A wealth of information goes into constructing the ST-RISK™ fragility functions. Fragility functions indicate the variability of economic loss associated with different levels of ground shaking. In past earthquakes, observed damage has been recorded for wide ranges of ground shaking. If these observations are plotted for a particular level of shaking and fit to a curve, the result will be a graph that looks like the following:
The smooth curve representing the loss variability is known as a probability distribution
density function (PDF). At each level of damage a probability of occurrence can
be determined from the curve. You can also obtain a mean level of damage, a median,
and a 90th percentile. The 90th percentile damage is also referred to as the Probable
Maximum Loss (PML) for this particular level of ground shaking.
ST-RISK™ uniquely blends insurance loss data with post earthquake observed loss obtained as part of engineering reconnaissance information to generate damage functions that most accurately reflect reported losses from earthquakes.
The benefit of using both types of information is that paid insurance claims alone do not necessarily reflect potential losses to owners. Insurance-based losses pose significant obstacles if they are the sole basis of forming damage functions:
- Insurance claims usually don't accurately represent losses that fall below the deductible. With relatively high deductibles, usually at least 10 percent, a significant number of structures are damaged below the deductible level. In most cases damage resulting in losses less than the deductible is not reported to the insurance company.
- Insurance claims often reflect damage that includes pre-existing conditions. Although it might be appropriate to include these losses as part of paid, insured losses, they are not appropriate in estimating real losses to a property owner.
- Often political pressure will cause insurers to be liberal with their payment of earthquake claims, particularly if they are covered by reinsurance. Any such overpayments would translate into an overestimation of loss for a property owner.
It is therefore preferable to rely on both sets of information deriving damage functions, as ST-RISK™ does. In this way ST-RISK™ is unique and more appropriate for use by anyone interested in understanding seismic risk for individual buildings.