Pretty much any expense a company makes has to be cost-justified in some fashion, and often that process involves calculating a return-on-investment (ROI). If one can recoup the expense over a reasonable period of time, and, ideally, continue to save money, then it probably makes sense to pull the trigger on the investment.
That said, not every investment presents an easy to perform ROI calculation. Spending money on insurance comes to mind. Companies spend vast amounts on insurance to protect their buildings, intellectual property, equipment, employees’ health, and all sorts of other items. Performing an ROI calculation on whether the expense is worth it essentially comes down to calculating how likely it is that some sort of problem will occur and the cost of the resulting damage. The higher the potential financial loss, the more the company is likely to spend on insuring against it.
Uninterruptible power supplies (UPSs) require a similar ROI calculation because, in effect, they offer insurance against downtime caused by some sort of electrical disruption.
UPS systems can protect against two types of downtime, in different ways. The first is a utility outage that results in loss of power to an entire facility. Facilities that maintain a “high availability” posture, meaning they take steps to ensure redundancy to protect against common failures, can expect maybe one whole-facility outage per year (at least in North America; your mileage may vary, as they say). Critical facilities such as hospitals fall into this category. Other, less critical facilities that take fewer steps to ensure power availability can expect probably three to five outages per year. (And some may experience far more, such as this school district in Alabama that experiences some 22 lightning storms per month during the summer.)
In a whole-facility outage situation for a high availability facility, the role of the UPS is typically to provide power to critical systems until backup generators come on line, which usually takes just a few seconds. Still, it’s a critical period because systems and processes may need to be restarted, and that can result in far more downtime and potentially the loss of important data.
For outages at lower-availability facilities, the role of the UPS may be to provide power to critical systems for some length of time, maybe an hour or two, until the power returns. Or, the UPS can provide power long enough to allow for graceful system shutdowns, preventing the loss of data and enabling faster system restarts.
The other type of downtime is a partial outage, one that affects only a portion of the facility. Tripping a feeder breaker, for example, could bring down a sizeable portion of a facility and may happen as many as 10 times per year. More localized events, such as a tripped circuit breaker or fuse, may affect only a single room or a piece of equipment. But such events are fairly common, occurring maybe 10 to 100 times per year, depending on the size and complexity of the facility. The judicious use of UPSs to back up critical areas and equipment close to the load can protect against these localized outages.
Also, keep in mind that on-line UPS systems deliver clean and continuous power, isolating equipment from power disturbances, not just outages. Many systems have proven to last longer and have a greater MTBF (mean time between failure), when operating on UPS-generated power.
UPSs are available in a wide enough range of sizes and power capacities to protect against pretty much any outage imaginable. The trick is to determine the level of protection that makes the most sense given the risk tolerance of the company to different sorts of outages, and in different areas.
The ROI calculation, then, involves determining how much downtime will cost the company, which will make it easier to calculate how much it’s worth to protect against that downtime. It is also important to remember that, for brief outages, the effective downtime can be far greater than the outage itself. It takes time and, depending on the application, some level of intervention to restart systems.
Cost of downtime figures vary wildly by industry. A white paper by Vision Solutions, which provides high availability solutions for IT systems, quoted these cost per hour figures for various industries:
Brokerage Service: $6.48 million
Energy: $2.8 million
Telecom: $2.0 million
Manufacturing: $1.6 million
Retail: $1.1 million
Health Care: $636,000
The white paper was published in 2008, so you can bet those figures are significantly higher today. But the point is the same: downtime is expensive.
The expense can come from various areas, including lost production in a manufacturing facility, or lost productivity caused by an idled workforce in any company. A sudden loss of power may also cause damage to equipment and processes and, as mentioned, loss of data in IT systems. It can even compromise employee safety.
To determine its own level of risk, companies can examine data they collect from various systems. In manufacturing environments those systems may include sequence of event recorders (SERs) and manufacturing execution systems (MESs) while in data centers it may be data center infrastructure management (DCIM) tools. The idea is to identify those machines or systems that have historically had issues or are most at risk.
Companies should also determine which systems need to be available through an outage, including control and monitoring systems. Hospitals, for example, routinely identify critical systems that must be protected against any kind of power outage, whether whole facility or more localized.
A thorough risk assessment may also turn up potential improvements in the UPS architecture, such as whether a centralized vs. distributed structure makes more sense and areas where more layered protection may be needed.
Only after identifying the cost of downtime in your organization and the level of protection various systems warrant can a company determine how much UPS insurance makes financial sense. To learn more, click here to visit our Critical Power web site.