Blackbird and Factbird are on a magazine, “Jern-Maskinindustrien”, written by the journalist Laurids Hovgaard. Now English version is available to read.
Simple data collection optimizes production
Blackbird, a startup company, makes it possible to obtain production data by using a simple system without PLC and Scada Integrations.
By Laurids Hovgaard (Original text is in Danish. Translated by Blackbird ApS)
When Royal Unibrew, a Danish brewery, and emendo, a consulting company, wanted to optimize filling lines at the brewery, they came up against a brick wall: data about production rates was nearly impossible to collect. Sure the data about units produced and downtime was in the PLC or SCADA system, but building out integrations to retrieve this data could cost a fortune! So Finn Hunneche, partner in emendo, started to search for a solution that could be installed without PLC integrations.
“After a week of searching, I had to give up – I just could not find it anywhere. It became clear we had to develop a solution by ourselves. Then Blackbird was founded as a spin-off of emendo and we started to develop Factbird, a box that could collect this data without PLC or SCADA integrations. Production data is sent to the cloud via a SIM card without using any cables and wires, so manufacturing companies can see how production lines are running in real time on a mobile, tablet or PC” said Finn Hunneche.
Since the Factbird box and sensor do not have to be integrated into existing systems, it does not take long to install the system and get started, which also keeps costs down.
“You can register which batch should be run and continuously monitor whether production runs as planned or not. It is just like GPS which predicts when you should arrive at your destination based on big data collected over the time”, said Mads Petersen, CTO for Software Development at Blackbird.
By being able to access the production data of your lines on blackbird.online, you do not have to sit and go through all the video to find the cause of a production breakdown.
Blackbird’s data shows exactly what time the stop occurs, and then you can watch the video of that specific period of time.
Factbird was launched in March this year, and there are about 20 companies who use it since then. Some use Factbird only to collect production data, while others have worked with blackbird to create custom solutions for their specific needs.
Many of production staff have their own opinion about how much downtime is happening every day and why it happens.
At Royal Unibrew, it was expected that there would be many more stops as the line speed is increased – perhaps so many that the total production volume would fall. However, to everyone’s surprise, it was found out that in most cases the number and the length of the stops were not affected by the increased speed. That is, the collective perception turned out not to be true. The line actually benefited from running at a higher speed.
The result was that Royal Unibrew started running the line at maximum speed, and consequently the output was increased.
“Instead of judging by own opinion, we provide fact-based overview which makes it easier to optimize”, said Finn Hunneche.
This fact-based approach is exactly what Mette Geisler, Production Manager at Premiere Is, uses her Factbird for.
“I have used Factbird mainly to identify the number of stops, their duration, and the time of start and stop on the line. “Factbird” has become an important tool to start my day at work. In seconds, I can get an overview how the production has been running for the last 24 hours. – Once I have checked my Factbird and read the team’s journal, I am well prepared for the status meetings”, said Mette Geisler.
Machine learning by data
Today, Blackbird’s biggest challenge is neither hardware nor software, but knowledge, both for Blackbird itself and what the Internet of Things technology can be used for in the industry.
“One of the biggest challenges in IoT technology today is that many do not know how to use the technology. Today Factbird is used primarily for monitoring production lines, but in the future, we will hopefully be able to use machine learning to help production machinery that predicts and optimizes processes based on algorithms instead of pre-programming. This means that the machines will learn from data instead of just collecting it. IoT technology becomes essential if Denmark, a high labor cost country, wants to manage the global competition. If you can reduce idle time, while being innovative, then you come long way”, said Finn Hunneche.