Published on: 

Enrique Herrera, Industry Champion at OSIsoft, discusses the evolving trend

of digitization and what the turbomachinery sector needs to do to prepare for it.

What does OSIsoft do?

We make the PI System, which captures and normalizes data from operational equipment, such as sensors, power supplies, manufacturing tools and even large things like ships. This enables people to boost production, save money, improve safety or conduct analytics.

How does the PI System enable people?

• Xcel Energy saved $46 million over six years with better wind forecasting

• In 2015, ArcelorMittal boosted iron ore production from 23 million tons to 26 million tons by fine tuning its supply chain

• Syncrude saves $20 million a year through predictive maintenance

• Dong Energy did the same in off shore wind • China National Bluestar optimized 12 of its sites. At one site, it saved $200,000 in steam costs alone

• Maynilad, the water company for Manila, reduced leakage from 67% to 32% by tracking pressure losses.

What do you do in the turbomachinery field?

We produce software for operational intelligence that acts as the data layer. Operators want to know the health of their systems. Are they running correctly or is there something out of whack? How can they achieve more efficiency? We serve up the data that lets them answer those questions.


How is the trend toward digitization impacting the turbomachinery sector? Digitization is the future for everyone in industrial equipment. We’re seeing companies embed digital subscription services into pumps, tractors and other heavy equipment. These systems can alert a customer to an impending failure and avoid a costly breakdown. Potentially, everyone will be better off than they were before. Lower costs for the customer and higher revenue for the vendor as we saw at the beginning of the PC revolution. It’s rare, but it bodes well for growth.

How does the turbomachinery industry prepare for this change?

The first step is technical. They must embed these tools into their products. The second stage, which can be tougher, is cultural. What applications do people want? How do you design an interface so people keep using the data streams? How do you establish the return on investment? What are the consequences for those that choose not to make the change? They will be left behind. The results from early applications like predictive maintenance are already in. It works. Companies that don’t move to embed technology for this into their products will find they don’t have buyers because their customers will have already reorganized their operations to take advantage of the technology.

What will be the benefits for those that make the transition?

Happier customers and likely better sales. With digital you can lower down time and OPEX while establishing a long-term relationship with your customer. There aren’t a lot of downsides.

Where do you see the turbomachinery sector going in five years or more based on more use of digitization?

It’s hard to overestimate the impact that digitization will have. Industry uses approximately 54% of the world’s energy — that’s more than transportation, commercial buildings and consumers combined. Yet, if you’ve ever been in a factory you know there are ways to save. Maintenance is a big problem, too. U.S. manufacturers alone lose more than $20 billion a year due to unplanned outages and most of them could be stopped with the right technology. Everyone sees the potential. The exciting part is that we are finally at the point where the technology is making it possible to take action.

Tell us more about your software.

PI System captures and organizes machine data. Our PI Integrators take PI System data and translate it so it can be ported to SAP HANA, Microsoft Azure or other cloud platforms for deep analysis. Without Integrators, that data have to often be sorted by hand. PI Integrators automates it. One of our customers, Cemex, used to spend over 740 hours collecting and prepping data for performance reports covering its 70 plants. Now, with PI Integrators, it takes less than an hour.

So, is everything going to the cloud?

No. The latency and bandwidth costs are too daunting. Imagine trying to control an automated mining tractor via the cloud, or an offshore oil platform. We’re going to see a mix of on-premise and cloud systems for a long, long time.