This Faculty Early Career Development (CAREER) proposes research on a comprehensive input-modeling framework and software for the simulation of stochastic systems with complex interactions and interdependencies. The premise is that searching among a list of input models for the true, correct model is neither a theoretically supportable nor practically useful paradigm for general-purpose input modeling tools. Since simulation inputs form the core of every stochastic simulation and large-scale discrete-event stochastic simulation becomes a tool that is used routinely for the design and analysis of many service, communications, and manufacturing systems, the need for highly flexible input models, which can capture the important features present in data or known to the analyst, and which are easy to use, adjust, and understand for all practical purposes, is more critical than ever. This work will investigate and construct such input models, incorporating complex interactions and interdependencies for multivariate time- series input processes that occur naturally in the simulation of many stochastic systems.

The proposed framework will also result in automated and statistically valid algorithms to fit input models to multivariate time-series data and generate realizations from these models quickly and accurately in order to drive stochastic simulations. It will additionally facilitate fast sensitivity analysis and enhance the use of simulation in the derivation of emergency responses. If successful, the research will substantially improve the fidelity of practical simulation models, leading to improved system representation necessary for accurate results and better decisions. The target applications will include global supply chain design, operational risk modeling, and telecommunication systems and provide the basis for a series of educational activities in stochastic simulation that will elevate this important area to the core of the curriculum in the Operations Management program of the Tepper School of Business at Carnegie Mellon University and encourage business students to use simulation in the decision-making process. The educational plan will additionally include design of simulation courses for the graduate program curriculum, research support for undergraduate students, and mentorship activities.

Project Start
Project End
Budget Start
2006-02-01
Budget End
2011-01-31
Support Year
Fiscal Year
2005
Total Cost
$400,000
Indirect Cost
Name
Carnegie-Mellon University
Department
Type
DUNS #
City
Pittsburgh
State
PA
Country
United States
Zip Code
15213