8.6 Need for Validation
These assumptions, simplifications, oversights, and limitations cannot be so gross as to prevent the model
from representing reality at a given confidence level. If any of these potential problems do prevent the
model from representing reality, the practitioner may have serious problems. The validation process
assists the practitioner in knowing whether it is appropriate to proceed with the simulation study or go
back to the drawing board. Unfortunately, if you do have to go back to the drawing board, validation
does not have the ability to tell you specifically where to look.
8.7 Two Types of Validation
There are two major types of validation of interest to the simulation practitioner. The first of these is
face validity. Face validity means that the model, at least on the surface, represents reality. The second is
statistical validity. Statistical validity involves a quantitative comparison between the output performance
of the actual system and the model (Law and Kelton, 2000). The simulation practitioner must achieve
both types of validity to have confidence that the model is accurate.
8.8 Face Validity
Face validity is normally achieved with the assistance of domain experts. A domain expert is simply an
individual or group of individuals who are considered knowledgeable on the system under study. Provided
that the same group of people who commissioned the simulation study in the first place are knowledgeable,
it is of great benefit for the practitioner to use this group as domain experts. This approach helps:
• Instill a sense of ownership in the model
• Prevent last-minute “why didn’t you…” questions
• Reduce the number of project progress inquiries