3 Warehousing

Logistics
Warehousing
Overview
Examples
Purposes and functions
Strategic decisions
Warehouse layout
Receiving and Shipment Zones: Sizing
Storage Zone: Capacity
Storage Zone: Sizing (Length, Width, Height)
Product Allocation Dedicated Space
Order batching
Order picking
Recent developments
Typical covered warehouse
Purposes and functions
Receiving
Holding
Receiving
Receiving
Receiving
Holding
Holding
Holding
Picking
Picking
Picking
Picking
Batch forming
Shipping
Packaging
Shipping
Strategic issues
Type of storage racks / type of pallet
Material handling
Warehouse lay-out
Storage racks
Standard Pallet Racking

Narrow Aisle Pallet Racking
Double Deep Pallet Racking
Power-mobile pallet racks
Drive-In Pallet Racking
Etc.
Standard Pallet Racking

Offers direct and fast
accessibility to each
individual pallet.
Low space utilisation.
Utilises standard fork
trucks and hand
trucks.
Narrow Aisle Pallet Racking

Offers direct and fast
individual pallet.
The aisle width needs
only to be slightly larger than
the pallet load,
therefore high utilisation of
warehouse space.
Double Deep Pallet Racking

Using a double deep lift truck,
the operator places the back
pallet into the system first,
then places another pallet in
front of it.
This gives twice as many
pallets stored for a given
number of aisles and offers a
similar space utilization to
Narrow Aisle Racking,
however, individual access to
every pallet is compromised.
Mobile Pallet Racking

Pallet racks are mounted on
mobile bases, these travel on
tracks laid into the floor.
Offers direct but slow
individual pallet.
Very high space utilisation.
Drive-In Pallet Racking

Works on a first in /
last out (FILO) basis.
Limited access to
individual pallets.
Maximum utilisation of
warehouse space.
Uses standard pallet
trucks.
Material handling
Manual
Forklift trucks (Straddle truck, Reach truck,
Counterbalanced truck, Free-path narrow-aisle machine, Rising
cabs, Two-deep reach, Stacker cranes
Conveyors
Articulated CB truck
Counterbalanced fork-lift truck
Reach truck
High track stacker truck very narrow aisle
Powered pallet truck
Conveyor
AGV with Conveyor
Warehouse layout
Warehouse layout: Zones

Usually 3 types of zones in a warehouse:
receiving zone(s)
storage zone
(sometimes divided into reserve and forward zone)
shipping zone(s)
Warehouse layout: Issues
Receiving and Shipment Zones: Sizing

Storage Zone: Capacity
Storage Zone: Sizing (Length, Width, Height)
Product Allocation Dedicated Space
Order batching
Order picking
Sizing the Receiving and Shipment Zones
receiving zone usually larger than the shipping zone
determine the number of truck docks

ddaily demand of orders
ttime for load/unload a truck
q(average used) truck capacity
Tdaily time available to load/unload
dt
nD
qT
Storage Capacity
Inventory of product j at time t
dedicated space
each product is stored in a specific area
assign storage location to products
random space: products can be stored anywhere
the products are randomly distributed over the
warehouse
product
1
2
3
4
5
6
7
total space:
batch size
safety stock
1000
200
2500
400
11600
1250
850
300
1750
500
4500
1000
2600
950
md max I j (t )
j 1
mr max I j (t )
t
j 1
space
dedicated
random
1200
700
2900
1650
12850
7050
1150
725
2250
1375
5500
3250
3550
2250
29400
17000
Storage Zone Length, Width, Height (1)
Height is usually determined by the S/R (Storage / Retrieval) technology
m
x , y
wx , w y
n x , n y , nz
number of required stocking positions

occupation of a unit load
width of the side aisles and the central aisle
number of stocking positions in each direction
Lx x wx n x
1
2
Ly y n y w y
average travelling distance :

