Turn big data into smart data. Data provides insight into the ins and outs of assets. It's a solid foundation for predictive maintenance. But why should you avoid big data? And how do you make your data smart? Read our blog, "How do I avoid big data?
Big data is hot
Organizations want to know as much as possible. Sensors are stuck everywhere and everything is logged. That produces a lot of terabytes of data, data is the new gold! A advantage for the asset manager, because now you feel like you know everything. A line of thought that has taken root in many organizations in recent years. The result is a true big-data explosion.
In some organizations, it works.
McLaren's Formula One team pulls 25 terabytes of information per car from a single race. A team of data analysts studies that in real time, supported by a battery of data servers. Computing power that could reduce infant mortality in a children's hospital by 30 percent, as Peter van Manen of McLaren Electronics tells (and proves) in this Ted Talk. Talk about the power of data.
Meanwhile, there are also opposing voices
Organizations are revisiting the usefulness of big data. With a big pile of data, you still don't necessarily know much. After all, what good is all that data if you're not set up to analyze and apply it all? How do you figure out what information you actually need from all your assets? And how are you going to set up your processes so that you can actually do something with all that data?
Source: hxdbzxy, Shutterstock
Therefore, a breakaway is now emerging that rejects big data and focuses on "smart data.
Smart data: that is data that contains real information, on which an asset manager can set up predictivemaintenance. Data that is linked to the criticality of your system. But how do you get that smart data? You have to go back to the core business of your organization and its critical factors.
The following questions are relevant:
- Which system parts are critical to my core business?
- What information do I need to direct?
- Are there other factors that may have an influence?
In the auto industry, you see this already happening. Critical systems such as brakes and oil have been monitored for years. If something is wrong with those, you get a message on your dashboard and have to go to the garage. At the same time, the on-board computer is able to analyze your entire driving behavior. Are you a quiet Sunday driver or a highway driver, do you always drive alone or do you tow large trailers? Do you drive in areas of high humidity, or in extreme cold?
Combining the technical parameters on wear and usage with driving behavior and conditions creates a reliable and predictable breakdown risk of the critical parts. Yet maintenance intervals are still set at a number of 20,000 or 30,000 kilometers. By using all the data you have, you can determine service intervals accurately and specifically for each car and its driver. In doing so, you save costs and reduce risks.
This is exactly how predictive maintenance works.
Collect applicable data linked to the criticality of your system. It may still be a lot of (big) data, but it is all relevant information on the basis of which you can steer and make decisions. It prevents downtime and unnecessary maintenance. Exactly what you have in mind as an asset manager.
On to Industry 4.0!
Interesting in this context is "The Internet of Things. This phenomenon allows objects themselves to store data in a network or cloud and share it with other objects. This too produces big data that must be reduced to relevant (smart) data. This phenomenon is already being called the fourth industrial revolution, and it offers fantastic opportunities in life cycle management and predictivemaintenance.
This is still in the future. The challenge is already big enough to get from big data to smart data: data linked to your core business and the criticality of your systems. A step that is going to give many asset managers peace of mind.
