I would imagine a moving barrier, the width of the shed.
Along its length are a series of gates, with funnelling type
walls/fences that 'encourage ' the chickens through the access
ways as the barrier moves slowly along the length of the shed.
As each chicken passes through the barrier big enough for only
one bird a optical sensor establishes that 1 bird is present and a
rapid, multiple sample weighing is performed on the floor plate of
this access way.
The sample weights (say several hundred) for each bird are
averaged to minimize errors and the weight recorded, to be
averaged as the total bird count is completed.
Oh yes... the birds are counted too of course.

This should work.
The birds should get used to the routine of this wall moving slowly
from one end of the shed to another once a day.
_____________________________

Lance Allen
Technical Officer
Uni of Auckland
Psych Dept
New Zealand

www.psych.auckland.ac.nz

_____________________________