Mat Prob Pertemuan 4
Mat Prob Pertemuan 4
DISTRIBUSI DISKRIT
Poisson Distribution
l=1
l=5
l = 10
Poisson Distribution
• Expected
Poisson Distribution
6
Poisson Distribution
Probabilitas seorang mengidap penyakit darah yang langka adalah 4 orang per
1000 penduduk.
Pada suatu daerah yang mempunyai populasi 1000 orang, berapakah
probabilitas :
•Terdapat 5 orang yang terkena
•Paling sedikit 10 orang yang terkena
•3 sampai 8 orang yang terkena
Diketahui:
7
Poisson Distribution
0,195 0,195
0,156
0,147
0,104
0,073
0,060
0,030
0,018
0,013
0,005 0,002
0,001
x
Poisson Distribution
0,195 0,195
Tepat 5 orang yang terkena P(x =
5) = ?
0,156
0,147 P(x = 5) = 0,156
0,104
0,073
0,060
0,030
0,018
0,013
0,005 0,002
0,001
x
Poisson Distribution
0,195 0,195
Paling sedikit 10 orang yang
terkenq
P(x ≥ 10) = ?
0,156
0,147 P(x ≥ 10) = 0,005 + 0,002 + 0,001
P(x ≥ 10) = 0,008
0,104
0,073
0,060
0,030
0,018
0,013
0,005 0,002
0,001
x
Poisson Distribution
0,195 0,195
3 sampai 8 orang terkena
P(3 ≥ x ≥ 8) = ?
P(3 ≥ x ≥ 8) = 0,195 +
0,156
0,147 0,195 +
0,156 +
0,104 +
0,104
0,060 +
0,073 0,030
0,060 P(3 ≥ x ≥ 8) = 0,74
0,030
0,018
0,013
0,005 0,002
0,001
x
EXPONENTIAL DISTRIBUTION
DISTRIBUSI KONTINYU
Exponential Distribution
Pada distribusi Poisson, diformulasikan:
dimana
x = jumlah kejadian
= rata-rata kejadian per satuan waktu
λ = 0,5
λ = 0,1
λ = 1,0
Exponential Distribution
• Probability Density Function
• Expected
WEIBULL DISTRIBUTION
DISTRIBUSI KONTINYU
Weibull Distribution
• Dari proses Poisson, peluang dalam kurun t waktu tidak terdapat
kejadian , R(t)~Reliability, diberikan
--- η = 5, β = 1
--- η = 5, β = 2
--- η = 5, β = 3
P(x)
t
0 5 10 15 20
Weibull Distribution
• Probability Density Function
• Expected
Data from a certain call center shows a historical average of 10 calls received
per hour. What is the probability of receiving 0, 10, 20, or 30 per hour in this
center?
We can use the Poisson distribution to describe this process because:
1.The number of calls per hour can take values 0, 1, 2, ….etc. No decimal
numbers can occur.
2.The occurrence of one event does not affect the probability that a second
event will occur. There is no reason to expect a caller to affect the chances of
another person calling, and so the events occur independently.
3.We may assume the average rate (the number of calls per hour) to be
constant.
4.Two calls cannot occur at the same time. It means that at each sub-interval,
like second or minute, either a call occurs or not.
This process is not a perfect fit for the Poisson distribution. For example,
the average rate of calls per hour may decrease in the night hours.
Practically speaking, the process (the number of calls per hour) is close to the
Poisson distribution and can be used to describe the process’s behavior.
Using the Poisson distribution can help us to calculate the probability of 0,10,20 or
30 calls per hour: