You are on page 1of 3

Queuing Theory Equations

Definition
λ = Arrival Rate
μ = Service Rate
ρ=λ/μ
C = Number of Service Channels
M = Random Arrival/Service rate (Poisson)
D = Deterministic Service Rate (Constant rate)

M/D/1 case (random Arrival, Deterministic service, and one service channel)
Expected average queue length E(m)= (2ρ- ρ )/ 2 (1- ρ)
2

Expected average total time E(v) = 2- ρ / 2 μ (1- ρ)

Expected average waiting time E(w) = ρ / 2 μ (1- ρ)

M/M/1 case (Random Arrival, Random Service, and one service channel)
The probability of having zero vehicles in the systems Po = 1 - ρ

The probability of having n vehicles in the systems Pn = ρ Po


n

Expected average queue length E(m)= ρ / (1- ρ)

Expected average total time E(v) = ρ / λ (1- ρ)

Expected average waiting time E(w) = E(v) – 1/μ


M/M/C case (Random Arrival, Random Service, and C service channel)

ρ
Note : must be < 1.0
c

The probability of having zero vehicles in the systems


_1
⎡ c −1 ρ n ρC ⎤
Po = ⎢∑ + ⎥
⎣ n = 0 n! c!(1 − ρ / c ) ⎦

The probability of having n vehicles in the systems


ρn
Pn = Po for n < c
n!
ρ n
Pn =Po n − c
for n > c
c c!

Expected average queue length


ρ c +1 1
E(m)= Po
cc! (1 − ρ / c )2

Expected average number in the systems


E(n) = E(m) + ρ

Expected average total time E(v) = E(n) / λ

Expected average waiting time E(w) = E(v) – 1/μ


M/M/C/K case (Random Arrival, Random Service, and C service Channels and K
maximum number of vehicles in the system)
The probability of having zero vehicles in the systems
−1
⎡ ⎛ ⎛ ρ ⎞ K −c +1 ⎞⎤
⎢ c −1 ⎜1− ⎜ ⎟ ⎟⎥
ρ ⎛ ⎞ ⎛ ρ c
⎞⎜ ⎟⎥
Po = ⎢∑ ⎜ ρ n ⎟ + ⎜⎜ ⎟⎟⎜ ⎝ ⎠
1 c
For ≠1 ⎟⎥
c ⎢ n =0 ⎝ n! ⎠ ⎝ c! ⎠ ρ
⎢ ⎜ 1 − ⎟⎥
⎜ c ⎟
⎢⎣ ⎝ ⎠⎥⎦

−1
ρ ⎡ c −1 ⎛ 1 ⎞ ⎛ ρ c ⎞ ⎤
For =1 Po = ⎢∑ ⎜ ρ n ⎟ + ⎜⎜ ⎟⎟(K − c + 1)⎥
c ⎣ n =0 ⎝ n! ⎠ ⎝ c! ⎠ ⎦

1 n
Pn = ρ Po for 0 ≤ n ≤ c
n!
⎛ 1 ⎞
Pn = ⎜ n -c ⎟ ρ n Po for c ≤ n ≤ k
⎝ c c! ⎠

⎛ρ⎞
Po ρ c ⎜ ⎟ ⎡
⎛ρ⎞ ⎤
k − c +1 k −c
⎝ c⎠ ⎛ρ⎞ ⎛ ρ⎞
E ( m) = 2 ⎢
1− ⎜ ⎟ − ⎜1 − ⎟(k − c + 1)⎜ ⎟ ⎥
⎛ ρ ⎞ ⎣⎢ ⎝ c ⎠ ⎝ c⎠ ⎝ c ⎠ ⎦⎥
c!⎜1 − ⎟
⎝ c⎠

c −1
(c − n ) ρ n
E (n) = E (m) + c − Po ∑
n =0 n!

E ( n)
E (v ) =
λ (1 − PK )

1
E ( w) = E (v) −
μ

You might also like