New ACM Transactions on Modeling and Computer Simulation (TOMACS) Area
Bayesian and Decision Methods for Simulation
Stephen E. Chick, Area
Editor
stephen.chick_at_insead.edu
(_at_ is for antispam)
Area Statement
Bayesian statistical and
decision theory techniques have gained attention in the discrete event and
stochastic simulation community, as a result of their potential to improve
the efficiency and quality of decisions that are based upon simulation
analysis. Examples include the use of Bayesian probability distributions to
more fully quantify input parameter and model uncertainty, experimental
designs to reduce that uncertainty, techniques to improve the efficiency of
ranking and selection/optimization procedures, response surface modeling, the
estimation of conditional expectations, inverse problems, and multi-attribute
methods.
The Bayesian and
Decision Methods for Simulation Area welcomes papers that extend the theory
of Bayesian methods to apply to the special sampling and modeling
environments found in discrete-event, stochastic and system dynamics
simulation models. The area also welcomes papers that show innovative uses
of Bayesian or decision-theory tools in cutting-edge applications, such as
health care, service delivery, telecommunications, etc. Successful
application papers will identify unique features of the problem that require
new approaches to decision making (e.g., uncertainty analysis in high-profile
policy contexts), and go beyond the application of standard tools to
well-known problems. Simulation has been widely applied to solve problems in
Bayesian inference (such as Markov Chain Monte Carlo methods for exploring
posterior distributions). Papers that use or develop MCMC methods are
appropriate for this area to the extent that they apply Bayesian methods to
address decision problems that are studied with discrete-event, stochastic or
system dynamics simulation models (as opposed to focusing on the application
of MCMC simulation to Bayesian models).
When preparing a paper
for the Bayesian Methods for Simulation area, authors are encouraged to
describe the extent of generality of the proposed methods.
Application-specific papers should provide sufficient background for a
general reader to appreciate the key features of the problem, and enough
detail for the reader to be able to fully comprehend the model (or at least
references to sufficient detail) and the solution approach.
Updated: Aug 2006