 |
| credit: Reliable Plant |
Do you know how often your equipment fails – and do you have the information available to prevent the equipment from failing?
In every industry, Big Data is quickly becoming a trend-leading issue and an opportunity. Providing data based on real-time decision making from a huge amount of available data is a competitive advantage to equipment manufacturers, as well other unrelated industries such as retail.
For example, in the aerospace and defense industry and the defense and security sector, most procurement contracts for large military equipment are approved based on relevant and well-analyzed maintenance data and resulting maintenance concepts for the period of operational use of the equipment.
With the application of Big Data, many aerospace and defense service organizations believe that they can predict possible equipment failure better by having a platform capable of handling a vast amount of data quickly in real time.
By installing the evaluated equipment in a large number of so-called end-systems, a process called “Failure Reporting and Corrective Action System” (FRACAS) helps gather information regarding the failure in question. Fleet data collected, evaluated, and presented in real time empowers service organizations and their engineers to better be prepared to handle any maintenance task which may occur due to any equipment failure.
Personally, I agree with Alan Epstein, VP of Environment and Technology at Pratt & Whitney when he states:
“Maintenance was always on condition; we just use computers to understand the condition now.”
This is only one side of the coin. Here we are talking about data from existing fleets.
This is only one side of the coin. Here we are talking about data from existing fleets.
The other side is about predictive simulation of possible and probable failures from a process called “Failure Mode Effects and Criticality Analysis” (FMECA) for end-systems in the design phase. In this process, possible failures are simulated and accumulated from the submodule level up to the end-system level to evaluate the possible effects of any equipment failure. This bottom-up approach helps prevent unnecessary corrective actions and preventive maintenance tasks. However, a vast amount of data needs to be evaluated quickly to influence design.
I am a strong believer that the essence of both processes is that we want to increase the mean time between failures (MTBF) to an absolute maximum to help ensure end-system 100% (or as close to 100% as possible) availability and readiness and keep maintenance downtimes to an absolute minimum. You cannot just use one of the approaches described above – you need both. We need to look at maintenance-relevant issues in their totality and holistically to reach meaningful results every time. Past technology was not able to provide this data fast enough to help ensure expected system availability. We need to provide a way to increase end-system availability to secure profitability.
According to Martin Frutiger, head of IT Tools at SR Technics, “If a component is performing better than expected, its replacement or overhaul can be delayed. If performance is not as good as it should be, it can be overhauled earlier.”
According to Martin Frutiger, head of IT Tools at SR Technics, “If a component is performing better than expected, its replacement or overhaul can be delayed. If performance is not as good as it should be, it can be overhauled earlier.”
It’s one thing to evaluate fleet data to generate meaningful maintenance data, but it’s another to start the process long before any fleet exists. This can be done by simulating possible failures in the early design phase to help prevent them from happening in the first place.
In this context, a specific application of the SAP HANA platform at John Deere has proven to be a platform that supports exactly the above. Together with The Applied Research Laboratory at Penn State University, we have done a proof of concept to demonstrate the power and ability of this platform in the context of system logistics simulations, fleet data evaluation tasks, and challenges.
For more information about predictive maintenance in aerospace and defense manufacturing, read the full article with insights from Martin Frutiger and Alan Epstein, as well as presentations of the discussion about the future government.
