The Future of Quality Management in Modern Manufacturing: AI-Driven
Quality — or rather, the lack of it — is a major source of manufacturing waste. ASQ shows that $861 billion is lost annually due to out-of-spec and poor-quality products. That’s essentially the same amount that consumers spent online with U.S. merchants in 2020. It’s also more than the estimated GDP for Switzerland this year. But what makes this staggering figure even more troubling is the fact that waste due to quality issues and yield losses is preventable.
Most of our customers have reached a kind of “KPI plateau,” where they might be able to make incremental improvements in quality, but it never lasts. It seems like the big wins were all harvested decades ago. Defect rates, material yield, first pass yield — they’ve hit a wall.
Manufacturers know that the answers to reducing — and ideally, predicting and avoiding — this kind of waste are there for the taking. They’re contained in the massive volumes of diverse data they collect from sensors, systems, and various other sources scattered across processes, plants, and functions. But they can’t easily get at that data, so they can make sense of it and put insights into action.
What’s the hang-up? Legacy technology that wasn’t made for a data-driven world, of course. I’m talking about the inflexible systems in manufacturing that are based on old constructs of painstaking, manual analysis, overly simplistic rules, and cumbersome user experiences.
Amplifying old systems with advanced technology that’s easy to use
Anyone reading this who is responsible for quality management is probably rolling their eyes and thinking, “Yeah, no kidding. So, what can you do about it?”
Good news: There’s a way off the KPI plateau. The answer to decreasing the overall cost of quality is to use an artificial intelligence (AI) solution that can take full advantage of all the data manufacturers collect and create value from it. That solution also needs to work seamlessly with existing systems, including manufacturing execution systems (MES) and enterprise resource planning (ERP) solutions. Because otherwise, nobody will use it.
The solution I’m describing is Quality Flow, a next-gen Quality Management System (QMS) from Noodle.ai. It helps manufacturers optimize product quality and improve profitability by using prebuilt data engines that can analyze 1) millions of parameter combinations, 2) across multiple production stages, 3) at varying levels of product and process hierarchy.
Quality Flow is powered by Explainable AI (XAI). We know engineers will never act on information they don’t understand. Black boxes are unacceptable. So, the XAI engines in Quality Flow explain to the experts using the application how Quality Flow detected a pattern and why it recommended a particular solution. The four engines specific to Quality Flow are:
- Quality Sentinel, for discovering and diagnosing a broad set of quality and yield issues and their top root causes
- Quality Precog – Defect, for predicting and simulating risk for high-value product defects
- Quality Precog – Spec, for predicting spec variability for high-value product properties
- Quality Pathfinder, for recommending actions to minimize defect risk or spec variability
Quality Flow powered by XAI was built to deal with the complexity of modern manufacturing, and truly understands quality management workflows for manufacturing. The quality management teams at leading companies who use the application have quickly learned to trust its insights and recommended actions to augment their decisions.
That trust has led to some impressive, bottom-line results. On average, Noodle.ai customers using Quality Flow have seen:
- More than 50% reduction in high-value defect rates
- 25% to 35% reduction in Total Cost of Quality (TCQ)
- Up to 35% decrease in quality variability
Real-world stories that further underscore the impact of Quality Flow
One of the best ways to understand the impact Quality Flow powered by XAI can have on your operations is to read success stories about manufacturers using the application to pinpoint costly issues and identify the best approaches to resolve them. You’ll find two case studies in our new publication, Quality Flow Unlocks Hidden Profitability in Manufacturing, which describe how:
- A top global producer of high-performance, advanced-engineered materials for semiconductor, aerospace, automotive, and other industries used Quality Flow to overcome a low production yield issue due to multiple defect types that it had been trying to solve for years. Within the first four months of using Quality Flow, the manufacturer saw a $1 million return on investment through a 40% reduction in lamination defects.
- A flexible mini mill that produces over 3 million tons of advanced steel for automotive, electrical, and other complex uses was struggling with high variability in the mechanical properties of its end products. These quality issues, like tensile strength, often led to a significant number of coils being downgraded and sold to secondary markets — or totally scrapped. Through its use of Quality Flow, this manufacturer drove 35% improvement in quality variability and is now saving millions of dollars annually.
Noodle.ai understands the complexity of modern manufacturing — and the problems your business wants to solve. We know what it takes to keep a plant running, and we understand the value of expert insight. We built Quality Flow with humility because we understand how hard it is to make good product. But we were also confident that we could make a big difference.
There’s a lot of hype in the AI space. But this isn’t hype: Quality Flow powered by XAI actually works.
To find out more about the results companies are seeing with Quality Flow powered by XAI, download a copy of Quality Flow Unlocks Hidden Profitability in Manufacturing, available now on our website.
Also, visit the Quality Flow page to discover more about this transformative solution.
- Noodle.ai’s Gartner 2021 Trifecta: Our Third Mention this Year, Now in Market Guide for Data Analytics and Intelligence Platforms in Supply Chain
- How Consumer Packaged Goods companies can improve Overall Equipment Effectiveness (OEE) with AI
- Gartner Says: AI is a Prerequisite to Supply Chain Digitization. Read Why!
- Progress is as Simple as a Bowl of Cereal: Joining Forces with Noodle.ai to Create Perfect Flow
- Deep Probabilistic Decision Learning Returns Perfect Flow to Operations