Industrial assets, from complex manufacturing plants to remote and mobile capital equipment, are subject to an asset availability ceiling. While this ceiling varies by industry, peak system availability is typically 85-95 percent. Unfortunately, the widespread acceptance of these ceilings masks the hidden—and significant—costs associated with unplanned downtime.
For typical heavy process industries, these costs can represent 1-3 percent of revenue and potentially 30-40 percent of profits annually. For large capital equipment, the costs may be 1-3 percent of asset value per year. With millions of dollars in savings at stake, the cost of unplanned downtime warrants further investigation.
Patterns in equipment availability
Industry studies show that large complex assets typically achieve 85-95 percent availability. Of greater interest is that nonavailability is split evenly between planned downtime (scheduled maintenance) and unplanned downtime (breakdowns). Because unplanned downtime is so pervasive and no clear way exists to eliminate the problem, the 2-5 percent of nonavailability is accepted as normal even though it represents a significant cost burden. The cost of downtime can be categorized as follows:
Lost revenue. The greatest impact of unplanned downtime is revenue loss. This is typically the result of demand exceeding supply. The loss of revenue due to downtime is especially egregious, because the cost is not just the loss of the typical 3-10 percent profit margin on the lost revenue. It is actually the value of the total revenue lost, less the direct avoided costs of production (generally materials or energy).
Consider this example: an airline flight is cancelled due to mechanical problems and all passengers fly on competing airlines. The only costs the airline avoids due to the cancelled flight are the fuel burned and possibly crew costs. However, no revenue was collected so this becomes a downtime cost. In this example, fuel and crew costs may be approximately 30 percent of revenue, so the cancellation results in a cost of 70 percent of the potential revenue for the flight—much higher than assuming the cost is the typical 6–7 percent airline net profit times the potential revenue. The same logic also applies to plant downtime.
Carrying excess capacity. A typical strategy to address an asset availability barrier is carrying excess production capacity. This may entail building a plant slightly larger than necessary so product can be inventoried to cover unplanned downtime, or carrying spare units to replace those that fail. Both solutions have costs: capital to purchase that additional capacity and additional maintenance expenses associated with a larger facility.
In this model, it is assumed that excess capacity is equal to the amount of unplanned downtime, with a cost equal to that fraction of asset value. This amount then is annualized based on an expected equipment life and discount rate. To calculate maintenance costs on this excess capacity, a rule of thumb can be used. For most long-lived equipment assets, life-cycle maintenance costs are roughly equal to capital costs. In this model, the maintenance costs of excess capacity are determined by a multiple of capital costs. If maintenance costs are known, the correct multiple can be entered; however, for the following examples, it was assumed capital and maintenance costs were equal.
Disruption and recovery costs. The recovery cost associated with returning to normal business operations also must be considered. This could include overtime for emergency repairs, airfreight for materials or spare parts, loss of product due to off-quality operations, etc. Since these costs are situation-specific, it is difficult to use a rule-of-thumb or balance-sheet-based calculation to develop an estimate.
So, for the purpose of this simple model, recovery costs are included as a fraction of the maintenance costs of the asset (which are estimated as a multiple of capital costs). For the following examples, it is assumed that recovery costs are 3 percent of the total maintenance costs, although the model does allow for any percentage to be specified. It also is assumed that the recovery costs are constant even if excess capacity is available; if unplanned downtime occurs, the costs to recover should be the same if revenue can be recovered or not.
Simple downtime cost model–plant example
With the three major elements of downtime now identified—loss of revenue, excess capacity, and recovery—a simple model can be used to calculate the hidden cost based on the amount of excess capacity available to recover lost revenue. In this approach the worst case is assumed to be 0 percent excess capacity, meaning no revenue can be recovered, resulting in a cost equal to the lost revenue less direct-avoided cost of production plus recovery costs.
One hundred percent excess capacity means enough exists to allow full recovery of revenue lost to downtime, so the cost becomes excess capacity costs plus recovery costs. Ratios of excess capacity between 0 percent and 100 percent indicate partial revenue loss and partial excess capacity cost; therefore, these costs are linearly interpolated for situations of partial excess capacity. As mentioned previously, recovery costs are assumed to be the same regardless of the degree of excess capacity.
To develop the costs required for this model, financial statement ratios are used. This approach allows the hidden cost of downtime to be calculated without needing to delve into excessive detail. Fig. 1 shows an example for a heavy process plant. In this case the needed ratios were taken from the financial statements of a large U.S. petrochemical company, for a hypothetical plant valued at $100 million.
The key ratios required to calculate the hidden cost of downtime are return on productive assets, return on sales, and tax rate. The return on productive assets is net profit divided by book value of net physical plant and equipment. Return on sales is net profit divided by total sales. For a given asset, in this example the $100 million plant, the profitability of the plant is calculated by multiplying the return on productive assets by the asset value, and total revenue is calculated by dividing the profit by the return on sales. With total revenue estimates, the asset value, and assumptions on avoided costs of production, recovery costs, maintenance cost multiplier, etc., it is possible to plot the hidden cost of downtime based on a percentage of unplanned downtime vs. excess capacity required to cover that unplanned downtime. (Exact formulas and an interpolation table used for the calculations in this article can be found at www.smartsignal.com/hidden_cost_asp.html.)
As shown in Fig. 2, cost is plotted against total profit from this particular facility, demonstrating that the hidden cost is a large portion of total profit. The percentage of total profit is very high in this case because the business is both low margin and capital intensive. If capacity constrained, cost of lost revenue is high—due to low margins—and if the business is capital intensive, the cost of excess capacity is also very high. Therefore, for this kind of business, the hidden cost of downtime represents a substantial drag on profitability and elimination of downtime can deliver significant cost savings.
Simple downtime cost model–equipment example
This same approach can be applied to individual pieces of capital equipment. If the asset value is known, balance sheet ratios for return on productive assets and return on sales can be used to determine the hidden cost of downtime for an individual piece of equipment. Fig. 3 shows a similar example for an expensive productive asset, such as a locomotive. The ratios used are typical of U.S. freight railroads.
For the case of equipment assets, it is interesting to look at the hidden cost of downtime as a percentage of asset value. In Fig. 4, the cost of downtime is plotted as a function of asset value, showing for this kind of asset in this business that the hidden cost of downtime runs 2-3 percent of asset value per year.
As these examples illustrate, the cost of unplanned downtown can be significant. However, if the availability ceiling can be broken, organizations can achieve significant returns. One solution is to use predictive maintenance software, which can identify emerging problems before they lead to unplanned downtime. Part II of this article, to be published next month, will review a predictive condition maintenance solution and show how it is being used to break the availability ceiling and reduce the hidden cost of unplanned downtime. MT
David R. Bell is vice president of business development for SmartSignal, Inc., 4200 Commerce Court, Suite 102, Lisle, IL 60532; (630) 245-9000.