Lx
Ly
2
x 12 wx nx 12 y n y wy
space requirement : nx n y nz m
wy
1
2
3
4
5
6
2 3
nx 6
4 5
wx
I/O
n y 12
wx
7
8
9
10
11
12
Lx
Ly
Lagrangean relaxation (LR)

min
nx , n y
x 12 wx nx 12 y n y wy
nx n y nz m
such that
LR x 12 wx nx 12 y n y wy (nx n y nz m)
x 12 wx n y nz 0
1
2
y n x n z 0
x 12 wx
n y nz
y
2nx nz
x 12 wx
ny
nx
y
nx n y nz m
n*x , n*y
Storage Subsystem Length, Width, Height (2)
m y
n
*
x
2nz x 12 wx
n*y
2m x 12 wx
y nz
Product Allocation Dedicated Space (1)
divide the warehouse into M (square) grids of same storage

capacity
n
each product
j (j=1,..,n)
requires mj grid squares
m
j M
assume j 1
(add dummy product)
pj is the total loads moved per time unit for product j
t k ...travel time from grid k to port (proportional to the distance)

c jk ...per period travel cost for storing product j in grid k
c jk
pj
mj
tk
Integer program:
x jk number of products j assigned to grid k

n
minimize:
c
j 1 k 1
jk
x jk
subject to
M
x
k 1
n
x
j 1
jk
mj
j 1,..., n
jk
nz
k 1,..., M
x jk {0,1,..., nz } j , k
Without loss of generality (we can always set

that
nz 1
M M ' nz ), we assume
Binary program:
1 assign product j to grid k
0 do not assign
x jk
n
minimize:
c
j 1 k 1
jk
x jk
subject to
M
x
k 1
n
x
j 1
jk
mj
j 1,..., n
jk
k 1,..., M
x jk {0,1}
j , k

1 if k S j
x jk
0 if k S j
where S j ...set of grids assigned to product j
Rewriting objective:
minimize
n
c
j 1
k 1
jk
x jk
j 1
kS j
jk
j 1
pj
kS j
tk
pj
j 1 m j
kS j

Optimal Solution
[Order Grids]: Compute the tk for all grids and sort them in nondecreasing order, such that
t[1] t[2] t[3]
[Order Products]: Put the products in non-increasing order, such
that
p[1] / m[1] p[2] / m[2] p[3] / m[3]
[Assign Products]: For j=1,n assign product j to the first mj grid
squares still available.

Optimality Proof
Take any solution with a different order. Then there must be two
grids stocking products k and l with t[k] < t[l] and p[k] / m[k] < p[l] /
m[l].
Swapping the grids increases cost by
t[k] p[l] / m[l] + t[l] p[k] / m[k] t[k] p[k] / m[k] - t[l] p[l] / m[l]
= ( t[k] t[l] )( p[l] / m[l] p[k] / m[k] )
<0
So, the suggested solution cannot be optimal. Q.E.D.
Example (1)
product
# of grids
total loads moved
required (mj)
per day (pj)
Ratio pj/mj
10
100
10
150
30
160
20
Example (2)
tk
6
1.5
4.5
I/O
Example (3)
tk
5
4
3
5 8
4 7
3 6
products
8
7
6
I/O
3
3
2
3 1
3 1
2 3
1
1
1
2
2
3
2 1
3 1
3 1
1
1
I/O
3
4
5
3 6
4 7
5 8
6
7
8
p[2] / m[2] p[3] / m[3] p[1] / m[1]

Other issues
Clustering of products with correlated demand
Balancing of workload for pickers/machines
Minimising risk of damage
Order batching (orders are combined to batches)

Heuristics
Most start with a seed order and then expand the batch
with orders in close proximity so long as vehicle capacity
is not exceeded.
Different proximity measures: number of common
locations/aisles, sum of distances for each
location/aisles in candidate order to closest
location/aisles already in batch.
Some rules have been proposed that consider due
dates.
Remark: order batching may not always be possible, in
which case each order is retrieved individually.
Order Picking (1)

One-dimensional presentation of Two-dimensional situation!
x
x
Start
Travelling Salesman Problem (TSP):
How to determine the travelling times?
Order Picking (2)

crossovers only allowed at ends of aisles
x
x
Start
Order Picking (3)

Routing methods
S-shape, start aisle 1
Largest gap, start aisle 1
Optimal, start aisle 1
Order Picking (4)

Routing methods
Largest gap: (go into an aisle until the largest
gap) for warehouses without cross aisles.
S-shape: (traverse always through an aisle if
products in it need to be picked) if many
locations have to be visited (more than 3 picks
per aisle). Also to prevent congestion or to
enforce one-way traffic.
Optimal method
Order Picking
So far, we have assumed a Picker-to-product
environment, but sometimes it is Product-to-picker
(Order-to-picker):
AS/RS
Carousel
Walk/ride and pick
system (W/RPS) (may
visit multiple aisles)
Automated Storage Retrieval

