling quick generation of updated lead-out criteria without significant processing delays during active production.
Solution Approach 2:
The machine learning model continuously learns from incoming production data in real-time, updating lead-out criteria without interrupting the production flow. This continuous learning process allows the system to adapt to changing conditions while maintaining uninterrupted manufacturing operations.
3Manufacturing precision
If comprehensive data analysis of preform properties is performed, then manufacturing precision is improved, but loss of information increases due to data complexity
Solution Approach 1:
The system extracts and focuses on the most critical preform parameters that have the strongest correlation with bottle quality outcomes. By identifying and prioritizing key parameters through initial data analysis, the system manages data complexity while maintaining high manufacturing precision for the most influential factors.
Solution Approach 2:
Different levels of data analysis are applied to different parameters based on their importance. Critical parameters receive more detailed analysis and stricter quality thresholds, while less important parameters use simplified evaluation criteria, optimizing both precision and data manageability.
Data Source
AI summary
A method for product guidance in a stretch blow molding and/or filling system for bottles, and a corresponding production system are described. Preform parameters of preforms provided for stretch blow molding are measured in an automated manner and initial data acquired in the process is stored. Bottle parameters of the blown empty bottles and/or subsequently filled bottles are measured and machine error states are optionally detected. Results data acquired in the process are individually associated with the preforms and stored. At least one lead-out criterion applicable in the downstream production operation is calculated for deciding whether or not to lead out faulty preforms or bottles, based on the initial data and results data. The lead-out criterion is additionally updated in an automated manner while taking into consideration the acquired initial data and results data.