Facility Location Considering Capacity

Lecture Delivered by:

Prof. SP Sarmah
Adding Capacity to the Model
 Introduction
• In previous model, it is assumed that each facility
could be as large as it needed to be to service the
demand assigned to it.
 Formulation considering capacity
• Objective function
Minimize  dist
iI jJ
i, j d jYi , j

• Subject to Y i, j  1; j  J
iI

X
iI
i P

Vol
iI
Y  capi X i ; i  I
i, j i, j

Yi , j  X i ; i  I , j  J

Xi and Yi,j are binary numbers

 Meaning of constraints
• The formulation is very much similar to the “distance only”
problem.

• In this formulation, one new capacity constraint is added.

• This constraint ensures that a warehouse is never assigned

more demand than it can handle.

• If the warehouse capacity is not infinite, this constraint also

will ensure that if a customer is assigned to a warehouse, that
warehouse must be opened.

• In constraint the term capi measures the capacity of warehouse

i.
• The other term voli,j measures the effective volume of demand
for customer j being assigned to warehouse i.
• A distributor in Brazil provide services to 25 customer
regions.

• Each region has a different demand. The total demand is

approximately 100 million units.

• He would like to locate five warehouses to service this

demand.

• First, the model is run without capacity constraint considering

only minimizing weighted distance and the results for the best
five warehouse is shown in the next slide.

• The selected locations for warehouses are Santos, Juiz de Fora,

Recife, Anapolis and Belem.
 Throughput per warehouses
ranges from 7.3 million unit
upto 41.5 million units

No Capacity Model Output

 Analysis
• Now consider the capacity constraint to more evenly balance
the throughput of each facility

• Each warehouse is having maximum capacity of 20 million

units. (total approx. 100/5 = 20 million unit)

• Two scenarios were reviewed to understand the impact.

• In the first case, model is run for the same five warehouses
selected previously.

• The result shows no feasible solution exist.

• The 20-million-unit capacity constraint is forcing each

warehouse to hold almost exactly that amount.
 Analysis
• We have set model in such a way that customer will receive
product from only one warehouse location.

• If the customer demand can not be equally split up in 20

million unit bucket, then this problem becomes infeasible.

• Sao Paulo Region customer actually has a total demand of

approximately 29 million units which can’t be fulfilled by only
one warehouse.

• Therefore, the solution is infeasible.

 Analysis
• INFERENCE: It is to be ensured that not only enough
capacity for the overall demand, but we must have enough
capacity at single instances of the warehouses for specific
sourcing rule (like single sourcing)

• Now consider capacity upto 30 million units per warehouse.

• The solution is compared with the solution of without capacity
model and shown in figure below.
• No capacity solution on left and with capacity solution right.

Due to rearrangement of allocation, average distance travelled from

warehouse to customer increases
Capacity Scenarios Solution Comparison
 Capacity Scenarios Solution Throughput Comparison

• Now, Let the model to select any five warehouses with

capacity constraint.
Original Versus Five- Warehouse Solution Map Comparison

Original Versus Five-Warehouse Throughput Comparison

 INFERENCE
• Capacity constraints don’t necessarily change the
locations of facilities, but they do have the impact of
changing the warehouse to customer assignments.

• Every supply chain has capacity constraints, but

nobody always want to model them.

• Capacity constraint and entities in the model consume

capacity in large chunks, may create a problem which is
very hard to solve.

• Combination of service constraint and capacity

constraint increases the possibility of infeasibility.
COST BASED FACILITY
LOCATION MODEL
COST BASED MODELS FOR FACILTY LOCATION
AND CAPACITY ALLOCATION
Objective :
to maximize the overall profitability of the resulting supply chain network
while providing customers with the appropriate responsiveness
profit = revenue – costs
= (revenue from sale of the product/service) -
(costs from facilities, labor, transportation, material, and
inventories)

Decision Phases
I. Location and capacity of facilities
(long-term)
II. Assignment of current demand to available facilities and
Identification of lanes of transportation
(intermediate-term)
 The Capacitated Plant Location Model
Inputs :
n = number of potential plant location
m = number of markets or demand points
Dj = annual demand from market j
Ki = potential capacity of plant i
fi = annualized fixed cost of keeping factory i open
cij = cost of producing and shipping one unit from factory i to
market j
(cost includes production, inventory, transportation, etc.)

