Productivity gains from using a modelling approach

[Graham_Moir]

I was wondering whether anyone has any hard examples/case studies and figures/numbers that provide a persuasive cost/benefit case for using an end to end modelling tool for software development, whether that's EA or something else.  

Intuitively we all know it's the way to go, but intuition on its own doesn't go down well with senior management.

[qwerty]

Everything you need to known is written in either The Mythical Man Month or in Peopleware. These are the factors that county, not using modeling vs. not modeling.

Probably there are no numbers available since you would need to run a project  twice. You should read The Deadline which probably makes clear why there isn’t such a study.

q.

[Graham_Moir]

Well I think there's a big difference between using modelling and an element of automation, versus manually written Office documents, a couple of standalone Visio diagrams and trying to hold all the information in your head.  Not to mention then doing traceability completely manually in an Excel spreadsheet.  

There is some information out there - see this from late last year,

    http://tinyurl.com/ps54ppx

just wondered if anyone had anything else.

[Robert Sheridan]

Rather than re-running a project how about re-running a workshop modelling some requirements using just Visio and Word and then using EA or an equivalent.  

I have had to use Visio and MS Office recently after years using EA and every simple frustration we encountered we would not have encountered using EA, and that did not include having to manually do traceability.

[Graham_Moir]

Thanks Robert, so from your recent experience, what would you say the productivity loss was in having to go back to Office/Visio?   30% (effort and duration) seems to be the figure that keeps popping up in what I can find on the web.  I realise you probably don't have any actual objective figures, but a reasoned subjective opinion would still be useful.