Stop wasting energy: Use predictive maintenance

By Jonathan Cooper


Industrial Exchange

August 18, 2017

Maintenance techniques targeting energy efficiency can result in an annual savings of up to 20 percent. Why, then, did two-thirds of energy generated in the U.S. last year go to waste?

According to a U.S. Department of Energy study, maintenance techniques targeting energy efficiency can result in an annual savings of up to 20%. Why, then, did two-thirds of energy generated in the U.S. last year go to waste?

Plants, factories, facilities, and commercial and residential buildings account for nearly 40% of the total amount of American energy consumed. Running faulty or substandard equipment is often the culprit, and energy leaks are easy to overlook. When you add it all up, you can see how underperforming equipment can account for a significant percentage of operating costs.

The solution is to replace or repair malfunctioning equipment. Until now, that’s been a matter of scheduling and guesswork. Switching to a predictive maintenance (PdM) approach from a preventive maintenance (PM) approach offers a new way to drastically reduce costs. PdM is an early warning system for when to replace or repair malfunctioning equipment. It reduces overhead, prevents unplanned downtime, and provides insight into real-time equipment conditions.

In the industrial sector, inefficient equipment and maintenance practices can cause energy expenses to skyrocket. For example, effective HVAC maintenance alone can reduce energy costs by 5% to 40%, while malfunctioning machines or less-than-optimal equipment function can increase energy spending by 30% to 60%. Another culprit, motor-driven equipment, is responsible for 64% of total building electricity consumption.

The total cost of ineffective equipment can range from $40 to $150 billion per year, and it’s not uncommon for these inefficiencies to result from even the simplest problems. For example, minor bearing wear or even slight misalignment can cause a standard AC motor to consume considerably more energy than it should.

While preventative maintenance allows machines to run inefficiently, wasting both energy and money, these issues are detectable with PdM. When problems occur on the machine level, they can be wasteful, but when expanded out to an entire warehouse or production facility, costs are more pronounced and sometimes painful.

Until recently, PdM remained a cumbersome and costly approach to equipment maintenance, requiring custom-installed monitors for each piece of equipment and highly trained technicians to install and operate accompanying devices. With the advent of high-powered, connected smartphones and tablets, cloud computing, big data, and an increasingly technologically savvy workforce, PdM is no further from most warehouse floors than the maintenance teams' pockets.

Much the same way that on-site servers and data centers are affordable and accessible for the enterprise, PdM has come into the age of the cloud. And with this reduced barrier to entry, the case for cutting energy costs almost presents itself.

Typical PM methodology relies on a rigid schedule of manual equipment inspections, replacements, and repairs, as delineated by pre-set calendar intervals. With a PdM approach, condition-based monitoring collects real-time data through vibration and ultrasound analysis to determine if and when parts require maintenance for optimization.

Instead of ignoring malfunctions and running equipment to failure, PdM delivers a comprehensive analysis of all machines and equipment. Technicians gain a high-level view of energy use, locating and identifying opportunities to correct energy leaks and cut energy costs across the board. As PdM gains traction across building types, it’s likely that we’ll finally see less energy wasted and more money saved.

Jonathan Cooper is the Head of Business Development at Augury, a New York- and Israel-based company that is bringing predictive maintenance technology to new markets and creating the mechanical diagnostics platform for the Internet of Things.