IoT has made it easier than ever for machines to communicate with each other and their human counterparts. The Industry 4.0 revolution is an enabler of this communication, as it allows the data generated by these connected devices to be captured remotely so that trends can accurately reflect what’s happening in your facility or plant.
Maintenance is an important part of keeping your equipment running smoothly. With sensors, manufacturers and companies are able to eliminate unplanned downtime by implementing AI-based device algorithms that ensure the health and safety for each machine in their fleet. This is called predictive maintenance, and it includes functions like reducing downtime and breakdowns of machines by sensing abnormal vibrations, out-of-spec voltage draws, unusual torque on motors, and many other details that are often otherwise invisible and undetectable, causing inefficiencies or eventual machinery breakage.
Machine Learning at the Edge
Edge machine learning is a powerful tool that can be used to derive valuable insights from sensor data that is otherwise discarded or underutilized from traditional application. There are many different types of machine learning and signal processing algorithms, each with their own strengths and weaknesses. For example, a condition monitoring algorithm accesses the machine’s current condition to detect and diagnose faults in the machine whereas a prognostics algorithm forecasts when a failure will happen based on the current and past state of the equipment and sensor.
Engineers with domain expertise in industrial maintenance can use Edge Impulse Studio to select the appropriate combination of algorithms for the specific maintenance problem they are trying to solve, combining their own insight with highly capable machine learning algorithms.
The five human senses can be mimicked by digital sensors, allowing for predictive maintenance to improve machine efficiency and decrease downtime. Edge Impulse’s platform works with all these types of data and use cases:
- Sight (computer vision): With the help of thermal sensors, technicians are able to pinpoint areas in need for maintenance.
- Sound (audio): By using sonic and ultrasonic technologies, it is possible to pinpoint potential trouble spots by “listening” for friction that’s outside of normal operating parameters.
- Touch (vibration): The data equivalent of human touch is vibration analysis, which can detect vibrations in machinery and establish both normal and sub-optimal results.
- Smell (TVOC): A sensor could be used to discern a gas leak on a factory floor or a key utility asset.
- Taste (TVOC): Imagine a world where you can identify and taste liquids just by their odor. This could be useful for industries like pharmaceutical manufacturing or beer brewing. It can be also be employed to recognize increases in toxic substances in assets.
- New sixth machine sense (all data): With neural networks and sensor fusion capabilities, machines can detect data from the environment that is invisible to humans.
Business Case Examples for Edge Predictive Maintenance
Predicting asset depreciation and maintenance timeline: The security and building access industry have been experiencing increasing pressure due to the global pandemic, and it’s imperative for customers to understand when a security door or component might fail. By anticipating maintenance, companies can reduce unplanned out-of-service intervals, allowing for minimal disruption in office buildings where there is huge traffic of people.
The Edge Impulse platform and solutions engineering team enables companies to make more accurate predictions about when devices might fail, which lets them optimize their fleet maintenance and use service crews most effectively. This saves the companies money by letting them lower overall asset downtime and allows customers to be more satisfied with their product and services.
Lowering cost and gaining more ROI: A global shipping company was looking for a way to lower their costs and increase efficiency. Focusing on predictive maintenance allowed them to proactively address any issues before they became costly or caused unsafe conditions in order to avoid downtime on ships.
With predictive maintenance, you can monitor your equipment while it’s running: This means that there is less downtime for inspections and repair jobs because the monitoring process takes place during operation instead of waiting until something breaks or wears out.
A shipper company leveraging the power of predictive maintenance can reduce their ships’ times in dock and optimize spare parts storage worldwide so that they had right-sized inventory wherever it was needed. Logistics companies can do the same with machine learning and advance their responsiveness to quickly changing conditions for increased safety with less downtime.
Benefits of Processing with Edge Devices
Data complexity: If you’ve got a factory or manufacturing floor with hundreds of cameras and sensors in it then there’s just no way to send that information across the Internet to the cloud for processing — it’s going to overwhelm whatever kind of connection you have.
Latency: This is the time it takes for something to happen after a key event happened. It’s important in industrial and manufacturing because if there are sudden changes, such as a potential machine malfunction — then those cloud-based compute devices won’t be able to make decisions or predictions quick enough. Cloud processing is simply too slow. Predictive models running on the edge is the way to go.
Cost: The economics of cloud computing are getting better and cheaper all the time, but it still costs money. Edge Computing can reduce data consumption by sending less information to a server in a remote location, which saves energy as well as provides faster network speeds for users on competitive websites who do not have this advantage over them yet.
Reliability: The local processing of an asset-monitoring system means that it will be able to work even when connectivity goes down. Edge machine learning is great for both on- and off-grid industrial assets.
Privacy: With edge compute, sensitive live operational sensor data does not need to leave the facility or be shared with third parties.
Management Awareness
The transition to predictive maintenance will require significant changes in how organizations work. For example, they may need new processes, new silicon hardware, and new procedures for collecting and analyzing data — all while trying not disrupt current business activities. Whether you are a company with data or in need of data collection help, from model development to deployment at the edge, Edge Impulse will give your organization the edge.
Please reach out or book an exploratory call to learn more about how we can help your team with your machine learning needs.