Input model risk, which is also called “input uncertainty,” refers to the effect of not knowing the true, correct distributions of the basic stochastic processes that drive a computer simulation . Examples of input models include interarrival-time and service-time distributions in queueing models; bed occupancy and patient characteristic distributions in healthcare models; distributions for the values of underlying securities and assets in financial models; and time-to-failure and time-to-repair distributions in reliability models. When the input distributions are obtained by fitting to observed real-world data, then it is possible to quantify the input model risk, or to choose input models that hedge against this risk. In this chapter, we define input model risk and describe various proposals for addressing it that had their origins at the Winter Simulation Conference.
|Original language||English (US)|
|Title of host publication||Advances in Modeling and Simulation|
|Subtitle of host publication||Seminal Research from 50 Years of Winter Simulation Conferences|
|Editors||Andreas Tolk, John Fowler, Guodong Shao, Enver Yücesan|
|Publisher||Springer International Publishing|
|Number of pages||18|
|State||Published - 2018|