ai quality in manufacturing
From medical devices to automotive, manufacturing is using AI to gain insights that will help improve product quality and user experience.

October 28, 2019

Artificial Intelligence (AI) is fast becoming the quality control specialist in manufacturing for a variety of industries, transforming how inspections, maintenance and assembly are conducted – among other things.

In actuality, by providing more eyes and ears in the manufacturing process, AI is helping to address the growing challenges of product defects and recalls. According to the corporate insurance carrier, AGCS, “defective products not only pose a serious safety risk to the public but can also cause significant financial and reputational damage to the companies concerned. Defective product incidents have caused insured losses in excess of $2B over the past five years, making them the largest generator of liability losses.”

But, consider if you could predict which products could be defective before they leave the plant and get to the root cause of the problem. Then you could have a chance to reduce defects and prevent them from happening.

By applying predictive analytics, companies can take a proactive approach to avoid potential problems before they happen, and look at the variables and data to know with a high degree of probability which products could be troublesome. Applying AI-based predictive analytics frees up medical device manufacturers, for example, from spending 90 percent of their time focusing on quality assurance and addressing problems, so they can spend more time on strategic approaches and innovations.

AI, and more specifically machine learning, is already being used among leading medical device manufacturers to help ensure that products are defect-free before they leave the plant.

For example, a leading medical device manufacturer in Puerto Rico is using machine learning software to conduct predictive analytics, using a combination of historical and current data to identify discrepancies, variances and the smallest weakness that could cause a specific product to fail – well before it leaves the factory floor – which would be almost physically impossible using manual methods. The secret sauce, however, to this type of an algorithm is big data, which is collected during the manufacturing process and used to train the machine learning tool to spot anomalies and provide automatic self-correction.

And AI is not just improving the quality of the final products, but it’s also helping to keep the equipment that produces them in tip-top shape. AI solutions can predict when factory equipment will need maintenance, by communicating with smart sensors that can indicate if a piece of equipment requires maintenance.

Industrial IoT Data Powers Predictive Analytics

In addition to product quality applications, predictive analytics, which is driven by machine learning, holds promise in all areas of engineering and design. For example, predictive analytics can help an engineer understand how upgrades or changes to specific parts or products could impact their operation or safety.

Industrial Internet of Things (IIoT) systems, which use data gathered from sensors on devices, continue to provide valuable insights on how manufacturers can improve or change their products to keep pace with changing consumer or professional needs. When machine learning algorithms are trained on this data, product design can be enhanced to take into account actual, in-the-field experiences, as opposed to hypotheses.

Making the AI Leap

There’s no doubt that the shift to AI-driven manufacturing is a transformative step for many companies. It not only requires new skill sets and technology, but also a total realignment of company culture, and it’s fast becoming a necessity for corporate survival. So how can companies move toward AI-driven transformation on the manufacturing line? Below are three key strategies:

Focus on your data assets. Effective machine learning solutions are only as good as the data provided. Manufacturers should determine where all their data resides, what data may be lacking and work to centralize and augment it to feed to algorithms that continually learn and become smarter.

Identify the business problem. Before building a predictive model, you need to know what you are trying to solve and work backwards: Which parts or products are you evaluating? Are you trying to determine the likelihood of a device failing? Or, do you want to improve usability or the user experience? You should conduct a pilot program to determine the ROI, time savings and other benefits you could achieve before moving forward with such a transformative project.

Commit for the long haul. Unlike other digital transformation initiatives, AI is never actually completed. The key to AI success is constant training. AI solutions need to be continuously fed new data in order to become smarter and more relevant, so it’s important to retain in-house data scientists or partner with firms that can provide this ongoing expertise to continuously improve the algorithms that run your programs.

From medical device manufacturing to automotive, the manufacturing industry is under enormous pressure to improve quality and eliminate product defects, while working to innovate in an increasingly competitive global marketplace. AI is quickly taking on a major supporting role, giving manufacturers greater insights to improve product quality and address user experience, so they can spend their time innovating the next big thing.

carlos melendez wovenware
Carlos Meléndez

As COO and co-founder of artificial intelligence and software engineering services firm, Wovenware, Carlos Meléndez helps companies achieve customized digital transformation to propel their businesses to the next level. With expertise in business strategy and software engineering, Carlos has a strong track record bringing AI, machine learning and deep learning solutions to organizations across a variety of industries. More information can be found at: www.wovenware.com.