Scribd
Upload
86 views18 pages

Transportation and Logistics Optimization

This document discusses optimization approaches for one-to-many distribution with identical customers and fixed vehicle loads. It presents a continuous approximation (CA) approach that models the problem as a continuous time-space system to formulate the logistics cost function. The CA approach defines decision variables like headway, load per stop, and stops per tour as functions over the continuous space. It considers costs like motion, pipeline inventory, and holding costs. Upper and lower bounds on the optimal solution are derived by excluding or ignoring the pipeline inventory cost.

Uploaded by

ssregens82
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
86 views18 pages

Transportation and Logistics Optimization

This document discusses optimization approaches for one-to-many distribution with identical customers and fixed vehicle loads. It presents a continuous approximation (CA) approach that models the problem as a continuous time-space system to formulate the logistics cost function. The CA approach defines decision variables like headway, load per stop, and stops per tour as functions over the continuous space. It considers costs like motion, pipeline inventory, and holding costs. Upper and lower bounds on the optimal solution are derived by excluding or ignoring the pipeline inventory cost.

Uploaded by

ssregens82
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

Transportation and Logistics

Optimization
Chapter 4.
One-to-Many Distribution
Identical Customers with Fixed
Vehicle Loads
Assumptions
full vehicles with capacity Vmax
Relax, now v
max
represents actual vehicle
load
know identical demand curve D(t)
N identical customers
When would vehicles not be full?
restrictions on route length
cant travel around filling up vehicles
pipeline inventory: non-negligible
EOQ in One-to-One Dist.
Transportation cost per item

c
]

|
H

+c

Holding cost per item


c

E
1
+c

(E
1
+t
m
) = (c

+c

)E
1
+c

t
m
= c
h
E
1
+c

t
m
Under uniform headway: E
1
= E

Total cost per item

c
]

|
H

+c

+c
h
E

+c

t
m
Optimum headway:
Minimize
c
]

|
H

+c
h
E

Economic Order Quantity (EOQ) in headway


Pipeline Inventory Cost
Decision
Pipeline Inventory Cost

m
: waiting cost of item in the vehicle
Directly dependent of the number of stops
Logistics cost = transportation cost +
holding cost + pipeline inventory cost
Decisions
L: number of dispatches (h:headway)
Dispatch timing {t
l
: l = 1, 2, , L}
n
s
: Number of stops
v:Delivery load per stop
Pipeline Inventory Cost
Location-dependent limits on route length
if pipeline inventory cost is not negligible
Cost per item

m
1
2
d
s
s s
On average, half the total items are in the vehicle

m
c
i
s
d
2
1
2
s s
n
s
2
Pipeline Inventory Cost
Total Pipeline Cost
=
c

s
#itcm - km 2 +
1
2
c

t
s
(#itcm) + c

t
s
(#itcm - stops 2)
=
c
i
s
kNL 6
-
1
2
L
mcx
+2L N t
mcx
2
+
1
2
c

t
s
(N t
mux
) +c

t
s
(
NL
mcx

2
)
Pipeline Inventory Cost per item
Total # of item=N t
mux
t
mux
=
L
mcx
n
s
z
p
= (
c
i

s
+
c
i
t
s
2
) + (
c
i
k6
-
1
2
2s
+
1
2
c

t
s
)n
s
CA Approach
Variable x : an area location in continuous target area
Assume that in the time-space continuum, system
parameters are slow-varying. (Approximation)
Considers the area around point (t, x):

i
t =
i
o x = o
r x = r
Define decision variables/functions
Headway H(t,x)
Load per stop v(t,x)=
i
t E t, x
Stop per tour n
s
(t,x)
Only two decision variables are independent
Let :
mux
be the vehicle capacity

i
n
s
B :
mux
CA Approach
Cost formulation; Cost per item
Motion cost
Pipeline Inventory Cost
Holding Cost
Total Cost (function of v and
s
)
'
'
1
'
1
'
1
'
1
2
2
1
s
s
s s d
s
d m
c
H n D
c
H D
c
H D
k c
H n D
r c z +
|
|
.
|

\
|
+
|
|
.
|

\
|
+
|
|
.
|

\
|
+
|
|
.
|

\
|
= o
s
s i i s i i
p
n
t c
s
k c t c
s
r c
z
|
|
|
.
|

\
|
+ +
|
|
.
|

\
|
+ =
2 2 2
2
1
o
) ( ) ( H c H c c z
h r i s
= + =
|
|
.
|

\
|
+ + +
|
|
.
|

\
|
+
|
|
|
.
|

\
|
+ + + + + =
2
'
' 2 2
1
) (
1
) (
1
) (
1
) 2 (
2
1
2
1
s i i
s
h
s
s i i
s
s s d
s
d
t c
s
r c
c v
D
c
n
t c
s
k c
v n
c
v
c
v
k c
v n
r c z
o
o
CA Approach
Logistics cost function (LCF) per item
CA Approach
Logistics cost function (LCF) per item
Formula (4.20)
CA Approach
Upper Bound
Exclude pipeline in optimization
CA Approach
Lower Bound
Ignore pipeline altogether
Formula 4.20 then decreases with
s
for any v:
always choose largest
s
such that
s mux
CA Approach
The gap is:
CA Approach
Practically, it is often optimal to use full
vehicles
CA Approach
When is near 1, we should not use full
vehicle
Solution
CA Approach

You might also like