Académique Documents
Professionnel Documents
Culture Documents
Cumulative-Sum Chart
If each point on the chart is the cumulative history (integral)
of the process, systematic shifts are easily detected. Large,
abrupt shifts are not detected as fast as in a Shewhart chart.
m
maxθ Σ pdf(xi,θ)
i=1
m 1 xi-µest 2
maxµest Σ 1 e2 σ -
i=1 σ 2π
m k
f B(x i) f B(x i)
Sm = Σ log - min k < m Σ log >L
i=1 f G(x i) i=1 f G(x i)
or
f B(x m)
Sm = max (Sm-1+log , 0) > L
f G(x m)
m
S m = Σ (x i - µ 0)
i=1
where µ 0 is the target mean of the process. This can be
monitored with V-shaped limits.
Advantages
The Cusum chart is very effective for small shifts and when
the subgroup size n=1.
Disadvantages
The Cusum is relatively slow to respond to large shifts.
Also, special patterns are hard to see and analyze.
Lecture 14: CUSUM and EWMA 8
EE290H F03 Spanos & Poolla
Example
15 UCL=13.4
10
5
0 µ0=-0.1
-5
-10
LCL=-13.6
-15
0 20 40 60 80 100 120 140 160
120
100
80
60
40
20
0
-20
-40
-60
0 20 40 60 80 100 120 140 160
Lecture 14: CUSUM and EWMA 9
d= 2 ln 1- β
δ
2 α
θ = tan -1 δ δ= ∆
2A σx
d
θ
50 3
40 2
30 C 1
B
20 0
10 A -1
0 -2
-10 -3
0 20 40 60 80 100 0 20 40 60 80 100
Lecture 14: CUSUM and EWMA 13
Tabular CUSUM
A tabular form is easier to implement in a CAM system
SH( i ) = max [ 0, xi - ( µo + K ) + S H( i - 1 ) ]
SL( i ) = max [ 0, ( µo - K ) - xi + S L( i - 1 ) ]
SL( 0 ) = S H( 0 ) = 0
K = ∆/2 H d
θ
H = dσx tan(θ)
Cusum Enhancements
To speed up CUSUM response one can use "modified" V
masks:
15
10
-5
0 20 40 60 80 100
10
-5
-10
0 20 40 60 80 100 120 140 160
sample
UCL = x + 3 σ λ
( 2-λ)n
LCL = x - 3 σ λ
( 2-λ)n
Lecture 14: CUSUM and EWMA 19
relative 1.0
importance 0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 10 20 30 40 50
EWMA 0.6
EWMA 0.1 -> age of sample
15
EWMA Comparisons
λ=0.6 10
-5
-10
0 20 40 60 80 100 120 140 160
15
10
5
λ=0.1
0
-5
-10
0 20 40 60 80 100 120 140 160
sample
Lecture 14: CUSUM and EWMA 21
Summary so far..
While simple control charts are great tools for visualizing
the process,it is possible to look at them from another
perspective: