Manual for Life Cost Based FMEA
5.1 Modeling Time
A Monte Carlo simulation is applied to the Life Cost-Based FMEA to perform a sensitivity analysis on the variables associated to failure cost: detection time, fixing time,
delay time, quantity, minimum fixing time
and parts cost.
An example of a triangular distribution with
shown in Figure 4. A triangular distribution using minimum, mode, was used. There are many distribution systems one can use for the
and maximum value simulation; however,
with limited past history data and using estimated selected. There are several commercial software simulation; we have used Crystal Ball v2000.2 in
variables, a triangular distribution was programs for applying a Monte Carlo our analysis.
20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0%
1.52 1.85 2.17 2.50 2.82 3.14 3.47 3.79 4.12 4.44 Fixing Time
Figure 4. Triangular Distribution (Min: 1.2, Mode: 2.5, Max: 4.5)
For each time category, the users must decide what the minimum, mode, and maximum values will be. These values are acquired using empirical data or expert opinion. The steps to applying Monte Carlo simulation in the Crystal Ball software are as follows:
Move the cursor to the cell that requires a distribution analysis
Select “Define Assumption” icon on the menu bar
From the popup window, select “Triangular Distribution” and click “OK” button.
Input the 3 values : Min, Likeliest, Max
Repeat steps 1 through 4 for all unknown variables.
Move the cursor to the cell that sums up the total cost
Select “Define Forecast” button on the main menu
Type in Forecast Name (e.g. “Labor Cost”)
Go to the menu bar and under “Run” select “Run Preferences”
Type “5000” under maximum number of trials and select “OK” button.
Select “Run” under the run menu bar.
Several popup windows should appear that correspond to the forecast name that
was specified in step 8. (Figure 5)
FMEA MANUAL By S. Rhee and C.M. Spencer