Over the last several months, reports have piled up on the missteps big data solutions are having with IoT Industrial Applications.
GE Digital, announced last spring that it was taking a step back from its GE Predix investments because the platfrom underperformed two years in a row, with a 25% decline in share price in 2017. Similarly, Uptake Technologies announced in December that Valor Equity Partners ended its $35 million investment after just three months. It’s largest customer Caterpillar also halted its investment.
The reason: data science alone cannot predict failures far enough in advance to provide the value needed to lower costs. It’s simple, data science solutions rely on data anomalies from sensors to indicate the onset of failure. By the time the alarm is sounded, it’s too late to take action to prevent the failure.
It’s like going to the doctor because you’re sick. Yes, the doctor will give you medication to treat the ailment, but it’s too late to prevent the illness from happening. You’re already sick.
Yes, there’s value in this, because you’ll feel better but there’s not enough notice to change outcomes.
Sentient Science brings a unique perspective to the industries we serve because we combine materials science with data science to predict the lifecycle of parts, then monitor the health and recommend life extension actions that prevent failures from happening. The business benefit comes from preventing the failures from happening in the first place; not creating alerts after failures already occurred.
Sentient’s solution looks at the accumulation of stresses at the microstructure level to predict the initiation of failure, providing longer lead times needed to effectively plan for when and where parts will be needed so that operators can reduce safety stock.
While data science stumbles, we’re growing and expanding our customer base in the electrification and transportation markets. We’re also bringing employees from those companies into our small 70-person operation.
Why? Because software engineers who are building out these data science-only platforms see the $1 billion mistakes that are happening. They know that the materials science models are needed to provide true value, but the Original Equipment Manufacturers will not share out the asset configurations and the materials being used because their largest revenue comes from their aftermarket parts and services divisions.
So today, asset configuration (or the Bill of Materials) are rarely shared from the OEM to the operator customer, and without that knowledge, the operator will always be a step behind, operating under costly corrective maintenance practices.
DigitalClone Live is shifting the balance of powerfrom the OEM to the operator and creating supply chain efficiencies that are beneficial to both the consumer and the producer. As the industries we serve begin to see this shift, capital expansion can accelerate because there’s significant reduction in costs to both sides. This results in new business opportunities, more open data sharing and accelerated growth within the industries.
To learn more about our approach or to speak with one of our materials scientists, request a demo.