Definition
Face Validity is a qualitative validation technique where subject matter experts (SMEs) review the model’s structure, logic, and behavior to determine if it is a reasonable representation of the actual system.
Why It Matters
No matter how statistically sound a model is, it will fail if it is not “believed” by the people who must use it. Face validity is the primary tool for building stakeholder trust. By involving experts in the validation process, you ensure that the model doesn’t just pass a math test, but also passes the “reality check” of those with decades of front-line experience, catching subtle errors that data alone might miss.
Core Concepts
- SME Engagement: Involving the people who operate the system daily (e.g., floor supervisors, dispatchers, technicians).
- Subjective Assessment: Relies on “gut feeling” and experiential knowledge rather than hard statistical data.
- Turing Test: A variation where SMEs are shown data or animations from both the real system and the simulation and asked to identify which is which.
- Scope: Checks if the model captures the “vital few” interactions and doesn’t contain glaring logic errors (e.g., unrealistic speeds or capacities).