Numerical solutions for Bayes sequential decision problems
SIAM Journal on Scientific and Statistical Computing
A fully sequential procedure for indifference-zone selection in simulation
ACM Transactions on Modeling and Computer Simulation (TOMACS)
Simulation Budget Allocation for Further Enhancing theEfficiency of Ordinal Optimization
Discrete Event Dynamic Systems
New Two-Stage and Sequential Procedures for Selecting the Best Simulated System
Operations Research
New developments in ranking and selection: an empirical comparison of the three main approaches
WSC '05 Proceedings of the 37th conference on Winter simulation
Update on economic approach to simulation selection problems
Proceedings of the 40th Conference on Winter Simulation
| Hi-index | 0.00 |
This paper summarizes a new approach that we recently proposed for ranking and selection problems, one that maximizes the expected NPV of decisions made when using stochastic or discrete-event simulation. The expected NPV models not only the economic benefit from implementing a selected system, but also the marginal costs of simulation runs and discounting due to simulation analysis time. Our formulation assumes that facilities exist to simulate a fixed number of alternative systems, and we pose the problem as a "stoppable" Bayesian bandit problem. Under relatively general conditions, a Gittins index can be used to indicate which system to simulate or implement. We give an asymptotic approximation for the index that is appropriate when simulation outputs are normally distributed with known but potentially different variances for the different systems.