Page 293 - Industrial Process Plant Construction Estimating and Man Hour Analysis

Appendix B 273

where
y¼hours required for the nth unit of production
a¼hours required for the first unit
b¼natural slope
The power function y¼ax b is transformed from a curved line on arithmetic
^
scales to a straight line on log-log scales; let
y ¼ log y
a ¼ log a
x ¼ log x

Taking logarithms of both sides, logy¼log a+xlog b appears like y¼a+bx.

Calculating sample size
Sample size: absolute precision

2 ∗ ∗
n ¼ z p 1 pÞ=e 2
ð
where
z ¼ z value (1.96 for 95% confidence level)
p ¼ percentage expressed as a decimal (0.2)
e ¼ acceptable error percentage as a decimal (.05¼ 5%)

Calculating error limits for a sample size
Given sample size is calculated by
2 ∗ ∗ 2

n ¼ z p 1 pð Þ =e
Determine the limit of error, e
∗ ∗ ^
e ¼ z p 1 pðð Þ=nÞ 1=2

Work sampling method
Model: Hs¼(N i ) (Ht) (RF) (1+PF&D)/N
where

Hs¼standard man-hours per task
N i ¼observation of event i
Ht¼total man-hours worked during sample study
RF¼rating factor
PF&D¼personal, fatigue, and delay allowance
N¼number of random observations during sample study