Facility Location Models
MGO 636 Supply Chain Modeling Design & Optimization
1
Outline
• Background
• Major Factors & Framework
• Network Design Models
– Plant Location Model
– Demand Allocation Model
– Plant Location Model with Multiple Sourcing
– Plant Location Model with Single Sourcing
MGO 636 Supply Chain Modeling Design &
Optimization
2
Distribution Network: Plant, DC, Warehouse
Customer
DC
MGO 636 Supply Chain Modeling Design &
Optimization
3
Network Design Decisions
• Facility role: What facility performs what function?
– Packaging at the manufacturer or warehouse?
• Facility location: Where should facilities located?
– Tim Hortons opened up a coffee shop at Alfiero in 2008.
– UB’s Pharmacy school was relocated to Downtown Campus.
• Capacity allocation: How much capacity should be allocated? – Best Buy in Galleria Mall stocks more products than other two stores. • Supply and market allocation: Who serves whom?
– Market: Wal-Mart, Target, Tops, etc.
– Supply: Local and global supplier, etc.
MGO 636 Supply Chain Modeling Design &
Optimization
4
Outline
• Background
• Major Factors & Framework
• Network Design Models
– Plant Location Model
– Demand Allocation Model
– Plant Location Model with Multiple Sourcing
– Plant Location Model with Single Sourcing
MGO 636 Supply Chain Modeling Design &
Optimization
5
Major Factors
• Strategic: cost vs. responsiveness economies of scale
• Technological:
– Why Intel has manufacturing plants only in 5-6 countries, whereas Coca-Cola has bottling plants all over the world?
• Macroeconomic: tariffs, tax, exchange rate
–
BMW plant in SC, GM plant in TN
• Infrastructure: electricity, phone lines, suppliers
– Factories in China near Shanghai
• Political: stability
• Competitive
– Positive externality: collocation benefits all firms
• Suziki in India develops car suppliers which can also supply Nissan in India
• Shopping Malls
– Negative externality: collocation hurts each firm (see next
MGO 636 Supply Chain Modeling Design &
Optimization
slide)
6
Negative Externality: Hotelling’s Model
Hotelling may won Nobel Prize some day.
0
a a b
1-a-b
1 b Suppose customers (preferences, e.g. sugar content in coke) are uniformly distributed over [0,1] and go to the closest firm.
How much does firm at a get, how about firm at b?
Both firms maximize their market share if they move closer to each other and locate at a = b = 1/2
MGO 636 Supply Chain Modeling Design &
Optimization
7
A Framework for Network Design
COMPETITIVE STRATEGY
INTERNAL CONSTRAINTS
Capital, growth strategy, existing network
PRODUCTION TECHNOLOGIES
Cost, Scale/Scope impact, support required, flexibility
COMPETITIVE
ENVIRONMENT
PRODUCTION METHODS
Skill needs, response time
FACTOR COSTS
Labor, materials, site specific
MGO 636 Supply Chain Modeling Design &
Optimization
GLOBAL COMPETITION
PHASE I
Supply Chain
Strategy
PHASE II
Regional Facility
Configuration
TARIFFS AND TAX
INCENTIVES
REGIONAL DEMAND
Size, growth, homogeneity, local specifications
POLITICAL, EXCHANGE
RATE AND DEMAND RISK
PHASE III
Desirable Sites in Each Region
PHASE IV
Precise Location &
Capacity Allocation in Each Facility
AVAILABLE
INFRASTRUCTURE
LOGISTICS COSTS
Transport, inventory, coordination
8
Outline
• Background
• Major Factors & Framework
• Network Design Models
– Plant Location Model
– Demand Allocation Model
– Plant Location Model with Multiple Sourcing
– Plant Location Model with Single Sourcing
MGO 636 Supply Chain Modeling Design &
Optimization
9
Network Design Models
• Objective
– profit maximization
– cost minimization
• Key Information needed
– Location of supply sources and markets
– Location of potential facility sites
– Demand forecast by market
– Facility, labor, and material cost by site
– Transportation costs between each pair of sites
– Inventory costs by site
– Sales price of product in different regions
– Taxes and tariffs
– Desired response