الفهرس | يوجد فقط 14 صفحة متاحة للعرض العام |
المستخلص Discrete-event simulation concerns the modeling of a system as it evolves over time by a numerical (computer-based) representation in which the state variables change at only a countable number of points in simulation time. This technique can be used to model a real or proposed system so that the behavior of the system can be studied under specific conditions. Most simulations can be classified as either finite-horizon or steady-state. Finite-horizon (terminating) simulation models are ended at a specific time or by the occurrence of a specific condition. On the other hand, steady-state (non-terminating) simulations operate (at least conceptually) into the indefinite future; and in this case interest centers on long-run average performance. Analysis of the outputs generated by steady-state simulation models is the focus of this research. Three fundamental problems arise in analyzing output from a stochastic steady-state simulation. The first problem is caused by a transient in the initial sequence of responses that is due to the system’s starting condition. It is usually impossible to start a simulation in steady-state operation, thereby making it necessary to do the following: (a) start the simulation in some convenient initial condition that may not be typical of steady-state operation; and (b) select the duration of the warm-up period (i.e., the data-truncation point or statistics clearing time) so that beyond the warm-up period, the mean of each simulation-generated observation is sufficiently close to the steady-state mean. Initialization bias can cause not only a grossly misleading simulation-based point estimate of long-run system performance but also wildly optimistic indications of the inherent accuracy and reliability of the simulation-based results, where the term accuracy refers to the magnitude of the estimation error and the term reliability refers to the CI’s nominal (user- |