Insurers are under constant pressure--pressure to reduce the time-to-market for new products and to introduce new features that decrease risk; pressure to comply with government regulations; pressure to do more with less time. But compute resources haven’t kept pace with the demand.
In the past, reserves and risk-based capital could be calculated using deterministic formulas. With the industry moving towards principal-based reserves, companies are using stochastic methods for pricing and analysis. Stochastic methods combine probability and the interpretation of quantitative data. Many of these calculations iterate through thousands of scenarios and tens of thousands of policies, over a 30 year lifetime. This exponentially increases the amount of computing capacity necessary to perform these jobs. In many cases actuaries model these projections in Microsoft Excel. Desktops simply can’t keep up. Even with the latest multi-core technology and maximum memory allowed, a single pass through a complex scenario on a desktop computer can force a spreadsheet to run, at best, overnight and, at worst, for multiple days before completing on a single desktop machine. The ability to perform these projections within regulated timeframes is a huge challenge.