Minitab displays summary information about the current data:
The total number of units is the number of units shipped during the data collection period.
The observed number of failures is the total number of failures (or warranty claims) that actually occurred during the data collection period.
The expected number of failures is the total number of failures that the model predicts during the data collection period. If the expected number of failures differs substantially from the observed number of failures, the chosen distribution may not provide a good fit to the data, especially if the Least Squares estimation method is used.
The 95% Poisson CI is a 95% confidence interval for the expected number of failures.
The number of units at risk for future time periods is the number of shipped units that did not fail during the data collection period. The number of units at risk equals the total number of units minus the observed number of failures.
Example output
Summary of Current Warranty Claims
Total number of units
12000
Observed number of failures
69
Expected number of failures
68.5201
95% Poisson CI
(53.2630, 86.7876)
Number of units at risk for future time periods
11931
Interpretation
For the compressor data, the company shipped 12,000 units, of which 69 failed (resulted in warranty claims). The model expected approximately 69 to fail with a 95% confidence interval of (53.2630, 86.7876).
