In: Statistics and Probability
A fuel injector assembly plant that runs two shifts per day
simulated to estimate throughput. To validate the model, the
resulting simulated throughput will be compared against the
observed system behavior. By searching the previous year’s database
records, it was estimated that the average throughput of the fuel
injector assembly plant was 35.6 on a given day. Eight independent
replications of the model were run, each of 30 days’ duration, with
the following results for average throughput of the assembly
plant:
38.9 32.0 37.4 35.1 39.8 31.9 34.2 36.5
(a) Develop and conduct a statistical test to evaluate whether
model output is consistent with system behavior. Use the level of
significance α = 0.05.
(b) If it is important to detect a difference of 2 throughput per
day, what sample size is needed to have a power of 0.90? Interpret
your results in terms of model validity or invalidity. (Use α =
0.05)
a)
Ho : µ = 35.6
Ha : µ ╪ 35.6
(Two tail test)
Level of Significance , α =
0.050
sample std dev , s = √(Σ(X- x̅ )²/(n-1) )
= 2.9596
Sample Size , n = 8
Sample Mean, x̅ = ΣX/n =
35.7250
degree of freedom= DF=n-1= 7
Standard Error , SE = s/√n = 2.9596 / √
8 = 1.0464
t-test statistic= (x̅ - µ )/SE = (
35.725 - 35.6 ) /
1.0464 = 0.119
p-Value = 0.9083 [Excel formula
=t.dist(t-stat,df) ]
Decision: p-value>α, Do not reject null hypothesis
Conclusion: There is not enough evidence that
model output is consistent with system behavior
b)
hypothesis mean, µo = 35.6
Level of Significance , α =
0.05
std dev = σ = 2.9596
power = 1-ß = 0.9
ß= 0.1
δ= µ - µo = 2
Z (α/2)= 1.9600 [excel
function: =normsinv(α/2)
Z (ß) = 1.2816 [excel
function: =normsinv(ß)
sample size needed = n = ( ( Z(ß)+Z(α/2) )*σ / δ )² =
( ( 1.2816 + 1.960
) * 3.0 / 2 ) ²
= 23.01
so, sample size =
24