Design of Simulation Experiments FOR SIMULATION MODELING Overview
Simulation models often have many input factors and determining which ones have a significant impact on performance measures (responses) of interest can be a truly daunting task. The common approach of changing one factor at a time is very often incorrect, because for many models factors interact to impact on the responses. In this course we give a comprehensive presentation of design of experiments (DOE) specifically for simulation modeling, whose major goal is to determine which factors have the greatest effect on the responses. Another important use of DOE is to develop a surrogate model (a simplified model of the simulation model) based on the important factors to predict the model response for factor-level combinations that were not actually simulated due to execution-time or setup-time constraints, or because a prediction is needed in real time. A surrogate model can also be used to find the factor combination that optimizes the simulation response.
We discuss a simple and widely applicable approach for determining significant factors in the context of simulation modeling. Methods designed for physical experiments, which are discussed in university courses or implemented in statistical software, make assumptions that are rarely satisfied in practice. Students will analyze simulation-response data in class using a leading DOE software package, and be prepared to apply their knowledge the following week at work.
Each attendee will receive a copy of the book Simulation Modeling and Analysis (5th Edition, McGraw-Hill, 2015) by Dr. Averill M. Law as part of their registration fee – this book is widely considered to be the “bible” of simulation with more than 163,000 copies in print and 15,700 citations. Versions of this seminar have been presented to Lockheed Martin, Middle East Technical University/Roketsan (Turkey), NSA, Sasol Technology (South Africa), and the U.S. Navy.
What You Will Learn (look to the right for more detail):
- Factorial Designs for Determining Important Model Factors
- Fractional Factorial Designs for Reducing the Required Number of Simulation Runs
- Classical Surrogate Models Based on Central-Composite Designs and Regression for Prediction or Optimization
- Simulation Surrogate Models Based on Space-Filling Designs and Kriging (Gaussian Process Modeling)
- Critical Dangers of Using Standard Experimental Designs and Analyses for Simulation Modeling
- Commercial Software for DOE That Are Appropriate for Simulation Modeling
- Numerous Examples and In-Class Exercises to Illustrate the Mechanics and Applications of DOE