Systems AS/RS: S/R machine
for each aisle; pick-up and
delivery station located at the
end of each aisle
Product to picker:
Horizontal carousel
Horizontal carousel
Horizontal carousel
Picking place
What is the optimal route?
Horizontal carousel
Picking place
The optimal route admits at most one turn!
Horizontal carousel
Where to place fast moving (frequently
demand) and slow moving items?
Picking place
Horizontal carousel
Where to place fast moving (frequently
demand) and slow moving items?
Small orders: fast movers close to the
picking place.
Large orders: slow movers close to the
picking place.
Picking place
Horizontal carousel: large orders

Slow movers
Slow movers
x
x
x
x
x
x
x
x x
Fast movers
x
x
x
x
x x
Fast movers
Large orders:
Better to put
slow movers close
to the picking place.
Horizontal carousel: small orders

Slow movers
x
x
Slow movers
x
x
x
Fast movers
Small orders:
Better to put
fast movers close
to the picking place.
Fast movers
Horizontal carousel:
Nearest neighbour heuristic
In practice, the nearest neighbour heuristic is often applied for carousels.
On average, that performs quite well.
But what is the worst case behaviour?
Recent developments
pick-to-light, pick-to-belt
Enabling technologies: Sorting & Dispatch

Sorting
Use of high capacity
sortation systems with
automatic scanning
Ringsorter
Source: VanderLande
Source: PSB, 2000
Crossdocking
In a crossdock, goods arriving
from the vendor already have a
customer assigned, so workers
need only move the shipment
from the inbound trailer to an
outbound trailer bound for the
appropriate destination.
Crossdocking essentially
eliminates the inventory-holding
function of a warehouse while
still allowing it to serve its
consolidation and shipping
functions.
Shipments typically spend less
than 24 hours at the facility,
sometimes less than an hour.
Merge-in-transit
Merge-in-transit is similar to cross-docking, but differs in the way orders
are fulfilled. With merge-in-transit, it is essential to identify all the
component/shipments of a single order in the merging terminal and to
ensure that all the components are delivered at once, delaying the
earliest shipments if necessary.
In a cross-docking situation the emphasis is more on process
efficiency, as the shipments incoming to a terminal are forwarded with
the next delivery to the customer, often regardless of the order they
belong to.
The operational efficiency achievable with merge-in-transit is not as
high as with cross-docking, but it can be economically performed with a
wide product offering and a large customer-base.

3 Warehousing

Uploaded by

Copyright:

Available Formats

3 Warehousing

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

3 Warehousing

Uploaded by

Copyright:

Available Formats

Logistics

Typical covered warehouse

Purposes and functions

Standard Pallet Racking

Standard Pallet Racking

Narrow Aisle Pallet Racking

Double Deep Pallet Racking

Mobile Pallet Racking

Drive-In Pallet Racking

Counterbalanced fork-lift truck

High track stacker truck very narrow aisle

Powered pallet truck

AGV with Conveyor

Warehouse layout: Zones

Warehouse layout: Issues

Receiving and Shipment Zones: Sizing

Sizing the Receiving and Shipment Zones

receiving zone usually larger than the shipping zone

determine the number of truck docks

Storage Zone Length, Width, Height (1)

Height is usually determined by the S/R (Storage / Retrieval) technology

number of required stocking positions

average travelling distance :

Lagrangean relaxation (LR)

Storage Subsystem Length, Width, Height (2)

Product Allocation Dedicated Space (1)

divide the warehouse into M (square) grids of same storage

pj is the total loads moved per time unit for product j

Product Allocation Dedicated Space (2)

t k ...travel time from grid k to port (proportional to the distance)

Product Allocation Dedicated Space (3)

x jk number of products j assigned to grid k

Product Allocation Dedicated Space (4)

Without loss of generality (we can always set

Product Allocation Dedicated Space (5)

Product Allocation Dedicated Space (6)

where S j ...set of grids assigned to product j

Product Allocation Dedicated Space (7)

Product Allocation Dedicated Space (8)

So, the suggested solution cannot be optimal. Q.E.D.

p[2] / m[2] p[3] / m[3] p[1] / m[1]

Product Allocation Dedicated Space (8)

Clustering of products with correlated demand

Balancing of workload for pickers/machines

Minimising risk of damage

Order batching (orders are combined to batches)

Order Picking (1)

Order Picking (2)

Order Picking (3)

S-shape, start aisle 1

Largest gap, start aisle 1

Optimal, start aisle 1

Order Picking (4)

Automated Storage Retrieval

What is the optimal route?

The optimal route admits at most one turn!

Horizontal carousel: large orders

Horizontal carousel: small orders

But what is the worst case behaviour?