Bringing New Life to Preventive Maintenance in Additive Manufacturing

Can predictive maintenance strategies help prevent failure in additive manufacturing?

April Miller, managing editor of design technology

July 12, 2022

4 Min Read
preventive maintenance
Image courtesy of Illia Uriadnikov / Alamy Stock Photo

Preventive maintenance can help manufacturers extend the lifespan of additive manufacturing machinery and the parts that they manufacture. However, the required cost and time commitment can make it challenging to implement and practice correctly.

New artificial intelligence (AI) platforms can help businesses streamline preventive maintenance, making their strategies more efficient and easier to manage—or, in some cases, enabling predictive maintenance strategies that help businesses see failure coming.

How AI Can Streamline Preventive Maintenance

Businesses have access to more data than ever, including information on machine health, performance, and maintenance status.

Built-in machine telematics systems and IoT sensor fleets can gather vast amounts of data on parameters like machine temperature, timing, ultrasonic vibrations, and lubrication. In theory, this information could make preventive maintenance much easier. An IT or data science team could find direct correlations between certain operational parameters or events and machine health with the right analysis.

For example, it may be obvious that certain temperatures or vibration patterns signal a failing component that may cause cascading issues as it breaks down. For additive manufacturing equipment that requires careful temperature management, like a 3D printer, excessive heat may be a particularly valuable warning sign.

However, managing and applying this data can be difficult, even with the help of skilled data scientists.

New AI platforms enable businesses to analyze the massive amounts of operational data they can capture from their equipment and vehicle using technology like IoT sensors. They may use analyzed data to provide early notice on potential failures, more accurately identify possible issues and develop more-effective maintenance schedules.

The right platform enables technicians to respond to problems faster and easily identify the underlying causes of malfunctioning equipment. As a result, businesses may be able to reduce maintenance costs while also decreasing downtime and increasing equipment life span.

For example, manufacturers can use IoT devices to gather information on 3D printer performance—like temperature, vibration, or the speed of the 3D printer’s nozzle.

The AI platform can analyze information from these sensors, notifying managers immediately if sensor data suggests the printer is performing unusually or on the verge of failure. With this advance notice, technicians and managers may be able to more quickly respond to issues with additive manufacturing equipment.

Emerging Applications for Artificial Intelligence in Manufacturing

Similar AI-based technology can be adopted for various other purposes, including document automation, supply chain management, and manufacturing workflow optimization.

Businesses of all sizes can struggle with the data they collect, even when they know which KPIs are most important for some of these processes, like freight cost savings or shipment volume in supply chain management.

In part due to data disorganization, the average worker spends a significant amount of time and energy looking for things they need to perform essential tasks. These things may include resources like blueprints, design documents, and important metrics.

New AI tools can make data management and analysis much easier. They can reduce the labor needed to manage data and make it easier to uncover insights that can’t be found with conventional analytic techniques, making the technology a crucial part of business operations and allowing it the potential to add around $13 trillion to the global economy.

Predictive Maintenance and Going Beyond Preventive Upkeep

AI can also enable predictive maintenance forecasting algorithms that can see failure coming based on current machine performance. Businesses that use information from IoT devices and other sources can more effectively know when machines will fail and what steps they can take to maximize machine lifespan.

Over time, using a baseline built from historical operating data, the algorithm can predict future machine performance and failure. Predictive maintenance can significantly reduce the cost of machine ownership and prevent downtime.

The AI algorithm can make more accurate predictions with more data. This means predictive maintenance algorithms may become even more accurate over time, especially with updates to the underlying AI platform from its developers.

In addition to tracking the performance of their machinery, manufacturers can also use AI to predict the performance of parts created with additive manufacturing. In 2020, for example, U.S. Army researchers were able to develop a sensor that could detect when 3D-printed parts were beginning to fail, allowing for prompt replacement or maintenance.

By measuring the performance and degradation of additively manufactured parts, it may also be possible to gather data that manufacturers can use to improve production methods and inform design decisions like material choices.

Using AI to Revolutionize Preventive Maintenance in Additive Manufacturing

Manufacturers are under more pressure than ever to keep production lines running and boost productivity. Preventive maintenance is essential but requires a recurring investment of time and money.

New AI tools can help streamline preventive maintenance and even lay the foundation for a predictive maintenance strategy. Some tools can also help manufacturers monitor the performance of 3D printed goods—helping them anticipate failure and improve the quality of their products.

About the Author(s)

April Miller

managing editor of design technology, ReHack magazine 

April Miller is a managing editor of design technology at ReHack magazine as well as a contributing writer at sites such as Open Data Science and the Society of Women Engineers.

Sign up for the Design News Daily newsletter.

You May Also Like