It's all about the safety PLC-not!

摘要

This week I was asked to join a debate about functional safety. The argument was based on a paper that criticized the accuracy of the Markov model used to calculate failure probability of a safety integrity level (SIL) 3 certified safety programmable logic controller (PLC). I was drawn into the argument. While reading the paper, I concluded that this paper could be right. Even though the argument was against the model that I had made some time ago. The old model appeared to be slightly pessimistic, changing the safety PLC result by as much as 10%. I began investigating to identify the exact differences between my model and why it may have been more accurate. Then, I had a flashback.