Predictive maintenance has been touted as a “killer use case” for IIoT, Digital Transformation, Industry 4.0, or any other buzzword or marketing jargon that’s been used in the last decade or more. For folks that have been around the space for longer than we care to admit, glad to be invited to the party. Opting for a data-centric approach you unlock a whole new level of identification and accuracy. Here are five that create a strong business justification for most organizations.
Many pieces of equipment that are not properly maintained consume additional electricity
Something as simple as imbalance in an electric motor can lead to 20-30% increases in electrical consumption. In an assessment of about 500 motors that could result in excess spend of $1-2M per year. The same can be found for conditions like belt slip, again, resulting in far greater consumption just to meet base demand. Also, with the growing nature of electrical components in facilities, load can increase causing harmonics in your systems, again leading to an unclean electrical signal and additional consumption and spend.
Quality of production is directly tied to asset health
Even in instances where equipment is still functioning, if it’s not operating at proper levels, you’re likely to see decreases in your quality numbers. That could mean longer times to process a batch, rework, or excess scrap or consumption of materials. Small improvements in reliability can yield significant results. Just think, for every 1,000 units you need, a 10% decrease in productivity means you need 1,110 units produced to meet your goal. In long run batch and continuous environments that adds up to significant costs on an annual basis.
Increased capital to meet capacity
While not explicitly tied to quality, one expansion method is adding new equipment, lines, or even plants. If you’re operating to a high OEE or TEEP, you may need to invest, as there’s not a ton you can squeeze from the equipment. However, looking at many facilities, that’s simply not the case. With data centric reliability we can improve availability, performance, and quality. Using data and monitoring, optimizing the line speed to your desired quality and uptime becomes more predictable, because you can see the parameters in real time, catch anomalies before they impact function, leading to potential for significantly better overall productivity.
Failing to meet production goals and schedules
Unless you’re already running 24 x 7 a way to fulfill orders and production is to run additional shifts, whether with additional staff or overtime. As discussed above, if you’re not working to improve or maximize your OEE, you’re likely to spend more money meeting those targets. If you are running 24 x 7 and falling behind, you’re once again thinking about capital expansion to close the gap. Being able to see and manage your equipment and processes with better data will help weed out these inefficiencies and improve productivity to lower COGS.
Overspending on ineffective maintenance strategies
There’s no arguing that preventive maintenance (PM) is more cost effective than reactive or corrective maintenance. However, many PM strategies continue to rely on time based work orders, often dictated by manufacturer’s recommendations. Those PMs are not designed to optimize spend and performance, they’re meant for preventing failure during the warranty and selling spare parts. Great for them, not the best strategy for you. Moving to data-centric and condition-based work allows you to truly optimize spend and performance. This easily leads to removing 1 of every 3 PM work orders and in many cases over 2 out of every 3. The impact on your work order backlog and spares budget is tremendous.
Leveraging data has a massive impact on the operation of your facility. Actionable knowledge of your equipment can be established quickly and cost effectively. All of this is demonstrable, allowing you to truly experience the impact before committing additional time to expand and capture more of the benefits.