Decision Variables :
yi = 1 if plant i is open, 0 otherwise
xij = quantity shipped from factory i to market j
 The Capacitated Plant Location Model
Objective Function :
Minimize the total cost (fixed + variable) of setting up and operating the
network
n n m
Minimize  f y    c x
i1 i i i1 j1 ij ij
Constraints :
• Demand at each market j be satisfied.

 xij D j for j 1,2,...,m (all markets)

n

i1
• No plant can supply more than its capacity.
m
 xij  Ki yi for i 1,2,..., n (all plants)
j1
• Each plant is either open or closed.
yi ∈{0, 1}
• Non-negativity xij ≥ 0
 Locating Plants and Warehouses Simultaneously

Suppliers Factory Warehouses Markets

(h) Site (i) Site (e) (j)

Fixed cost =Fi Fixed cost = fe

Capacity=Sh Capacity=Ki Capacity=We Demand=Dj

Unit cost =chi Unit cost =cie Unit cost =cej

[numbers] [l] [n] [t] [m]

Decision Variables
yi = 1 if factory is located at site i; 0 otherwise
ye = 1 if warehouse is located at site e; 0 otherwise
xej = quantity shipped from warehouse e to market j
xie = quantity shipped from factory at site i to warehouse e
xhi = quantity shipped from supplier h to factory at site i
19
 Locating Plants and Warehouses Simultaneously

Objective :
To decide on plant and warehouse locations as well as quantities
shipped between various points that minimize the total fixed and
variable costs.

Suppliers Plants Warehouses Markets

(h) Site (i) Site (e) (j)

Fixed cost =Fi Fixed cost =fe

Capacity=Sh Capacity=Ki Capacity=We Demand=Dj

Unit cost =chi Unit cost =cie Unit cost =cej

[l] [n] [t] [m]

n t l n n t t m
Min Fi yi   f e ye   c x 
hi hi c ie iex  c x
ej ej
i 1 e 1 h 1 i 1 i 1 e 1 e 1 j 1

20
 Locating Plants and Warehouses Simultaneously

Constraints :
1. Total amount shipped from a supplier cannot exceed the supplier’s capacity
Supplier (h)

Capacity=Sh n
n  xhi
 xhi  sh for h 1,2,...,l (all supliers) Number=[ l ]
i1
i1

2. Total amount shipped out of a factory cannot exceed the quantity of raw material received
Factory (i)
l t
x 
 hi  xie for i 1,2,..., n (all factories) l t
h1 e1  xhi  xie
h1 Capacity=Ki
e1

3. Total amount produced in the factory cannot exceed its capacity Number=[ n ]

t
 xie  Ki yi for i 1,2,..., n (all factories)
e1

21
 Locating Plants and Warehouses Simultaneously
Constraints :

4. Total amount shipped out of a warehouse cannot exceed the quantity received from the factories
Warehouse
n m
 ie  xej for e 1,2,..., t (all warehouses)
 (e)
x n m
i1 j1  xie  xej
i1 j1
Capacity=We
5. Total amount shipped out of a warehouse cannot exceed its capacity
Number=[ t ]
m
 xej  We ye for e 1,2,..., t (all warehouse)
j1
Customer
6. The amount shipped to a customer must cover the demand (j)
t t
 xej  D j for j 1,2,..., m (all customers)  xej
e1
e1
Demand=Dj

7. Each factory or warehouse is either open or closed Number=[ m ]

y , ye  0,1
i
8. Non-negativity Constraint
xej, xie, xhi ≥ 0
22
ISE | IIT Kharagpur.