Dealing with quality improvement is a regular part of our business operations, often resulting indirectly from introducing enterprise automation. Companies typically aim to minimize losses, especially those tied to quality issues. This ongoing focus on quality management reflects its relevance and significance. In this article, we’ll delve into why quality management matters and how data enhances our improvement strategies.
Quality assurance has been and today still is an independent branch in many manufacturing enterprises. The separation of production management and quality assurance that has existed historically has frequently led to an uneven system environment. Separate message dialogs, like those for in-production data collecting and inspection, are the result. What should also be prevented is the user being unnecessarily confronted with two systems, especially as production data are also quality data.
Traditional Approach to Quality Control
Quality control methods have changed over the 20th century, initially rooted in statistics. But in the 1980s, due to global competition and customer demands, a new view on quality control emerged, which led to the introduction of approaches like statistical process control, Six Sigma, ISO 9000, and more.
Traditional Quality Control (QC) primarily focused on inspection, mostly led by the inspection department in industries. They used statistical techniques like sampling and statistical quality control (SQC) to ensure quality. SQC involves checking a sample of manufactured goods to understand the quality of the entire batch.
There are two main methods used in SQC:
– Control charts;
– Acceptance sampling.
Control charts track statistics related to different attributes of parts or products over time. They help determine if a process is running normally or not.
Acceptance sampling decides whether to accept or reject a batch based on a sample’s quality. It’s used for checking raw materials, deciding on shipments, and inspecting parts during manufacturing.
While still functional, we find traditional QC approaches to be too conservative and notably limited in scope, primarily relying on manual procedures and lacking thorough statistical analysis. Statistical sampling risks include delivering defective items to customers and overlooking faults during inspection, which applies to both control charts and acceptance sampling. Recently, there has been a shift towards 100% automated inspection to meet the demand for guaranteed quality.
It’s worth noting that Total Quality Management has been around for a while and deserves a closer look on its own. At the same time, there’s an increasing need to present information from different parts of the company in a clear, organized manner and to establish functional links between them. Simply having access to high-quality data isn’t enough anymore; tracking quality data requires having access to all data. This applies not only to quality managers but to all departments, from purchasing and development to customer service staff, who should all use a unified database.
Production Data and Traceability
A modern approach to quality control prioritizes the use of production data, reflecting the industry’s shift towards data-driven decision-making. It refers to information gathered at every stage of the manufacturing process, facilitating product traceability within the supply chain. This comprehensive data collection involves storing manufacturing details in a central database and closely monitoring products as they progress through each production step. This data serves various purposes, such as reducing costs linked to product recalls, maintaining quality standards, and enhancing overall production efficiency.
Data collection methods can vary between manual and automated processes, differing in technology but not in principle. Whether through machines or manual input, the key is to ensure convenient interfaces, effective visualization, and timely notifications.
Manual data collection involves:
- Reading part markings using handheld or fixed scanning devices;
- Operators confirming steps or completion manually;
- Using handheld tools equipped with data collection technologies for inputting data.
Automated data collection includes:
- Direct communication with the existing machines and devices by the means of the relevant interfaces;
- Comprehensive integration with the 3rd-party Applications, used for various purposes;
- Integration with the specific fail-safe controlling modules.
In summary, data plays a crucial role in manufacturing operations, with production data serving as the foundation for maintaining quality standards throughout processes. By capturing information at every production stage, production data enables comprehensive traceability and analysis, essential for maintaining and improving product quality standards. Through efficient tracking of all operational aspects, you can prevent issues such as unplanned downtimes, human errors, quality losses, equipment breakdowns, and unexpected material shortages.
Enhancing Quality Control with AI
Artificial Intelligence (AI) is bringing about a major revolution in industrial inspection, significantly enhancing efficiency through AI data processing.
In manufacturing, AI utilizes data alongside machine learning and deep learning algorithms to automate processes and enhance the speed and accuracy of operations. The key processes involve developing and testing AI models, gathering and pre-processing industrial data, and implementing them in production. These algorithms are then integrated into a variety of applications aimed at improving manufacturing procedures, enhancing product quality, and optimizing overall operational efficiency.
Today, AI offers great potential in addressing critical manufacturing challenges, such as:
– Demand Forecasting;
– Waste Reduction;
– Product Development;
– Efficiency Loss Forecasting;
– Predictive and Prescriptive Maintenance.
Undoubtedly, quality control can greatly benefit from Artificial Intelligence. AI systems continuously track manufacturing data, alerting us to quality issues and providing analysis for ongoing improvements. The future of quality management lies in predictive analytics empowered by AI. The ability of AI to learn and adapt over time is a key advantage. By merging data from different sources and considering specific variations, AI models can accurately predict and identify potential quality problems.
In conclusion, we are witnessing a paradigm shift in large-scale companies due to the revolutionary potential of AI data processing. In addition to improving precision and productivity, AI is driving the evolution of smart manufacturing and predictive maintenance, while also advancing comprehensive quality assurance measures. Furthermore, MES are set to play an important role in the broader digital transformation journey, particularly in optimizing maintenance processes and implementing Statistical Process Control for top-tier quality assurance.
