FMEDA calculation question

[stanley_wu]

Hello, for our system level FMEDA calculation, we need MCU's total safety related failure rate, single point failure rate , and multi point failure rate. But in the S32K144 FMEDA summary sheet, I found that the ∑λSR have deviation from SPF and MPF.

For example of clock module, the ∑(λSPF+λRF) is 0.154 fit, and the DC_spf is 94.68%, then the SPF should be near to 0.154/(1-94.68%) = 2.895 fit. The total safety related failure rate should be 2.895 + ∑λMPF_det (2.717) = 5.612 fit, which is not ∑λSR (2.892).

So if ∑λSR is the total safety related failure rate, why it has so big deviation from λ_SPF + λ_MPF?

I'm looking forward for your clarification!

Thanks in advance!

[nxf51601]

Hello, Following is the calculation done in below query: "For example of clock module, the ∑(λSPF+λRF) is 0.154 fit, and the DC_spf is 94.68%, then the SPF should be near to 0.154/(1-94.68%) = 2.895 fit. The total safety related failure rate should be 2.895 + ∑λMPF_det (2.717) = 5.612 fit, which is not ∑λSR (2.892)". Following below should be the calculation to be done based on ISO26262: In Summary sheet, DC_spf denotes Single point fault metric (SPFM) and for clock module SPFM is 94.68%. As per ISO26262 standard, SPFM is calculated as shown below: 1- ∑(λSPF+λRF)/∑λSR From the statement, “SPF should be near to 0.154/(1-94.68%) = 2.895 fit” you tried to calculate ∑λSR which is 2.895 fit and interpreted as ∑(λSPF+λRF). However, while calculating Total safety related failure rate, you shouldn’t add again ∑λSR which is 2.895 fit and also you missed adding ∑λMPF_undet. In query calculation, you tried to calculate ∑λSR as shown below which is incorrect: ∑λSR = ∑(λSR) + ∑(λMPF_det) + ∑(λMPF_undet) = 2.895 + 2.717 + 0.021 = 5.633fit (which is wrong) Actual method for calculating ∑λSR as shown below: ∑λSR = ∑(λSPF+λRF) + ∑(λMPF_det) + ∑(λMPF_undet) = 0.154 fit + 2.717 + 0.021 = 2.892 fit Kind Regards, Rupal