What is Probability of Failure on Demand (PFD) in instrumentation

What is Probability of Failure on Demand in instrumentation


The aspect to be verified is the Probability of Failure on Demand (PFD). The PFD of the complete SIS loop including the initiator, logic solver and final element shall be calculated.
The calculated PFD value should be verified as better than the minimum required PFD value as shown in the Table 1 by a factor of 25%.
PFD calculation is based on the Simplified Equation of ISA-TR84.02 Part-2. Within the calculation, the common cause factor and systematic error factor are considered.

Definitions of Terms Used for PFD Calculation

SIS: The Safety Instrumented System, which is generally composed of sensor, logic solver and the final element.
β: The fraction of single module or circuit failures that result in the failure of a Additional module or circuit. Hence two modules or circuits performing the same function fail. This parameter is used to model common cause failures that are related to hardware failures. (Dimensionless)
MTTF: The mean time to the occurrence of a failure. (Hours)
PFD: The probability of the SIS failing to respond to a process demand, it is commonly referred to as the probability of failure on demand. It is also  referred to as safety unavailability or fractional dead time. (Probability)
TI: The time interval between periodic off-line / on-line testing of the system or an element of the system. Dimension (Time)
XSA: The scoring factor for sensors and actuators and provides credit for automatic diagnostics that are in operation. (Dimensionless)
YSA: The scoring factor for sensors and actuators that corresponds to those measures whose contribution will not be improved by the use of automatic diagnostics. (Dimensionless)
Z: Frequency and level of coverage of automatic diagnostics factor. (Dimensionless)
λDD: The dangerous (e.g., fail in a direction that would defeat the purpose of the SIS), detected failure rate for a module or portion of a module. Detected dangerous failures are failures that can be detected by on-line tests.  (Failures per hour)
λDU: Is used to denote the dangerous undetected (covert) failure rate for a module or a portion of a module. Dangerous undetected failures cannot be detected by on-line tests. (Failures per hour)
λFTO undet: The rate failures causing Fail-To-Operate (FTO) failures, undetectable by automatic self-test. (Failures per hour)
TIF: The probability that a component which has just been functionally tested will fail on demand, (applies for FTO failures only), and Systematic error. (Probability)
DCPST: Diagnostic coverage factor by contribution of partial stroking test.

2 thoughts on “What is Probability of Failure on Demand (PFD) in instrumentation”

  1. I would like to thank you for the efforts you have put
    in penning this blog. I’m hoping to check out the same
    high-grade blog posts by you later on as well. In fact, your creative writing abilities has inspired me to get my very
    own website now 😉

    Reply

Leave a Comment

error: Content is Protected.