*Per rodney disgust
Two of the most common concerns about adopting smart manufacturing are price and lead time. But, behind them, there is another even more fundamental: the fear of change. Most small and medium-sized industries have factory floors full of old machines with legacy systems, very old, and have a set of systems that have not yet been integrated. Not to mention the reliance on electronic spreadsheets to exchange information between these systems, in addition to many decades-old clipboards and processes that have not changed since then.
Imagine: older workers are retiring and nobody knows how some of these machines really work. Another dilemma that presents itself in many factories: why fix something that isn't broken?
Many claim that trends come and go, with companies investing in technologies that don't pay off.
But now, industries are realizing that, although the old way of doing things has not been totally overcome, something really must change to guarantee the evolution of things, and that they need more efficient processes to be able to compete in an increasingly business world. increasingly volatile. And as Boomers move into retirement, Gen Z will expect the technology they grew up with to accompany them on their journey.
To adapt to the new era of Industry 4.0, companies in this sector need a more accurate view of what is happening in their plants to make the best decision, and it would certainly be good if all reports were aligned for this change. Reducing scrap (or even knowing how much scrap they have) would help with sustainability and reduce losses. Preventing machines from breaking down would be a dream come true.
Some of this seems too good to be true and conquering this reality requires starting from scratch, learning new systems, buying new machines or changing how they work. So most companies are not willing to make the leap.
But, here's the good news: it really is possible to start modernizing your plant by taking small steps and it's possible to reap big benefits and still keep a lot of your old way of doing things.
How to do this? We'll explore one easy change you can make to improve data collection: installing inexpensive IIoT devices. Then we'll cover how to get the most out of this data collection.
Step 1: Turn Legacy Machines into Next-Gen Data Sources with IIoT
The problem with legacy machines is not that they work in isolation, like islands, but that they fail to communicate the data they collect with management systems and with other machines on the shop floor. The same goes for legacy software systems. They are fine on their own but are terrible team players.
All the benefits of Industry 4.0 depend on centralized data. One way to collect all the data involved in production is by implementing new machines capable of communicating with management systems. But, we know, this is not in the plans of most small and medium-sized industries.
A good path for small and medium-sized industries, until they can buy new machines, is to invest in affordable IIoT solutions to update their existing fleet. These IIoT solutions include sensors, scanners, thermometers, and more, and when added to old machines on the shop floor, you can start collecting massive amounts of data.
However, collecting the data is only the first step. The next step would be to centralize this data to ensure the necessary visibility. For this, it will be important to explore the modern data architecture.
Step 2: Data architecture for transform the data stories
After installing the IIoT sensors, what to do with the huge amount of information collected?
The key is turning data into “actionable insights” and updating your data architecture makes it possible to know what to do with it by turning unused data silos into data-driven pipelines.
Many companies have a good amount of data sources, but isolated from each other and inaccessible by decision systems. Data architecture helps solve this by connecting disparate data sources – whether machines or software systems – and contextualizing them using analytics-based data analysis techniques involving artificial intelligence and machine learning (AI/ML) to find meaningful insights that would be impossible to find without being connected to other data sources.
With the right data architecture, you can quickly analyze data collected from multiple sources. Even better, it's easy to tailor analytics reports to a specific role, or to a departmental staff member. For example, the CEO data dashboard can show cost reduction and ROI analysis. A plant manager's dashboard, on the other hand, can show uptime and throughput analytics, as well as alerts for preventive maintenance.
If everything so far seems complicated, the bad news is that it really is. Now, the good news is that it won't be difficult to achieve this architecture if it is possible to have data integration tools capable of carrying out the communication between the different data sources with the management and business intelligence systems.
A systems integration platform automates all the heavy lifting for the system bus and data sources. As this architecture is cloud-based, it is possible not to worry about the backend such as the cybersecurity involved.
Fortunately, implementing a modern architecture doesn't involve scrapping your old systems, machines, or processes so you can start your small and medium-sized industry modernization efforts. All this without interrupting operations or having to relearn outdated processes.
To complete, the implementation of a data architecture helps to ensure the ROI (return on investment) as quickly as possible, minimizing production interruptions. Teams can continue working as usual and make small adjustments that have big impact based on the data.
Then adjustments – sprint-based – help take the next step. That means you don't have to take on long-term debt and expect the solution to pay for itself in a decade. Instead, small and medium-sized industries will be able to see their investment pay off right from the start, without having to completely change their operations.
(*) rodney disgust is CEO of Magic Software Brasil.
EN 
PT 
