Monte Carlo Simulation
Using repeated probabilistic sampling to quantify uncertainty in availability, production, and lifecycle-value outcomes.
Definition
Monte Carlo simulation is a probabilistic simulation technique that uses repeated random sampling from input distributions to quantify uncertainty in an outcome. Instead of producing a single deterministic answer, it produces a distribution of possible outcomes — commonly summarized as P50, P80, and P90 values — that expresses how confident a forecast actually is.
Why a Single Number Is Rarely Enough
Failure rates, repair times, commodity prices, and production rates are rarely fixed values — they vary within a range. A deterministic calculation that plugs in average values for each input systematically understates risk, because it ignores how variability compounds across a system. Monte Carlo simulation addresses this by sampling each uncertain input thousands of times according to its actual distribution, running the full model for each sample, and reporting the resulting spread of outcomes.
A P80 availability forecast, for example, states the availability level the system is expected to exceed 80% of the time — a materially more useful planning figure than a single average-case estimate.
Where It Is Applied
Monte Carlo methods are used to quantify uncertainty in availability models, production forecasts, project cost and schedule estimates, and lifecycle-value calculations — anywhere a decision depends on a range of plausible outcomes rather than a single expected case.
How This Connects to Knar Global's Work
Knar Global uses probabilistic simulation within its RAMgen and Industrial Digital Assets methodologies to express availability, production, and lifecycle-value outcomes as ranges rather than single-point estimates — making the resulting expected value, and the confidence behind it, explicit for capital and operating decisions.