4 ms·
> They have tried various techniques for estimating time requirements, from wideband delphi to evidence-based scheduling and found that no technique by itself w
by jacques_chester 14y ago
> They have tried various techniques for estimating time requirements, from wideband delphi to evidence-based scheduling and found that no technique by itself works well for the group. Since they are software engineers, "we wrote our own tool", he said with a chuckle, that is a hybrid of several different techniques. There is "no silver bullet" for scheduling, and it is "unlikely you could pick up our method and apply it" to your domain.
It so happens that I'm working on a tool that amongst other things would allow combined PERT 3-point estimates and Delphi Wideband estimates.
However the most important lesson in estimating is: do not rely on one estimation method. Ever. Use as many as you can. For high stakes estimates, try to have at least one of each of the 4 main classes of estimate: judgement, analogical, parametric and decomposition/recomposition. My tool fits in the 4th category, which is IMO currently underserved.
Those of you in the machine learning world will recognise the vibe of ensemble learning.