Confidence Limits for the Optimum of a Simulation Response Function.
Abstract
This paper deals with stochastic simulation in which the system being studied can be controlled to some extent by setting the value of an input parameter. The system response for a given input will vary from simulation run to simulation run due to uncontrolled and random factors; we thus consider the average or expected response. This is usually called the simulation response surface. Since our work here involves only a single variable the term response curve might be more appropriate. The simulation 'response' will generally be identified with some measure of system effectiveness and the analyst will want to maximize (or possibly minimize) the response by choosing the optimum value for the input parameter.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1977
- Accession Number
- ADA061542
Entities
People
- David W. Robinson
- Michael A. Crane