The Astros and Yankees might be getting a lot of attention lately with this year’s World Series, but do know how important the Oakland Athletics were to baseball in the early 2000s? The same core ideas which helped reshape the way people think about baseball are the same principles that are helping manufacturing companies around the world.
It’s called predictive analysis and it doesn’t take much for your manufacturing floor to start gathering the type of data that allows your company to knock it out of the park.
For non-baseball fans, the 2002 Oakland Athletics ignored traditional player scouting norms due to their limited payroll which kept them from signing big name players. Instead, they opted for the newly minted “Moneyball” strategy which pointed them in the direction of players who would win games, not create highlights. The strategy has since been credited to building many championship baseball teams in the years since, but it all relies heavily on statistical analysis and computer generation.
Instead of a big name player fans knew, Athletics manager Billy Beane poured over statistics of individual players and favored those that were lesser known, but who could get on base. Instead of relying on traditional knowledge from scouts and coaches, the 2002 Athletics proved with the power of data analysis that slugging percentage and on-base percentage were much better indicators of a good baseball team.
The strategy worked. The Athletics immediately made it to the playoffs in 2002 and 2003, largely on the back of their statistical analysis. Billy Beane’s data collection and analysis changed the Athletics and baseball forever.
It’s impressive the amount of change realized by the Athletics through data collection and analysis. While they might not credit the Oakland Athletics as their direct inspiration, industrial manufacturing plants around the world are realizing that analysis of their machines is becoming more and more important as the digital transformation continues.
Manufacturing statistics have long been tracked in terms of productivity monitoring through systems like OEE, but the change is now how long it takes to act on those numbers. Real-time data collection through connected IoT systems is helping plant managers plan proactively with strategy for what happens when their machines need break down and need repairs. Industrial IoT systems are now building statistics managers can use to monitor their machines, giving them the ability to redefine “failure rate” into “predictive maintenance.” Minimizing machine downtime means plants can produce more of what they were made to do, which benefits every level of a business in the long run.
One unique example of this would be the industrial mining company that chose DATTUS to help monitor the health of its mining equipment. The enormous size and specialized nature of these mining machines makes every minute they are running critical. By bringing DATTUS in to help monitor the vibration and temperature levels of machines, the West Virginian outfit was able to better predict mechanical failure and save 20 hours of downtime in their first 6 months of operations. That’s somewhere in the ballpark of $340,000 saved (producing $17,000 per hour) by using DATTUS and putting predictive analytics to work in the right way.
For businesses not already moving in this direction, decision makers in the industry know this is the future of industrial management. In the 2016-2017 Industrial Analytics Report, survey respondents were asked “What role industrial analytics play in your organization?” 15% of respondents said it plays a crucial role in their organization now, but a whopping 69% said it would be crucial in the next 5 years. The ability to collect data from industrial machinery and analyze it in an effective manner has never been more important and companies around the world are already preparing their “Moneyball” strategies to adapt.
The path toward an effective manufacturing IoT starts here. If you’re interested in a live demonstration of how DATTUS can help equip your plant, click here.