At a former employer we began monitoring the production and defects vs
overtime and found that at above 55 hours per week the defects and errors
outweighed any benefit.  Of course there was a huge range between
individuals.  We worked 10 hours per day and 4 on saturdays but cut back to
50 hours after 3 or 4 weeks as we found the number of defects increased.
After the 3 or 4 week period we made sure the operators had a full weekend
to recover and further cut back to 5 days per week and 10 hours per day and
kept this as a max figure.  On the opposite end of the range we had some
employees that could and desired to work 12 or 14 hours per day for totals
of 60 to 80 hours per week and their work did not suffer.  In the interest
of health however we felt that these kind of hours were not acceptable.
Interestingly during the biggest pushes where we would work 12 hours per
day and 6 to 8 on saturdays after a couple weeks the absentee rate would go
up and by the 3 or 4th week it hit 30%.  Nerves were raw and everyone got
fatigued.  We felt that if over 50 hours per week were required it was time
to increase the staff.  The department of labor (OSHA) I believe has some
statistics on this as well.

Denis Meloche
Heraeus Cermalloy

At 11:18 AM 8/21/97 -0500, you wrote:
>Charles Barker@IO-US
>08/21/97 11:18 AM
>
>Does anyone know of or has anyone on TechNet done any studies relating
>efficiency and actual increase in productivity to increased hours worked?
>
>At what point does the number of extra hours put in per day, times the
>number of consecutive days they are put in, begin to cause the increased
>number of errors and defects generated to wipe out any extra production
>achieved by working the extra time?  In other words, what is the point of
>diminishing returns?
>
>Does anyone have or know where I can find data correlating  to 8, 10, 12,
>14 or even 15 hr days? For 3, 4, 5, or 6 straight days?
>
>TIA
>
>