System Safety and Reliability are fundamental to system design and involve a quantitative assessment prior to system development. An accurate prediction of reliability and system safety in a new product before it is manufactured and marketed is necessary as it allows us to forecast accurately the support costs, warranty costs, spare parts requirements, etc.
No, reliability and safety are different system qualities.
The most concise discussion of this difference I have read so far is given in Nancy Leveson's "Engineering a safer world" book ( http://sunnyday.mit.edu/safer-world.pdf ). In section 2.1 from page 7, the assumption "Safety is increased by increasing system or component reliability" is refuted with examples.
It would seem that the obvious answer is that more information is better than less.. What truly matters is the correct use of reliability models and the data. In this sense, although more information is better than less, what matters is that the assumptions (e.g., asymptotic) are tested via simulations and that the data are validated according to well known protocols. Expert opinion are also critical although I suggest using a formal analysis of those opinions in terms of what assumptions the experts may, how credible the experts are, and so on. It is the combination of the sources of information and their formal analysis that matter in answering your question. Good luck, hope this helps.
System reliability depends on functional and quality based factors where as safety is external protection for improvement against failures
I recommend to read the basic definitions of dependability in "Basic concepts and taxonomy of dependable and secure computing", 2004. In the dependability community, safety and reliability are two attributes of dependability. Using these definitions, the answer to your question is : no. You can increase the reliability but stay at a law level of safety (imagine a system with a very low probability of failure, but when it occurs everyone on earth is dead... or the opposite, a super safe system but always failing after 5 min of use...). Actually, safety and reliability are binded to the concept of fault. And you can use a technique, e.g. fault removal (like test, or model checking), which actually increase both attributes, or one. Hope it helps !
In short, NO. More accidents occur due to other issues than equipment failure. More "reliable" equipment may not necessarily have different failure outcomes. It only affects the maintenance schedule. "Unreliable" equipment that is closely monitored and repaired is expensive, but not necessarily unsafe. Putting too much safety reliance on the "reliability" of a piece of equipment can still result in dangerous conditions.
I would say not always, but I'm not in agreement saying no.
Generally speaking how much the reliability impacts safety is an outcome of the system risk analysis that is a very important task if potential safety problem are present. In particular if high severity outcomes are possible as consequence of a component failure and there is no possible independent countermeasure to prevent the potential harm, this is a typical case where the reliability can strongly impact the safety. May be the confusion behind your question is due to the fact that safety is still a subjective concept, I mean, it depends on how do you represent the harms scale. So, if for your system the customer discomfort is an harm, then many reliability issues will impact safety.
There is a few debate and discussion pertaining to this matters (Reliability vs. Safety) that interested me :
https://www.linkedin.com/pulse/direct-correlation-between-reliability-safety-bob-latino
https://www.lce.com/A-Reliable-Plant-is-a-Safe-Plant-1266.html?lipi=urn%3Ali%3Apage%3Ad_flagship3_detail_base%3B1H3nPy1KS%2Fadof86TbHgEA%3D%3D
Hope this will help !
Not always. Safety is directly proportional to viability. And the components of vitality are reliability and stability.
No, reliability and safety are different system qualities.
The most concise discussion of this difference I have read so far is given in Nancy Leveson's "Engineering a safer world" book ( http://sunnyday.mit.edu/safer-world.pdf ). In section 2.1 from page 7, the assumption "Safety is increased by increasing system or component reliability" is refuted with examples.
Dear Jan Reich,
Arthur Schopenhauer said: “Every man takes the limits of his own field of vision for the limits of the world”. and this discussion is an example.
I doubt because Reliability and Safety are two different terms but used interchangeably. This is because, Reliability is 1- failure probability, whereas safety is probability that the system does not fail over time. In my opinion, given the yield, the safety of a system can be enhanced if the reliability of the components within it (depending on the arrangement of components, i.e., series, parallel, k out of n, and so forth) is increased. Without considering sensitivity, they actually go hand-in-hand, a reliable system is a safe system and vice-versa.
However, if sensitivity is considered, a slight sensitivity reduction may improve reliability but might do anything to safety for example- smoke detectors.
Safety functions need to be reliable in the same way the systems need to be reliable. Only that the systems also need to be safe! And, if you cannot have an operational fail-over behavior, when the reliable safety function kicks in to keep you safe, the reliablity of the whole system is affected ( for example: because your system shuts down ).
I agree with Alessandro Vasta that the use of the terms reliability and system is subjective, hence the confusion.
In Leveson's "Engineering a safer world", referenced by Jan Reich , it is thus helpful to look into the definitions used for reliability and safety (p. 10p).
But what I found to be more interesting is that the depicted problems of reliable components and unsafe systems result from specification failures or from ingnoring constraints (take the reliable toxin-releasing plant).
I came across the topic of specification failures a while ago when looking into formal verification of artificial intelligence systems. I think perhaps the following read could be of use for a better understanding of the inherent difficulties when designing the specification for complex systems:
Towards Verified Artificial Intelligence, Seshia et al., 2020
https://arxiv.org/pdf/1606.08514.pdf
Kind regards,
Martin
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability