Four datasets critical to a successful predictive maintenance program

April 9, 2019

Manuel Terranova, CEO Peaxy, recently spoke on a panel for the American Society of Naval Engineers (ASNE) on the University of Pennsylvania campus in Philadelphia. The Intelligent Ships Symposium is a biennial naval engineering conference that explores recent advances in Intelligent Systems and the challenges that lie ahead for the next generation of Intelligent ships. The discussion delved specifically into big data, machine learning and predictive analytics along with their ability to bring greater transparency and insight into the Fleet maintenance process. 

Here are some highlights from the discussion:

Many analytics programs look to the most obvious data sources stored in various business intelligence platforms. These are often provided by third parties as part of a hardware purchase. Unfortunately, this data is often incomplete, error prone and manipulated to provide summarized views and dashboards more suitable for executive level analysis.

In order to be successful, predictive analytics require access to four critical data sets:

 

  • Telemetry – unstructured data produced in real-time and at high fidelity by internet-connected machines with sensors.
  • Geometry – unstructured data that consists of engineering drawings and schematics of key pieces of equipment.
  • Simulations – unstructured data that provides real-time readings and results under various performance scenarios; absolutely essential to be able to predict equipment behavior and to drive informed decisions.
  • Service Records – generally structured data contained in “as built and as maintained” Bills of Materials and other records reflecting repair work performed on key pieces of equipment.


Since some of this data is often captured and stored at the “edge,” one of the biggest challenges in a predictive maintenance program is to migrate that data into a normalized compute environment, with a constant flow of updates. This often requires “heavy lifting” and the cooperation of multiple teams, and requires skills many organizations don’t possess in-house.

Another major challenge many organizations don’t anticipate is not having access to key data about their own machines. OEM’s often own this data and will not readily provide it unless a contract providing free access is established from the beginning.

Finally, companies need to incentivize the free sharing of data between organizations and break down artificial barriers and data fiefdoms. Key team members on a predictive maintenance program should be issued “hall passes” to allow them unfettered access to data regardless of its source.

What solutions does Peaxy offer?

Peaxy built one of the first combined physics and economic digital twins in 2016 for a Fortune 10 company. That effort delivered a digital twin of a complex system, including gas turbines, steam turbines, steam generation and power generation equipment all faithfully modeled and validated. While that first effort spanned over 200 days, we can now create twins in as little as 65 days, at a fraction of the cost.   

Peaxy offers predictive asset management for industrial equipment that turns operational data into insights that optimize productivity and increase revenue. Our Peaxy Lifecycle Intelligence (PLI) product is a modular, scalable, cloud-based asset management solution aligned with the needs of the value-driven enterprise. By rapidly turning operating data into financial insights, PLI lets operators minimize O&M costs and optimize performance, improving the lifetime value of industrial assets. PLI serves a wide range of use cases involving precision-engineered equipment, grid-scale battery installations, aviation components, gas turbines, steam turbines, wind turbines, compressors and propulsion systems.

Learn more here!