How to calculate Reliability from Failure Data

[AromaticsThailand]

I would like to know what is Reliability of this equipment at the moment if it have failure time interval as belo:

TTF
#1 8376.00
#2 2952.00
#3 3312.00
#4 2616.00
#5 1152.00
#6 888.00
#7 1896.00
#8 3552.00
#9 4824.00
#10 3384.00
#11 5256.00
#12 336.00

Failure Data from over 5 years part and unit of measure in hour.

[Pantelis]

Is it a single item or multiple … if single are you repairing it and putting it back into service… Depending on what the data represents (and what assumptions you make) you can use different reliability analysis methods to derive a model and reliability. The links below provide information on this, as well as tools to assist you.
1. To learn more on Life Data Analysis see: http://www.weibull.com/lifedatawebcontents.htm (http://www.weibull.com/lifedatawebcontents.htm)
2. If a single repairable system this may be more appropriate: http://www.weibull.com/relgrowthwebcontents.htm#Fielded%20Systems (http://www.weibull.com/relgrowthwebcontents.htm#Fielded%20Systems) or more specifically http://www.weibull.com/RelGrowthWeb/Repairable_Systems.htm (http://www.weibull.com/RelGrowthWeb/Repairable_Systems.htm)
You can download free (DEMO) software for the analysis from

Case 1 : http://www.reliasoft.com/Weibull/evaluate.htm (http://www.reliasoft.com/Weibull/evaluate.htm)
Case 2: http://www.reliasoft.com/RG/evaluate.htm (http://www.reliasoft.com/RG/evaluate.htm)

[AromaticsThailand]

These failure data for single equipment that I capture from CMMS

[Pantelis]

As per my prior posting then look at case 2. For this case then you can proceed with either:
"Traditional" Repairable Systems Analysis (see http://www.weibull.com/RelGrowthWeb/Repairable_Systems.htm (http://www.weibull.com/RelGrowthWeb/Repairable_Systems.htm) )
Or
Recurrent Events Data Analysis (see http://www.weibull.com/LifeDataWeb/recurrent_events_data_analysis.htm (http://www.weibull.com/LifeDataWeb/recurrent_events_data_analysis.htm))
Primary difference is what assumptions you are making about each repair action and its effect on the overall system reliability. In the first case you are assuming minimal repair (i.e. the systems reliability is not improved after each repair action), while the second analysis method "allows for" computing and including some improvement (i.e. restoration factor). Further details can be found in the links provided.

[AromaticsThailand]

I try to use failure data manually calculate weibull in Excel, I have one question if R^2 = 0.8652....Can I use weibull function to calculate reliability?

[Arai.M]

Per the previous posts, if we are talking about a repairable equipment, then a Weibull model is not a good assumption. The main assumption there is that you are repairing the component to an "as good as new" state. That is, whenever you have a repair, it is the equivalent of putting a brand new equipment in place. If you are repairing the component, then most likely, regardless of the R^2 value, a weibull distribution is not a good option as we are not talking about identical and independent events.

In the case of a repairable equipment, you are looking at case 2 in the previous post.