Page 140 - Safety Risk Management for Medical Devices

Risk Analysis Techniques 119

12.9.5 Ideal Function
This is the intended output of the System—the way the designers designed the
System. Naturally, things don’t always work the way you want them to, as Noise
Factors play a role.

12.9.6 Error States

These are the unintended outputs of the System. They are the result of Noise Factors’
influence on the operation of the System.

12.9.7 Workflow
Use the following steps to create a P-Diagram
1. Identify the intended function of the System under analysis, and the expected
Outputs
2. Identify the Input Signals—what are the inputs to the System? The things that
the System needs in order to produce its output—the Ideal Function.
3. Identify the Error States—what are the ways in which the System can produce
an output which is different from what is expected?
4. Identify the Noise Factors—what are the inputs, or influencers to the System
function, over which you don’t have control?
5. Identify the Control Factors—what are the inputs, or influencers to the System
function, over which you do have control?

P-Diagram analysis provides a systematic way to consider a function’s Error States
and Noise Factors, as well as the methods that can be used to control them.
As in FMEA, the level of granularity of analysis is the analyst’s choice. At the high-
est level “System” would be the entire medical device system. Or, “System” could be
subsystems, or lower level components of the System.
P-diagram analysis can help you identify Hazards and their Causes. By examining
the Noise Factors, you can create a causal chain that would explain how a hazardous
output could be realized. You will likely find overlaps between FMEA and FTA find-
ings and P-Diagram analysis. Ultimately it is your choice as to how many tools to use.
While extra analyses consume more resources, they also reduce the likelihood of miss-
ing some Hazards and their causal chains.

12.10 COMPARISON OF FTA, FMEA

FTA and FMEA are both useful and important analytical tools in risk management.
They both play a role, and it is best if they are used in a complementary fashion. As
each tool has strengths and weaknesses, the combinatorial application leverages the
best that each tool has to offer.