Page 317 - Solid Waste Analysis and Minimization a Systems Approach

17

BENCHMARKING AND EVALUATION

Based on the previous chapters’ waste characterization and significant variable
analyses, performance parameters were established for the 20 waste groups. The
performance parameters were established integrating the theoretical concepts of sta-
tistical quality control into the integrated environmental model. This involved apply-
ing control limits to the output of the regression models used to determine the sig-
nificant variables that influence solid waste. This novel approach to quality control
allowed for the monitoring and control of solid waste generation of U.S. businesses
and government agencies. In particular, confidence intervals were established for the
regression model using a control limit as the level of significance determined by the
t value. These established confidence intervals are the performance parameters for
U.S. company waste generation.
The following are the single variable upper and lower performance parameter
mathematics:

−
ˆ y − t s 1 + (x − ) x 2 <μ < ˆ y + t s 1 + (x − x) 2
0
0
0 α /2 Yx 0 α /2
n S 0 n S
xx xx
where ˆ y 0 = predicted value at x 0
t α /2 = value of t-distribution with n − 2 degrees of freedom.
s = unbiased estimate of standard deviation
n = sample size
x = value for independent variable
0
x = mean value for independent variable
xx ∑
S = n ( x − ) 2
x
i
i=1

295