SRC Forum - Message Replies
Forum: Reliability & Maintainability Questions and Answers
Topic: Reliability & Maintainability Questions and Answers
Topic Posted by: Reliability & Maintainability Forum
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998
Posted by: D. Gordon
Organization:Reliability Center, Inc.
Date posted: Thu May 10 13:25:06 US/Eastern 2001
Subject: Calculating Availability for Computer Systems
I have been searching for information about calculating availability for computer systems. When systems integrators carry out contracts they often have a system availability target of perhaps 99.8%. The people in companies who justify the figures perhaps use different methods of calculation from one company to another.
An example would be an Allen Bradley PLC system with 100 different modules that have quoted MTBFs, the system has some redundancy etc, what is the overall availability.
Standards exist (eg bellcore) for calculating reliability of components and cards etc, but when it comes to whole systems there doesn't seem to be any real consistancy and scarce information.
Please can you suggest the most comprehensive way of compounding the MTBfs of multipule components to attain an overall availability figure.
Reply Posted by: Bruce Dudley
Organization: Reliability Analysis Center
Date Posted: Tue May 15 9:49:13 US/Eastern 2001
The concept of availability was developed for repairable systems that are required to operate continuously, i.e., round the clock, and are at any random point in time either operating or "down" because of failure and are being worked upon so as to restore the operation in minimum time. In the original concept, a system is considered to be in only two possible states - operating as specified or failed and in repair. Availability is defined as the probability that a system is operating satisfactorily at any random point in time. Therefore, availability is a combination of reliability and maintainability parameters. The availability parameter can be calculated either as achieved availability which considers all down time; corrective, preventive, administrative and logistic or as intrinsic availability which evaluates only corrective maintenance times. The equations are:
A(achieved) = Uptime / Total Time
Where: Uptime is operating time within specified parameters
Total Time is operating time and repair time, which includes corrective, preventive, administrative, and logistic time
A(intrinsic) = MTBF / (MTBF + MTTR)
Where: MTBF is the mean time between failure of the system
MTTR is the mean time to repair each failure
Background information on availability theory is contained in MIL-HDBK-338 "Electronic Reliability Design Handbook".
Reply Posted by: Barry Ma
Organization: Anritsu Company
Date Posted: Thu May 24 12:56:32 US/Eastern 2001
"MTTR is the mean time to repair each failure", said Bruce Dudley in his reply "Availability" 5/19/2001.
This is a very clear and concise definition of MTTR, as opposed to many others. Thank you, Bruce.