Issue 28, June 2003

The Effect of Inspection Intervals 

Consider a little known airline, Odyssey Air, that uses inflatable life vests manufactured by ACME Life Vest Company aboard its commercial aircraft. Odyssey Air wants to study and understand the effect of different inspection intervals for these life vests. The vests are stored until they are required for use. Therefore, failures remain dormant until the system is needed or until failed vests are discovered during scheduled inspections. Scheduled inspections involve testing all vests on the aircraft. Vests found failed are discarded and replaced with new vests (thus resulting in a mix of vests of different ages aboard the aircraft). Odyssey Air is contemplating different inspection intervals for these life vests. They wish to study the effect of inspections performed annually, every two years and every three years.

Setup

Past replacement data was utilized and a dormant failure distribution for these vests was obtained with the following properties:

• Weibull life distribution
• Beta = 2.55
• Eta = 6.89 years

One way to approach this using BlockSim is to set up a single block with the given dormant failure distribution. Figure 1 shows the analysis set up in BlockSim.

[Click to Enlarge]

Figure 1: Setting up inspection properties in BlockSim

If a vest is found failed, it is replaced; thus, a corrective action needs to be set for the block. In addition, one can assume instantaneous replacement (i.e. zero duration) since the time to perform the inspection and replace the vest is not of interest for this analysis. Furthermore, and since the vests are replaced with new ones, a restoration factor equal to 1 can be assumed.

The corrective action is not initiated until the vest is found failed. Thus, the corrective action will be based upon an inspection. In BlockSim, the "Upon Inspection" option needs to be selected. In addition, for annual inspections, the inspection would be once a year, and so forth.

Once the problem has been set up, simulation is utilized to see the effect of the inspection intervals. In this case, the instantaneous or point availability, A(t), is what is of interest. Within the context of this example, this will represent the probability that a vest will be operational (non-failed) at a specific point in time.

Annual Inspection

Figure 2 shows A(t) when utilizing annual inspections. As can be seen on the plot, A(t) goes to 1 after each inspection, implying that 100% of the vests are in a non-failed state after the inspection.

Figure 2: Availability for 1 year inspection

From the plot, it can be seen that after 1.5 years, A(t) is approximately 98%. This implies that 2% of the vests on the aircraft are in a failed state at that point in time. Furthermore, the following can be noted:

• The percent non-failed decreases after each inspection.
• The rate of decrease of A(t) keeps on increasing after each subsequent inspection (since non-failed vests are not replaced and the population ages) until a periodic reversal point is reached at which most vests are replaced with newer ones, thus yielding a younger population.

Figures 3 and 4 repeat the analysis using two and three year inspection plans.

Figure 3: Availability for 2 year inspection

Figure 4: Availability for 3 year inspection

As can be seen from the plots, the availability of the life vests changes dramatically as the interval of inspection increases. Based on the selected inspection intervals, Odyssey Air can now select the inspection interval for the life vests that meets their required goals.

Copyright 2003 ReliaSoft Corporation, ALL RIGHTS RESERVED