Prediction Effort - how does that relate to time to complete projects?

Question

It's just a question, because I've looked through an old Software Engineering textbook and one of the things it talks about is measuring effort. But, what I am unclear on is what exactly is "effort" and how does that relate to time it takes to complete a project? I have no experience in dealing with effort estimation because this wasn't something that we went over in great amount of detail in class. For the semester that I took software engineering, most of our class was spent working on UML diagrams and then scrambling at the end of the semester to actually code the thing because we spent most of our time working on UML diagrams. So, I'm hoping that someone who might understands the effort estimation will be able to explain it to me in this thread so that I can put my curiosity to rest. Thank-you in advance.

Answer 1

Certainly this is not something that can be fully answered in a small post... but here are some insights:

First, you should estimate the Size of your project (specifically, of the software to be developed). Some well know metrics are Function Points, Use Case points, and thousands of lines of code (KLOCs).

Then, you use some estimation model (generally, a couple of equations) and input more information about the projects. For instance:

• previous experience with the technology
• complexity overhead due a distributed team
• etc.

This will give the Effort which generally is measured in person-months or man-hours.

Finally, you can user other modeling frameworks (more equations) to figure out how much time and money you will need. For instance, some models allow you to determine the optimum cost/time tradeoff, and also allows you to "play" with the variables. For instance, "how much money will it cost me to finish two weeks later"

Some links:

http://en.wikipedia.org/wiki/Putnam_model

http://en.wikipedia.org/wiki/COCOMO

http://en.wikipedia.org/wiki/Software_development_effort_estimation

Answer 2

You ask four experienced, reliable, professionals. You get an independent, carefully thought-out, reasoned estimate from each.

Then you add them together.