Use a total time on test (TTT) plot to visualize how well your model fits the data.
The TTT plot provides a way to assess goodness-of-fit for the power-law process.
- If the points on the TTT plot are close to the diagonal (45°) line, then a power-law process or a homogeneous Poisson process model is appropriate. Because the points follow the line, the system is stable.
- If the TTT plot is a curve that is either concave up or concave down, then a power-law process is appropriate. If the curve is concave down, then the system is deteriorating. If the curve is concave up, then the system is improving.
- If the plot shows no pattern, or a curve that alternates between concave up and concave down, then neither the homogeneous Poisson process nor the power-law process is adequate.
For the air conditioning data, the points on the TTT plot are close to the 45° line, which indicates that the power-law process or homogeneous Poisson process provides an adequate fit for the data.