SIS Book - Chapter 08 - PFDavg With IEC 61508 Formulas
SIS Book - Chapter 08 - PFDavg With IEC 61508 Formulas
RAMS Group
Department of Mechanical and Industrial Engineering
NTNU
(Version 0.1)
Learning Objectives
The main learning objectives associated with these slides are to:
I Introduce and explain the simplified formulas in IEC 61508, part 6 for
calculating the average probability of failure on demand (PFDavg )
I Introduce and discuss some of the related concepts and assumptions
Outline of Presentation
1 Introduction
5 PFD formulas
6 Inclusion of CCFs
7 Markov
Remark
IEC 61508 formulas may be explained in different ways, and the approach
selected in the following slides is not the only one. A more thorough
explanation of background for formulas in IEC 61508 is provided by Dr.
Fares Innal in his PhD thesis:
I Innal, F. (2008) Contribution to modeling safety instrumented systems
and to assessing their performance. Critical analysis of IEC 6|508
standard. PhD thesis. University of Bordeaux, Bordeaux, France.
General Assumptions
IEC 61508 (part 6) presents simplified formulas for selected voted configurations,
derived with basis in the following assumptions:
(G)
PFDavg = λ D,G tGE
where λ D,G is the Group failure frequency of dangerous failures, λ D,G and tGE is the
Group-equivalent mean downtime. Assumptions underlying formulas:
I Any parallel structure of channels constitutes identical components.
I Time to failure and repair times are exponentially distributed.
Visualization I: RBD
Recall that the average unavailability. Aavg , (from Chapter 5) of a system could be
expressed as:
Aavg = λ S · MDTS
where the system is represented by as “super item” with failure rate λ S mean
downtime MDTS .
In IEC 61508 formulas, Aavg corresponds to PFDavg , λ S to λ D,G and MDTS to tGE .
PFDavg=λD,G.tGE
λD,G.tGE
λDU λDD
Channel
MDT1 MDT2
We note that:
tCE = λD
{
λ DU
τ
2
+ MRT + λ DD
λD
tCE
· MTTR
where λ D = λ DU + λ DD
Lundteigen& Rausand Chapter 8.PFD formulas in IEC 61508 (Version 0.1) 8 / 26
Equivalent Mean Downtime
λ DU τ λ DD
!
tCE = + MRT + MTTR
λD 2 λD
λ DU τ λ DD
!
tGE = + MRT + MTTR
λD n − k + 2 λD
λ DU τ λ DD
!
tGjE = + MRT + MTTR
λD j + 1 λD
For the group we assume that the downtime of DD failures is not affected by how
many channels being down.
PFDavg = λD,G.tGE
(c)
λD,G
(i)
λD,G
nλD
tCE
tG(n-k)E (c)
(i)
tGE
tGE
1oo3 system
Upon first failure: F
OK
τ
F
τ
F
τ
The dangerous group frequency, λ D,G , is the average system failure rate.
Average unavailaility of
λDU λDD ONE channel
1oo2 system
2λD.tCE
Dangerous frequency of
{
tCE TWO channels:
(2λD.tCE) .(λD)
First channel
is down
Second channel
fails
“Super-item” λD,G = 2λ2D,G.tCE
tGE
λ DU τ λ DD
Note that tCE is as before and tGE = λD 3 + MRT + λD · MTTR
I The first failure occurs with rate nλ D . The probability that the second failure
occurs while the first one is down is (n − 1)λ D tCE , the probability that a third
failure occurs while two are down is (n − 2)λ D tGE2 , and so on till the group
failure (involving n-k+1 failures) occurs with probability that
(n − k + 1)λ D tGE(n−k) .
n−k
Y
(koon)
λ D,G ≈ λn−k+1
D k (n − j + 1)tGjE
j=1
λ DU τ λ DD
tGjE = ( + MRT ) + MTTR
λD j + 1 λD
Note that tG1E is what we previously have defined as tCE .
What is PFDavg ?
Recall that the general equation for PFDavg was:
Inclusion of CCFs
IEC 61508 includes the contribution from CCFs due to DU as well as DD failures
using the standard beta factor model.
I The fraction of the DD failure rate that is CCFs is denoted β D :
(i) (c)
λ DD = (1 − β D )λ DD and λ DD = β D λ DD
Inclusion of CCFs
CCFs are included as a virtual functional block in the reliability block diagram, one
for the CCFs associated with DD failures and one for CCFs associated with DU
failures.
1 2
DUind DDind DUind DDind
1 3
DUind DDind DUind DDind CCFDU CCFDD
2 3
DUind DDind DUind DDind
Inclusion of CCFs
Because CCFs are represented by single functional blocks, it is rather straight
forward to set up the contribution toPFDavg from CCFs (note that λ D(c) = λ DU
(c) (c)
+ λ DD :
λ (c) τ ! (c)
λ DD
CCF
PFDavg = (c)
λ D(c) tCE =λ D(c) DU
+ MRT + MTTR
λ D(c) 2 λ D(c)
τ
!
= βλ DU + MRT + β D λ DD MTTR
2
The contribution to PFD from the independent part remains as before, except that
the fraction (1 − β ) and (1 − β D ) must be extracted for the DU and DD failure rate
respectively.
The same formulas may be derived using Markov methods, (or more precisely using
approximate Markov steady state models based on multiphase Markov models).
Consider figure 3.19 in Fares Innal master thesis. This diagram can be modified for
different architectures, but this one is used to illustrate derivation of formula for a
xoo3 system.
Note: The repair/restoration rates (µ i ) are not all going back to state 1. For the
approximations on the next slide, the result would be the same once truncated.
Lundteigen& Rausand Chapter 8.PFD formulas in IEC 61508 (Version 0.1) 24 / 26
Markov
For a 1oo3 system, the PFDavg corresponds to the steady state probability
of state 4:
6λ D 3 6λ D 3
PFDavg = P4 (∞) = ≈
6 λD 3 + 6 λD 2 µ3 + 3 λD µ2 µ3 + µ1 µ2 µ3 µ1 µ2 µ3
For a 2oo3 system, the PFDavg corresponds to the steady state probability
of state 3 and 4 (but probability of state 3 is the dominating):
6λ D 2 µ 3 6λ D 2
PFDavg ≈ P3 (∞) = ≈
6 λD 3 + 6 λD 2 µ3 + 3 λD µ2 µ3 + µ1 µ2 µ3 µ1 µ2
Note: These equations may be time consuming to derive by hand. Here, the
software Maple was used to derive symbolic equations.
Meaning of µ i