Infrastructure Life Expectancy: Frequency Distribution?

[TheRobsterUK]

I'm currently working on a costing model for water storage tanks. The type of tanks I'm looking at have a certain life expectancy but due to a limited number of installations there isn't much empirical (observed) data about how long they can actually be expected to last before needing replacement.

I have collected estimates from a number of manufacturers who have each given an expected range of life expectances. Generally these are between about 25-50 years, as shown below:

http://www.sudsolutions.com/misc/tanks.JPG

Now what I'd like to do is build some sort of Monte Carlo simulation algorithm using the data in the table above. But I am not sure how to translate that data into a frequency distribution. Does anyone know a formula that I can plug the above numbers into in order to be able to get some kind of frequency distribution? Or am I going about this the wrong way?

Any ideas appreciated. :)

Cheers
-Rob

 

[FredGarvin]

You might want to post this question in the math forum. It may get more responses. I guess the place I would start is just to calculate the mean and SD of each and plot a standard deviation curve for each range. I did a quick histogram and the distributions don't look normally distributed though. This is where someone well versed in stats can really help. With such a small sample to pull from, what assumptions can be made that are valid?

 

[TheRobsterUK]

Thanks, I've reposted in the General Math forum as advised.

https://www.physicsforums.com/showthread.php?p=1313697#post1313697

With regards to assumptions, I can't really make any...this is the only data that's available so I don't have much choice other than to work with a small sample and assume that the data is reasonably accurate.