Industry Giant Caterpillar Delivers Amazing ROI From Their IoT and Big Data Initiatives – Here’s How

When it comes to big data and Internet of Things (IoT) initiatives most companies are still in the design or early adoption phases which make it hard to get a solid return on investment (ROI) figures. So it’s refreshing to share a story of an organization delivering real-world ROI for their customers by vastly ramping up their data collection and predictive maintenance analytics.

The Marine Division of Caterpillar serves fleet operators of tug boats and shipping vessels for whom fuel usage often drives the bottom line. James Stascavage, intelligence technology manager at Caterpillar Marine, served with the US Navy for 28 years before joining ESRG, which was acquired by Caterpillar in 2015. He shared “Often times when people look at data, they’re looking for the ‘grand slam’ – one thing that’s going to save them tens or hundreds of thousands of dollars. In reality, it’s the small improvements that can add up to big dollar savings across many vessels,” he told me.

As companies set out on data-driven transformations, they want a blueprint to uncover a ‘north star’ insight – one golden discovery that will guide them towards growth. Stascavage shared many “north star” discoveries customers have made using Caterpillar’s Asset Intelligence platform, which is built on Pentaho’s data integration and analytics platform.

Proving ROI on Big Data Investment

Shipboard sensors monitor everything from generators, to engines, GPS, air conditioning systems and fuel meters. In one example, Caterpillar is able to identify that fuel meter readings are correlated with the amount of power used by refrigerated containers.  This data can now be used to determine optimum operating parameters, by simply modifying power output from the generators.

By doing multivariate predictive maintenance analysis in Pentaho, the customer discovered that running more generators at lower power was a more efficient approach than maxing out a few. 

The savings from this were estimated at around $30 per hour, which may seem small, but the aggregated savings for a fleet of 50 ships operating 24 hours a day and 26 weeks a year would generate more than $650,000 in savings.

Another insight involved efficiencies that could be gained by optimizing hull cleaning. As a ship sails, inevitably its hull becomes contaminated with corrosion from salt water and rough seas, and befouled by barnacles, seaweed and other ocean life.

One customer knew that this was impacting the performance of their fleet but had no way of knowing its actual impact, or how efficiently they were spending their nearly $20,000 per year hull cleaning budget.

Data collected from ship-board sensors as the fleet performed manoeuvres under a variety of circumstances and conditions – cleaned and uncleaned – was used to identify the correlation between the amount of money spent on cleaning, and performance improvements.

Stascavage tells me “When we looked at their fuel consumption our recommendation was they should be doing the cleaning a lot more frequently – an interval of around 6.2 months as opposed to every two years.”

“They knew that more frequent cleaning would, say, quadruple the cost of their hull cleaning, but what they never knew was the cost of not doing the cleaning.”

In this case, the analytics said that across the customer’s fleet of 8 ships, inefficiencies due to dirty hulls was costing up to $5 million every year, and by switching to an optimized cleaning schedule they could expect to save up to $400,000 per ship.

Predictive Maintenance Analytics

Initiatives such as these are great examples of practical deployments of Internet of Things (IoT) technology within industry – a trend often referred to as the Industrial Internet. Predictive maintenance analytics using machine learning models built by data scientists using R, Python or Weka, such as those used by Caterpillar Marine, are being used across all fields of industry, as the value of measuring and recording as much as possible becomes clear.

Stascavage says “I think the best lesson we learned is that you can’t collect too much information. I know there’s sometimes a hesitancy – ‘why are we going to collect that? Why do we need it?’ But the days of data storage being expensive are gone.”

“There are relationships between pieces of data that the human eye just can’t see – relationships about relationships about relationships.”

“And if you don’t collect the data, keep the data and analyse the data, you might not ever find the relationships.”

Industrial IoT technology will increasingly make it easier to drive positive growth and increase efficiency in any area of enterprise, which involves machinery. This will lead to less downtime as predictive maintenance and fault-correction become common with less wastage – leading to a positive impact on environmental factors such as pollution and waste material generation, as well as bottom line profitability.

Learn More

To get a closer look at Caterpillar’s big data analytics implementation, please join my webinar on February 23rd, where I’ll be talking with James Stascavage about Caterpillar’s predictive maintenance platform, and how predictive maintenance the industrial internet of things is transforming businesses.

Thank you for reading my post. Here at LinkedIn and at Forbes I regularly write about management, technology and Big Data. If you would like to read my future posts then simply join my network here or click 'Follow'. Also feel free to connect on Twitter, Facebook or Slideshare. 

You might also be interested in my new and free ebook on Big Data in Practice, which includes 3 Amazing use cases from NASA, Dominos Pizza and the NFL. You can download the ebook from here: Big Data in Practice eBook.