Multiple failure mode analysis for Parametric Distribution Analysis (Right Censoring)

In This Topic

Multiple failure mode analysis – parameter estimates
Multiple failure mode analysis – percentiles

Multiple failure mode analysis – parameter estimates

The parameter estimates define the best-fitting parameter estimates for the distribution that you selected for each failure mode. All other parametric distribution analysis graphs and statistics are based on the distribution. Therefore, to ensure that the results are accurate, the distribution that you select must adequately fit the data.

You cannot determine from the estimated distribution parameters whether the distribution fits the data well. Use the distribution ID plot, probability plot, and goodness-of-fit measures to determine whether the distribution adequately fits the data.

Example output

Parameter Estimates

		Standard Error	95.0% Normal CI
Parameter	Estimate	Standard Error	Lower	Upper
Shape	1.97672	0.276587	1.50260	2.60044
Scale	891.929	90.8270	730.552	1088.96

Parameter Estimates

		Standard Error	95.0% Normal CI
Parameter	Estimate	Standard Error	Lower	Upper
Location	5.75328	0.271171	5.22179	6.28476
Scale	1.95933	0.238720	1.54311	2.48780

Interpretation

For the dishwasher data, the engineers selected a Weibull distribution to model spray arm breaks, and a lognormal distribution to model spray arm obstructions. The following parameters define the best-fitting distributions for each failure mode:

Shape = 1.97672 and Scale = 891.929 for spray arm breaks

Location = 5.75328 and Scale = 1.95933 for spray arm obstructions

Multiple failure mode analysis – percentiles

The percentiles indicate the age by which a percentage of the population is expected to fail. Use the percentile values to determine whether your product meets reliability requirements, or to determine which failure modes impact the overall reliability.

Use these values 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 if the distribution adequately fits the data.

Example output

Table of Percentiles

		Standard Error	95.0% Normal CI
Percent	Percentile	Standard Error	Lower	Upper
1	87.0276	30.6339	43.6548	173.493
2	123.896	37.7877	68.1466	225.252
3	152.497	42.3555	88.4796	262.833
4	176.847	45.7243	106.541	293.548
5	198.502	48.3870	123.105	320.077
6	218.260	50.5811	138.583	343.746
7	236.594	52.4406	153.227	365.317
8	253.812	54.0493	167.205	385.279
9	270.130	55.4632	180.636	403.963
10	285.703	56.7217	193.608	421.606
20	417.625	64.8194	308.086	566.111
30	529.457	69.7943	408.905	685.548
40	634.964	74.3928	504.686	798.871
50	740.979	79.9464	599.746	915.471
60	853.343	87.6525	697.736	1043.65
70	979.746	99.1411	803.489	1194.67
80	1134.71	117.529	926.234	1390.11
90	1360.10	152.029	1092.51	1693.23
91	1391.24	157.433	1114.50	1736.69
92	1425.26	163.497	1138.28	1784.59
93	1462.89	170.393	1164.31	1838.05
94	1505.19	178.371	1193.22	1898.73
95	1553.77	187.816	1226.02	1969.15
96	1611.28	199.369	1264.30	2053.50
97	1682.59	214.223	1311.01	2159.50
98	1778.36	235.032	1372.53	2304.18
99	1931.34	270.138	1468.25	2540.49

Table of Percentiles

		Standard Error	95.0% Normal CI
Percent	Percentile	Standard Error	Lower	Upper
1	3.30424	1.78563	1.14571	9.52940
2	5.63679	2.72980	2.18177	14.5631
3	7.91050	3.55915	3.27511	19.1066
4	10.2074	4.33709	4.43857	23.4741
5	12.5595	5.08849	5.67682	27.7867
6	14.9838	5.82646	6.99250	32.1079
7	17.4916	6.55916	8.38765	36.4772
8	20.0913	7.29230	9.86408	40.9221
9	22.7896	8.03022	11.4236	45.4641
10	25.5926	8.77646	13.0681	50.1206
20	60.5984	17.2863	34.6455	105.993
30	112.822	29.6226	67.4371	188.749
40	191.884	49.8160	115.359	319.171
50	315.222	85.4790	185.266	536.337
60	517.841	152.725	290.505	923.079
70	880.729	291.401	460.480	1684.51
80	1639.73	627.451	774.563	3471.28
90	3882.58	1807.19	1559.26	9667.69
91	4360.12	2080.97	1710.97	11111.0
92	4945.69	2424.60	1892.04	12927.8
93	5680.72	2866.84	2112.69	15274.7
94	6631.50	3454.60	2388.85	18409.2
95	7911.58	4269.92	2747.04	22785.7
96	9734.61	5470.91	3235.47	29288.6
97	12561.2	7407.95	3953.98	39904.9
98	17628.0	11054.7	5157.08	60256.0
99	30072.1	20656.8	7824.62	115575

Interpretation

For the dishwasher data, based on the distributions fitted to each failure mode, the engineers conclude the following:

1% of the spray arms fail due to breakage by 87.0276 cycles
1% of the spray arms fail due to obstruction by 3.30424 cycles

Overall, by 3.30048 cycles, 1% of the spray arms will fail. For the greatest improvement in reliability, the engineers should focus improvement efforts on minimizing spray arm obstructions.