Planning and estimating are absolutely critical to the success of software development. You need to know the effort involved in carrying out a plan before you can knowledgeable agree to it. For example, you might approve a project that will take 6 months and million to complete but deny one that will take 2 years and million.
But planning is a difficult task. Some groups do no planning which results in a lack of direction and always greasing the squeakiest wheel. Other groups plan to the finest detail and have a hard time adjusting to the inevitable bumps in the road. And even once you come up with a plan, estimating that plan can be even more difficult.
One way of understanding and dealing with variables that are involved in project estimation is to use the cone of uncertainty. The cone of uncertainty is based on an x,y axis that measures the certainty that one can have in their estimation compared to the level that the project is defined.
For example, in the initial product definition phase, there is a lot of certainty, so estimates can vary up to 4 times above or below the estimated time. The only way to drop this level of uncertainty is to define the project further.
Here are the following steps in this cone:
1. Initial product definition
2. Approved product definition
3. Requirements specification
4. Product design specification
5. Detailed design specification
6. Accepted software
As each step is completed, you can more closely estimate the required to complete the project. This estimation will become more and more accurate until you complete the project and know exactly how long it took.
No comments:
Post a Comment