Production Data Quality Indicator for In-Line Product Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current quality control methods in manufacturing, particularly for products like printed circuit boards, are complex, costly, and often require high safety standards, including the use of X-rays, which can be disadvantageous and costly.
Innovation Solution
A method that calculates a quality indicator based on production data using sensors and an adaptive algorithm, allowing for partial or complete omission of quality checks post-production, with the quality indicator being assigned to the product for classification and potential use of QR codes, barcodes, or RFID chips for tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If quality control is performed using imaging techniques (including X-rays) after production, then product quality can be determined, but the complexity and cost of the manufacturing process increases significantly
Solution Approach 1:
The patent applies preliminary action by calculating the quality indicator during the production process itself, rather than performing quality control after production. The quality indicator is computed from production data obtained during manufacturing steps, enabling quality determination before the product leaves the production line, thereby avoiding complex post-production imaging equipment
Solution Approach 2:
The patent substitutes mechanical/imaging-based quality control systems with a computational approach. Instead of using physical imaging devices (including X-ray systems) to inspect products, the invention uses a computing unit that processes production data through algorithms to determine quality indicators, replacing complex mechanical inspection systems with software-based evaluation
2Reliability
If X-ray imaging is used for quality control, then internal product quality can be assessed, but the cost and safety requirements increase significantly
Solution Approach 1:
The patent replaces harmful X-ray imaging with a non-invasive computational method. Production data collected during normal manufacturing processes is processed by algorithms to assess quality without exposing products or operators to ionizing radiation, eliminating the harmful effects of X-rays while maintaining quality assessment capability
Solution Approach 2:
The patent introduces production data as an intermediary that indirectly reveals product quality. Instead of directly imaging the product (which requires X-rays), the system uses production parameters and process data as intermediaries to infer quality characteristics, avoiding direct exposure to harmful radiation while still assessing internal quality
3Reliability
If comprehensive quality control inspections are performed after production, then defective products can be identified, but manufacturing time and productivity decrease
Solution Approach 1:
The patent performs quality determination preliminarily during production rather than after completion. By calculating the quality indicator from production data as manufacturing progresses, the system identifies potential defects in real-time without requiring separate post-production inspection steps, thereby maintaining manufacturing throughput while ensuring quality
Solution Approach 2:
The patent maintains continuous quality assessment throughout the production process by continuously calculating quality indicators from ongoing production data. This continuous evaluation replaces discontinuous post-production inspections, ensuring quality monitoring occurs without interrupting the manufacturing flow, thus preserving productivity while maintaining defect detection capability
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method and a device for producing a product (P) and to a computer program product. The product (P) is produced in at least one production step (P1, P2, P3). A quality control check (QM) is optionally carried out after at least one of the production steps (P1, P2, P3) to determine a quality index (QX) of the product (P) in question. To save on the quality control check (QM), a quality indicator (QI) of the product (P) in question is determined using production data (x1, ..., xn). The production data (x1, ..., xn) are advantageously provided by sensors (S1, S2, S3). The quality indicator (QI) of the product (P) in question is preferably calculated using an adaptive algorithm (A). The adaptive algorithm (A) can be taught and/or improved using quality indices (QX) of a quality control unit (QME) and the corresponding production data (x1, ..., xn). The adaptive algorithm (A) is preferably taught with the aid of a further computing unit (CL), in particular in a cloud.