coding performance improvement project
November 8, 2003 at 5:44 pm #26005
tan wen weiMember@tan-wen-wei Include @tan-wen-wei in your post and this person will
be notified via email.
hi everyone i need some guidance. i’m currently doin my last year for my course (IT Management) in college. i’m doin an assignment regarding the project entitled coding performance improvement. Mainly it is about creating a more efficient coding scheme to be used in order to save further ongoing maintenance cost of any software developed. The thing is i have only a rough idea of how to estimate the financial benifit of the project. Could anyone please enlighten me in depth how one should go about and calculate the financial benefits. thanks in advance.0November 8, 2003 at 7:16 pm #63178
John J. McDonoughParticipant@John-J.-McDonough Include @John-J.-McDonough in your post and this person will
be notified via email.
I’m assuming that by reducing maintenance cost you intend to save costs by reducing errors introduced in coding, as opposed to improving coding productivity. Perhaps it’s my perspective – coding costs are a fraction of development costs, which in turn are a fraction of lifecycle costs.
If you do some background on testing and software lifecycle cost, you will find that something like 75% of the cost of software is incurred after it’s delivered. I think this is the part you are after. I’d go do some research to get references, but I think Boehm has some of these numbers.
However, of that 75%, most (90%?) is related to requirements errors rather than coding errors. So you are after 7.5% of the total cost. Since lifecycle cost is about 4X development cost, you are looking at reducing the total cost by some fraction of the 30% of development cost that might be available. You would typically use development cost as the basis because that is generally what is estimated up front. Most people want to ignore the lifecycle cost. In spite of extensive documentation from numerous sources, most people also don’t believe that the development cost is only a small fraction of the total cost.
Now, I suspect that there are other factors besides coding errors introduced, and that you can’t hope to get at 100% of the coding errors, so you will need to make some further estimates. But the books on estimating and books on testing ought to give you some clues as to what you might claim. I don’t think it would take too much effort to arrive at a defensible estimate of the fraction of the development cost that you could save of the total cost.
0October 20, 2009 at 10:07 am #65340
The forum ‘Software/IT’ is closed to new topics and replies.