top of page

More Than Just a Buzzword: Top 5 Applications of AI in Manufacturing

Hype surrounding the technology of Artificial Intelligence (AI) has existed since the 1950s, when the term was first coined. Now, barely a day goes by without hearing or reading it somewhere. However, far from being just a buzzword, AI has found real applications in almost all types of business. The manufacturing sector is arguably one of the areas most ripe for AI-based disruption. In a 2018 report on behalf of the Federal Ministry for Economic Affairs and Energy (BMWi), it was predicted that the use of AI in German manufacturing firms would lead to additional gross added value of approximately EUR 31.8 billion in the years 2019-2023, which is one third of the expected total growth in the sector over this period.

Estimation of additional AI-induced growth from 2019-2023 in manufacturing
(Image courtesy of Institute for Innovation and Technology)

One reason why manufacturing holds such potential for AI applications is data. Due to Industry 4.0 and its associated technological innovations, vast amounts of data are constantly being garnered from IoT devices. According to a 2020 study by Deloitte on AI Adoption in Manufacturing, no other sector of the economy generates more data than manufacturing, with approximately 1,812 petabytes of data being produced every year. An increasing volume of data means an increasing need for data analysis. Where humans reach their limit, AI can lend a helping hand.

Annual data creation by industry
(Image courtesy of GP Bullhound, Deloitte Research)

By analysing data in an efficient and objective way, machine-learning methods can be very useful in the improvement of various manufacturing processes. As seen in the graphic below, maintenance seems to be the most prevalent use case, with quality coming in a close second. In this article, innovative technologies will be presented which can facilitate the application of AI in maintenance and quality, as well as in areas like worker safety, product engineering and process optimisation.

Share of use cases of AI in manufacturing

1. Worker Safety

Keeping workers free from harm is of utmost importance in the production environment but never has this been truer than now, with the ongoing threat of Covid-19. With AI-powered video analytics from Intenseye, a New York based start-up, ensuring the health and safety of workers becomes a bit easier. There is no need for external hardware; you simply connect your existing IP cameras with the software and input your safety rules. When rules are violated, real-time notifications are received.

Factory worker with AI-based safety software
(Image courtesy of Intenseye)

The machine-learning based software detects strange movements, danger zone violations and use of personal protective equipment, thereby preventing accidents from occurring. For those concerned about privacy issues, the company promises that the software does not rely on facial recognition, as the algorithm uses visual cues instead, and any notifications received are anonymous. In response to the pandemic, their software can now also detect whether workers are wearing face masks and maintaining minimum-distance guidelines.

2. Generative Design

Another process which is being simplified by AI is design. In generative design, powerful AI algorithms autonomously create hundreds of optimal designs when fed parameters such as size, weight, cost, preferred materials, and manufacturing process. The optimal solution can thus be reached at a significantly higher speed. Combined with additive manufacturing, the possibilities for time-saving are endless. This makes product differentiation much more feasible. Another advantage is that even engineers lacking experience are able to produce high-quality, ready-to-use designs with the help of this technology.

Creo Generative Design software screenshot
(Image courtesy of PTC)

Experienced engineers are set to benefit from this AI-based innovation too. Jacobs Engineering Group used Creo Generative Design by PTC to design life support backpacks for NASA astronauts. As the backpacks need to be robust yet lightweight, finding an optimal design can be a challenge. An innovative form of 3D CAD software like Creo Generative Design streamlines the design process and leads to better outcomes, especially because the algorithm is not likely to make biased decisions based on aesthetics, as an engineer might.

3. Process Optimisation

Another area in which human bias can lead to less than optimal solutions is process analysis. Aachen-based start-up IconPro are using their expertise in data science and AI to deliver innovative solutions for process optimisation. With machine-learning algorithms, their software analyses data from MES (Manufacturing Execution Systems) or ERP (Enterprise Resource Planning) systems in a thorough, unbiased manner.

Data analysis
(Image courtesy of IconPro)

AI-powered mining of production process data can enable rapid derivation of key trends and correlations between processes and quality. This helps to detect bottlenecks, minimise scrap, reduce throughput times and improve process flows. Superior to conventional rule-based algorithms, AI-based solutions offer efficient analysis of the growing volume of data present in production environments.

4. Predictive Maintenance

Many studies, including Capgemini's recent report on AI in manufacturing, have cited maintenance as the most suitable use case of AI. Machine-learning algorithms facilitate a novel kind of automated maintenance, namely predictive maintenance. Through these algorithms, standard machine behaviour can be learned and abnormal behaviour automatically detected, so that critical issues are pinpointed before failure occurs. Senseye, based in the UK, is a leader in providing cloud-based predictive maintenance software.

Senseye predictive maintenance software in factory
(Image courtesy of Senseye)

Their software, Senseye PdM, improves over time as it learns about machine behaviour and generates accurate estimations of Remaining Useful Life. Benefits include a significant reduction in machine downtime and in maintenance costs. The new “Attention Index” feature even prioritises assets according to importance, so that you are not overwhelmed by notifications. Such is the success of Senseye that its software has been integrated into Mindsphere, Siemens’ cloud-based, IoT operating system.

5. Quality Control

Manual visual inspections are costly, and due to the high level of concentration required, human error is rife. Innovative, AI-based technologies for optical quality control in production can revolutionise how defect inspection is carried out. California-based software company, Landing AI was founded in 2017 by Andrew Ng, a well-known figure in the AI sphere. With their software solution, it is possible to automate time-consuming visual inspection. AI-powered vision detection relieves workers of repetitive defect-inspection tasks, speeds up the inspection process and improves overall product quality.

As we have seen, the applications of AI in manufacturing are numerous, and harnessing its analytical power can lead to a huge boost in productivity. In the next few years, it is likely that more and more manufacturing firms will lock into its potential, rendering solutions like those described here essential for sustaining a competitive advantage.

Still not sure how the latest AI technologies or other Industrie 4.0 innovations could increase productivity in your production environment? If so, feel free to send us a message at with details of your problem. Drawing on our industry expertise and AI-enhanced search methods, we will conduct extensive research and suggest the best technology solutions tailored to your specific situation.


Tech Bites News

Vielen Dank für Ihr Interesse

Sind Sie auf der Suche nach passenden Lösungen?

Wir unterstützen Sie gerne.

Thomas (002).jpg



Thomas Kinkeldei


bottom of page