Barts blog: Practical Challenges in IIoT Implementation
Barts blog #3
IIoT implementations have a tendency to cause headaches, and not just from a management perspective – the technology side also presents challenges. At Sitech Asset Health Center, we help organizations in the process and chemicals industries with their digital transformation strategy, both with consultancy and implementation. Over the past year, we have been able to use digital monitoring to improve the reliability and performance of several of our clients’ assets. Following on from my previous blog post, “Innovation is a Case of Trial and Error”, in this post, I’ll look at a real-life case shared by a colleague.
The process began by working with the client to select the assets to work on and demo the usage of smart sensors. Using an existing failure analysis, produced with FMEA tooling, the assets and the signal types that needed to be monitored were determined. After elaborating the scope and practical boundary conditions the actions moved on to the implementation phase. And it’s in precisely this phase that the below shared, interesting challenges and learning moments occurred.
Selecting the Right Sensor
Finding the right sensor was the first challenge, as alongside capturing and sending data, the sensor needed to satisfy a number of additional requirements. Some sensors require ATEX certification. We operate on a chemicals site, where there is a high risk of explosion, and only ATEX-certified sensors are permitted in these environments. To be sure that sensors will fit the requirements, ask for the specs and certificates.
In addition, battery life was also a key decision-driver. Occasionally replacing the batteries in a few sensors is not a major problem, but dealing with that following the scale-up that we anticipate would be a different matter entirely. Even though the service life of a battery is typically more than one year, practice shows that it can vary. The service life of a battery heavily depends on the frequency at which data need to be transmitted – once an hour, or more than six times an hour? This is especially important in determining the service life of the battery.
Registering Sensors
Virtually everything that is integrated into a chemicals installation has a registration number. There aren’t usually any naming agreements for sensors, let alone for the (maintenance) system that they should be recorded in. And that can be a problem. If not everyone in the team is aware of which sensors need to be installed where, there is the potential for a lot to go wrong. For efficient implementation and later on in the maintenance phase, it’s important to compile a diagram and secure the data beforehand, showing where each sensor is installed.
Sending Data
Data transmission needed to be wireless; that was one of the basic principles underlying the project. The plant has very few signal/network cable junction boxes, so a wired configuration would have meant costly infrastructure installation. It was clear beforehand that the assets that had been selected were in an area with good LoRa coverage, so what could possibly go wrong? Well, the plant is full of metal, which meant that signal strength in area’s was poor. In some sensor positions, the signal was just not strong enough for stable data transmission. By analyzing the signal fields, in the sensor positions specifically, and adding signal boosters, we were able to fix the problem.
So the colleague who shared this story just would make the message, although sensor technology is not new an overall IIOT first time right implementation requires “some” experience or otherwise acceptance of learnings! Who doesn’t agree!
By sharing this case, I hope that I have demonstrated the steps along the road to Industry 4.0. We should not be afraid of the learning moments, as doing and then sharing experiences is the way we can accelerate digitization.
If you’d like to know more about this case or want to share your own experiences of digitization, I would be happy to talk to you.
Bart Kooyman
info@sitech.nl