By Dusty Weis, AEM Strategic Communications Manager

Like predators on the hunt, Silicon Valley technology startups are on the lookout for industries where they can swoop in, outmaneuver the established powers and capture market share.

And heavy equipment manufacturers should be on notice, according to Ganesh Bell, the president of Uptake Technologies and a pioneer in disruptive technologies. Because when these outside disruptors look at the construction and agriculture industries, they see easy prey.

“When you think about the collective market opportunity of the industrial world, it actually dwarfs that of the consumer world,” Bell says. “It’s only natural that startups rush to this, and so do investors.”

Bell cites figures from the World Economic Forum, which found that the agriculture industry could generate $570 billion in value over the next decade by integrating artificial intelligence, machine learning, asset performance management and Internet of Things (IoT) technologies. In the construction industry, WEF pegs the potential productivity gains these digital technologies could enable at $470 billion.

“I’ve been living and working most of my career in Silicon Valley,” Bell says. “But over the last decade, I’ve been so fascinated by this idea of businesses and industries being reimagined with software that I bet my career on it.”

Now at the helm of Uptake, a leading provider of digital industrial solutions, Bell says there’s no reason that heavy equipment manufacturers can’t incorporate some of the same strategies that startups employ. Speaking at the AEM Annual Conference, he outlined three steps that manufacturers can explore to lead in the digital transformation of construction and agriculture.

Choose Purpose-Built Technology Solutions

The enterprise application software landscape is vast and littered with acronyms like CRM, MES, BIM, SPC and ERP. Each of these software solutions is tailored to a specific facet of business operations, like customer relationships or financial planning—and that’s the problem, Bell says.

While the platforms may have changed, he says the solutions themselves do the same things they did 30 years ago.

“I would argue that the most valuable data in the enterprise is not what you type into these systems, it’s data from your operations and your products,” Bell says.

Enterprises need technology solutions that synthesize data from all facets of their operations, Bell says. In addition to the traditional data tracked by enterprise application software, OEMs should incorporate machine data from IoT sensors and analyze it with adaptive learning A.I. technologies, mining insights that require a complete understanding of a business’s operations.

“All that stuff has to be brought together, and that requires a completely different architecture,” Bell says. “You couldn’t take your old contact data and hope to do what Facebook did to understand your social network. The only way to achieve that level of analysis is to purpose-build a system for that.”

The point of enterprise software isn’t just to collect and analyze data, Bell says. Rather, if it’s purpose-built to pursue specific outcomes like improved reliability, reduced costs and an equipment-as-a-service business model, that’s when OEMs will be best-positioned to succeed.

Partner to Lead

There’s a direct correlation between the size of a data set and its value in creating profit-driving efficiencies, Bell says. And the simple fact he wants to drive home is that no one OEM has collected enough data to go it alone.

“In many of your companies, you have hundreds of use cases where you could apply machine learning and advanced analytics today,” Bell says. “But to do that, you have to utilize data from your suppliers, your dealer networks and your customer operations, and even curated data from industry sources.”

However, partnering with sources of data is only half the battle, Bell says. In the fast-paced realm of innovation, the old adage holds true with disruptors—if you can’t beat ‘em, join ‘em.

For example, Bell’s own company, Uptake, has compiled a data source called the Asset Strategy Library. Tapping into troves of machine data, they’ve mapped out the performance of 800 equipment asset types, and more than 10 million equipment components. In their analysis, they’ve identified 55,000 ways these machines can fail, laying the data foundation for advanced predictive maintenance systems.

All tallied, Bell says it would take 32,000 years of human experience to accumulate all the data in the Asset Strategy Library. That head start in data analysis could represent a significant advantage for any manufacturer choosing to partner with Bell’s company, or another like it, but he says Silicon Valley was built on such mutually-beneficial arrangements.

“If Amazon makes $27 billion, there’s probably an ecosystem of players that make another $27 billion off of them,” Bell says. “That’s why they’re winning. You have to partner, not just for competency or data, but for other things as well.”

Reimagine Analytics With A.I.

Of course, there’s no feasible way for a team of humans to process 32,000 years’ worth of data and come away with precision insights. But for modern supercomputers that can perform quadrillions of calculations per second, it’s not such a tall order.

While “artificial intelligence” is an intimidating buzzword, Bell says it’s really just about enabling machines to use calculations to learn from patterns in data. At Uptake, they wrote a software algorithm that analyzes machine data looking for anomalies that correlate with machine failure. And, by siccing that algorithm on 1.1 billion hours of operational data, they were able to produce a failure prediction software engine to monitor machines in the field and predict when they might break down.

When deployed, these kinds of A.I. constructs monitor real-time machine data, make predictions and keep track of their successes and failures, learning from their mistakes and increasing their accuracy with every day of operation.

“For one of our leading customers, in a matter of a year, our engine started predicting over 90 percent of all failures in a large locomotive,” Bell says. “We can learn at a speed that’s absolutely inhuman, and that’s the power of machine learning and A.I.”

Such A.I. platforms can be applied to any piece of equipment in any industry, Bell says, and eventually grow to the point where they create significant savings for operators through reduced downtime and maintenance costs. But that’s only part of the value.

Because, as Bell explains, the power of A.I. analytics can be applied to any facet of an organization’s business model. And all you have to do to benefit is follow the data.

“Amazon is now in businesses that they themselves wouldn’t have predicted five years ago,” Bell says. “And that is the power of following data, and analytics and digital technologies. Because you will unlock new value and revenue streams, but only the companies that bet big will win.”

Subscribe to get the AEM Industry Advisor in your email.