Use the probability plot to assess how well the distribution that you selected fits your data. If the points closely follow the fitted line, then you can assume that the distribution fits the data reasonably well.
For the Temp80 sample of the engine winding data, the points appear to follow the fitted line. Therefore, you can assume that the lognormal distribution is an appropriate choice for the data. The fitted line is based on a lognormal distribution with location = 4.09267 and scale = 0.486216.
The survival plot depicts the probability that the item will survive until a particular time. Thus, the survival plot shows the reliability of the product over time.
When you hold your pointer over the survival curve, Minitab displays a table of times and survival probabilities.
Use this plot only when the distribution fits the data adequately. If the distribution fits the data poorly, these estimates will be inaccurate. Use the distribution ID plot, probability plot, and goodness-of-fit measures to determine whether the distribution adequately fits the data.
For the engine windings data, the probability of the engine windings surviving at a temperature of 80° C for at least 50 hours is approximately 60%. The survival function is based on the lognormal distribution with location = 4.09267 and scale = 0.486216.
To describe product reliability in terms of when the product fails, the cumulative failure plot displays the cumulative percentage of items that fail by a particular time, t. The cumulative failure function represents 1 − survival function.
When you hold your pointer over the curve, Minitab displays the cumulative failure probability and failure time.
Use this plot only when the distribution fits the data adequately. If the distribution fits the data poorly, these estimates will be inaccurate. Use the distribution ID plot, probability plot, and goodness-of-fit measures to determine whether the distribution adequately fits the data.
For the engine windings data, the probability of the engine windings failing by 70 hours at a temperature of 80° C is approximately 60%. The cumulative failure function is based on the lognormal distribution with location = 4.09267 and scale = 0.486216.
The shape of the hazard function is determined based on the data and the chosen distribution. When you hold your pointer over the hazard curve, Minitab displays a table of failure times and hazard rates.
Use this plot only when the distribution fits the data adequately. If the distribution fits the data poorly, these estimates will be inaccurate. Use the distribution ID plot, probability plot, and goodness-of-fit measures to determine whether the distribution adequately fits the data.
For the Temp80 variable of the engine windings data, the hazard function is based on the lognormal distribution with location = 4.09267 and scale = 0.486216. At a temperature of 80° C, the hazard rate increases until approximately 100 hours, then slowly decreases.
For the multiple failure data, Minitab displays graphs for each failure mode.
The probability that spray arms will survive breaks for 200 cycles is 95%, and that they will survive obstructions for 1500 cycles is approximately 20%.
The hazard rate for breaks increases slightly over time, but for obstructions decreases over time.