A quick assessment of input uncertainty

Authors:
Bruce E. Ankenman;Barry L. Nelson
Affiliations:
Northwestern University, Evanston, IL;Northwestern University, Evanston, IL
Venue:
Proceedings of the Winter Simulation Conference
Year:
2012

Citing 4
Cited 0

Input Distribution Selection for Simulation Experiments: Accounting for Input Uncertainty

Operations Research
Calculation of confidence intervals for simulation output

ACM Transactions on Modeling and Computer Simulation (TOMACS)
Simulation input analysis: collecting data and estimating parameters for input distributions

Proceedings of the 34th conference on Winter simulation: exploring new frontiers
Efficient Nested Simulation for Estimating the Variance of a Conditional Expectation

Operations Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

"Input uncertainty" refers to the frequently unrecognized, and rarely quantified, impact of using simulation input distributions that are estimated or "fit" to a finite sample of real-world data. In this paper we present a relatively simple method for obtaining a quick assessment of the overall effect of input uncertainty on simulation output, and a somewhat more involved follow-up analysis that can identify the largest sources of input uncertainty. Numerical illustrations of both methods are provided.