1

I'm working in image processing and I often have R&D tasks. Sometimes these tasks take a lot more time to do than estimated initially. For example, I recently estimated a task to take 6 hours and it finally took me more than 70 hours...

What is the expected way to deal with these kind of task and EBS ? Am I supposed to set these tasks as "Scheduled tasks" so that they don't affect my estimation history ?

flag
If these types of tasks occur often and you need to account for them in your schedule, why wouldn't you want them to affect your estimation history? – db♦ Oct 1 at 13:57
Because they are outliers in the statistics. EBS is supposed to adapt whether you usually estimate half the time it really takes you to do a task for example. But in this case it is several orders of magnitude of difference. [Joel](joelonsoftware.com/items/2007/10/26.html) said to break down task in 16h chunks max but when doing R&D I don't know what will work and what won't. Actually I could separate my task in smaller ones but I would have to create new ones each time the previous one fails or does not give good enough results. – rold2007 Oct 2 at 9:16
The question is, do you want EBS to be able to predict how long a research task might take you? – Mike Weller Oct 16 at 9:50

Your Answer

Browse other questions tagged or ask your own question.