The shop floor is a bustling, busy place. An orchestra of people and machines hums along, moving production forward. Just as in a regular orchestra, synchronisation and harmonisation are essential. Knowing your place in the composition, following what other musicians are doing, reacting appropriately — this is what keeps an orchestra playing and the music flowing. Why should a shop floor be any different?

It shouldn’t. And that brings up the question: How do you integrate all the different types of machinery on your shop floor and visualise the current state of your equipment’s effectiveness?

Working in concert

True harmonisation requires machine connectivity. If you don’t have machines on your shop floor that can communicate their status and collaborate with each other, then you can’t obtain data from the shop floor, you can’t relinquish control over to the shop floor, it can’t talk to itself, it cannot make music — and you have no insight into what is actually going on. Connecting machines is therefore the most pragmatic, and also the most promising, starting point.

What does machine connectivity really mean? Many factories are already automated and controlled via a manufacturing execution system, but when you look at how many machines need to be connected to aggregate efficiency data, we see that true connectivity rarely exists. The shop floor houses machines from different manufacturers and in various stages of ageing. There’s often a glut of very modern machines, but there are also machines that are 10, 20, 30 years old, retained to perform a certain task … and not much else. Newer machines have the ability to share and retrieve data through interfaces, but the machines talk in different protocols — if they’re even able to talk. Evidently, there are quite a few speed bumps leading up to machine connectivity.

Conducting with confidence

For the machines that are actually able to send data, we need to know which protocol the data is in, and how we can access that data. And for the machines that can’t send any data yet — how can we make them send data? How do we retrofit them with the appropriate sensors? Once we have answers to these questions and have put the respective systems into place, we can ultimately harmonise the data formats.

Assessing the shop floor based on its maturity, as well as the potential data payoff, is key. There are some crucial first steps here: judging whether the quality of data surpasses the effort needed to obtain it, finding ways to access the machines and extract their data, harmonising that data, and translating it into a standard protocol, such as OPC UA [Unified Architecture] or MQTT. These initial assessments become the foundation of robust, productive machine connectivity.

In the beginning, the goal isn’t to get as much data as possible out of the shop floor — it’s to get the machines to send the right kind of data you need for certain business outcomes. This might include getting downtime status to improve efficiency. Or setting constant parameters in place that monitor machines, components and their performance so that a robust predictive maintenance model can be set up.

At this point, another question arises: Why do this? Why go through all the trouble of connecting the entire shop floor? And there’s a very simple answer: It’s the base for digitalisation. As soon as you have the shop floor basics connected, digitised and running, you’re no longer chasing the music. You’re conducting the orchestra with confidence — and business success.