MHI Solutions Improving Supply Chain Performance
BY , EVP OF INDUSTRY LEADERSHIP, MHI
THERE IS NO doubt about it, the dawn of the artificial intelligence (AI) age is no longer approaching. It has arrived—in full force.
As a result, end users must understand how predictive analytics and digital twins are influencing fundamental, traditional reliability practices. After all, they’re jointly strengthening reliability practices throughout the entire material handling industry.
How so? According to Kushagra Thakur, senior product manager, Forge APM, at MHI member Honeywell, they simultaneously add precision, context and foresight. In turn, distribution centers (DCs) and warehouses have greater visibility than ever before, enabling more accurate decision‑making.
“Predictive analytics detect microscopic shifts in vibration, temperature and electrical patterns long before failure symptoms appear,” Thakur said. “Meanwhile, digital twins extend this by simulating asset behavior under different loads and operating conditions.”
While reflecting on predictive analytics’ chief goals, Nathan Hibbs, material handling and intralogistics industry business development manager at MHI member Beckhoff Automation LLC, believes two especially stand out. First, it will detect any equipment degradation before a catastrophic failure ever occurs. And second, it will ensure DC and warehouse managers’ equipment components have the most usable life possible.
“In an ideal situation, this can enhance maintenance practices by allowing end users to reduce the amount of spare parts they need to keep on the shelf,” Hibbs said. “Predictive analytics should provide them enough warning to procure the components and schedule planned downtime.”
On the other hand, digital twins will enable end users to prepare for unexpected equipment failures (and even predict them ahead of time). Through digital twins, they can stress‑test their entire equipment systems in simulated environments or test the impact of scheduled maintenance on the whole system. By doing so, they can eliminate unintended consequences that might otherwise affect an entire facility rather than just one section.
“Automated systems also constantly feed real‑time data into a digital twin to validate and compare the real world versus expected values, which can inform and improve end users’ predictive maintenance strategies,” Hibbs added.
Over the coming years, the material handling industry will continue to move toward creating a digital simulation world within its systems, too. These systems typically comprise various pieces of equipment from different providers that work together in a facility, enabling many functions.
Through this digital simulation world, design trade‑offs and predictions will occur. Modeling, via digital twins, will be a key component of the upcoming digital simulation, according to Mark Arisman, manager of business development, control products, at MHI member NORD Drivesystems.
“Once we operate like this in the simulation realm, a feedback loop of real data and analytics will lead to more accurate modeling,” Arisman said. “This should result in more reliable systems—because we’ve already modeled and analyzed them in the digital world first—and large productivity gains in design.”
By combining predictive analytics and digital twins, DCs and warehouses will not only minimize downtime but also increase profits. According to Hibbs, they’ll reduce their spare parts, damaged and scrapped products, and any collateral damage their peripheral components may have otherwise experienced due to previous catastrophic failures.
Additionally, material handling systems as a whole will be simplified. In particular, end users will be able to model and predict behaviors more accurately than before. And, of equal importance, they’ll likely notice reliability gains, according to Arisman. These gains will occur in two main ways: enhanced maintenance and diminished downtime.
“They’ll elevate several core practices too, including condition monitoring through higher fidelity and earlier detection. Failure mode and effects analysis (FMEA) will also be elevated as new failure modes and early indicators will be uncovered,” Thakur said. “Reliability‑centered maintenance (RCM)—by matching maintenance actions more accurately to real conditions—and root cause analysis, via data correlation at scale, will be elevated as well.”
Simply put, according to Thakur, predictive analytics and digital twins aren’t replacing each of these core practices. Instead, they’re enhancing them, especially in terms of accuracy and impact.
Regarding timing, it really depends on who you ask. Some industry professionals like Thakur believe the implementation of predictive analytics and digital twins has already begun. In fact, he has witnessed many material handling operations actively adopting predictive tools.