UNIT – III

SUPPLY CHAIN NETWORK

Presented by
MOHAN RAJ
SAUBARNIKA
KRISHNAN
HARINI
BABU PRASATH
DEEPAK
DISTRIBUTION NETWORK DESIGN

 Distribution network design is the process of planning the flow of goods from a
manufacturer to a consumer.

 It involves decisions about where to locate warehouses, distribution centers, and

production facilities, as well as how to transport goods and manage inventory.

 A distribution network can be seen as the flow of goods from a producer or supplier to
an end consumer. The network consists of storage facilities, warehouses, and
transportation systems that support the movement of goods until they reach the end
consumer.

 The process of ensuring the consumer receives the product from the manufacturer is
done through direct sales or by following a retail network.
ROLE

Customer Satisfaction: Ensures timely delivery of products.

Cost Efficiency: Reduces transportation and warehousing costs.

Market Reach: Expands access to different markets and customer segments.

Flexibility: Adapts to demand fluctuations and market changes.

Competitive Advantage: Enhances service levels, leading to better market positioning.

A distribution network design role involves a number of tasks, including:

Understanding business objectives :

Distribution network design should align with business objectives and consider other
goals, such as reducing costs, improving service levels, and expanding markets.

Optimizing logistics :

Route planning and optimization can help ensure that goods are transported in a timely
and systematic manner.

Managing inventory :

Inventory management is important for ensuring that the right products are available at
the right time and place.
FACTORS INFLUENCING

Customer Demand: Geographical distribution and volume of demand.

Product Characteristics: Size, perishability, and handling requirements.

Lead Time Requirements: Time needed to deliver products to customers.

Transportation Costs: Mode of transport, distance, and fuel prices.

Warehousing Needs: Location and size of distribution centers.

Technology: Availability of tracking systems, inventory management, and automation.

Service Level Expectations: Desired speed and accuracy of deliveries.

Regulatory Environment: Compliance with laws and regulations in different regions.

While customer service consists of many components, we will focus on those measures
that are influenced by the structure of the distribution network. These include:

Response time:

Response time is the total time between when a customer places an order and
receives delivery.

Product variety:

Product variety is the number of different products / configurations that a

customer desires from the distribution network.

Product availability:

Availability is the probability of having a product in stock when a customer

order arrives.
Customer experience:

Customer experience includes the ease with which the customer can place and
receive their order.

Order visibility:

Order visibility is the ability of the customer to track their order from placement
to delivery.

Returnability:

Returnability is the ease with which a customer can return unsatisfactory

merchandise and the ability of the network to handle such returns.
OPTIONS

Direct Shipping: Products are shipped directly from the manufacturer to the
customer without using warehouses or intermediaries. This reduces inventory costs
but may increase transportation complexity.

Centralized Warehousing: A single distribution center serves all customers. It

reduces inventory management complexity but may lead to longer delivery times
for distant regions.

Decentralized Warehousing: Multiple distribution centers are placed closer to

demand regions, reducing transportation time and costs but increasing warehousing
costs.

Cross-Docking: Products move directly from inbound to outbound transportation

without storage. This speeds up deliveries but requires highly efficient
coordination.
VALUE ADDITION

Cost Reduction: Optimizes inventory levels and reduces operational expenses.

Improved Delivery Times: Enhances the speed of delivery and service reliability.

Scalability: Allows for expansion into new markets with minimal additional
costs.

Customization: Tailors distribution strategies based on regional customer needs.

Risk Mitigation: Diversifies logistics channels and enhances resilience.

MODELS FOR FACILITY LOCATION AND CAPACITY LOCATION:

1. SINGLE FACILITY LOCATION MODEL:

• Objective: Minimize transportation cost between a single facility and

multiple demand points.

• Approach: Uses continuous or discrete optimization to find the best

location.

• Example: Choosing a warehouse location that minimizes delivery

distances.
2. MULTI-FACILITY LOCATION MODEL:

Objective: Minimize total cost (facility operation + transportation) for multiple

facilities.

Approach: Optimizes location for several facilities considering overall cost.

Example: Determining where to place multiple distribution centers in a region.

3. p-MEDIAN MODEL:

Objective: Minimize the total weighted distance between demand points and "p"
facilities.

Approach: Selects "p" optimal facility locations.

Example: Placing "p" hospitals in a city to minimize the total distance patients travel.
4. p-CENTER MODEL:

Objective: Minimize the maximum distance between any demand point and the
nearest facility.

Approach: Focuses on ensuring all points are close to a facility.

Example: Locating emergency services like fire stations to ensure rapid response.

5. CAPACITATED FACILITY LOCATION PROBLEM (CFLP):

Objective: Minimize costs while respecting the capacity limits of facilities.

Approach: Considers both location and capacity constraints.

Example: Locating warehouses with limited storage capacity to minimize overall

costs.
6. CAPACITATED PLANT LOCATION PROBLEM (CPLP):

Objective: Minimize production and transportation costs while meeting facility

capacity limits.

Approach: Optimizes factory locations, considering production capacities.

Example: Choosing factory locations and capacities to minimize manufacturing

costs.

7. HUB LOCATION MODEL:

Objective: Optimize the location of hubs that consolidate shipments from multiple
locations.

Approach: Selects central hubs to minimize transportation costs between nodes.

Example: Designing airline hubs to reduce total flight distances.

8. DYNAMIC FACILITY LOCATION MODEL:

Objective: Optimize location and capacity over time, considering demand changes.

Approach: Incorporates future expansion or closure of facilities.

Example: Planning a distribution network that can expand based on future demand.

9. STOCHASTIC FACILITY LOCATION MODEL:

Objective: Account for uncertainty in demand, transportation costs, or facility

operation.

Approach: Uses probabilistic methods to factor in uncertainties.

Example: Deciding warehouse locations while considering fluctuating customer

demand.
10. ROBUST FACILITY LOCATION MODEL:

Objective: Design networks that perform well across different uncertain future
scenarios.

Approach: Focuses on minimizing risks associated with variability.

Example: Locating distribution centers in areas safe from geopolitical risks.

11. GREEN FACILITY LOCATION MODEL:

Objective: Optimize locations while reducing environmental impact.

Approach: Includes environmental costs, such as emissions and energy use, in the
optimization.

Example: Locating warehouses near renewable energy sources to reduce carbon

footprint.
IMPACT OF UNCERTAINTY ON NETWORK DESIGN

1.DEMAND UNCERTAINTY :

Impact:

Demand fluctuations are common and affect decisions about the size and location of
facilities, inventory levels, and transportation modes.

Mitigation:

Designs may include flexible capacity, scalable infrastructure, or multiple supply and
distribution points to adjust to varying demand.
2.SUPPLY CHAIN DISRUPTIONS:

Impact:

Disruptions such as supplier failure, transportation delays, or geopolitical risks can

impact the availability of materials or products.

Mitigation:

Diversification of suppliers and strategic buffer inventory in various locations help

reduce the impact of these uncertainties.
3.COST UNCERTAINTY:

Impact:

Fluctuating transportation, labor, and material costs can lead to inefficiencies if

the network is too rigid to adapt to changes.

Mitigation:

Flexible contracts, dynamic transportation options, or leasing can offer cost

agility. Additionally, cost-sensitive network designs can help address fluctuating
prices by using adaptable resources and services.
4.REGULATORY AND ENVIRONMENTAL UNCERTAINTY:

Impact:

Sudden changes in regulations, trade policies, or environmental laws can disrupt how
goods are moved, stored, or produced.

Mitigation:

Designing networks that comply with international standards and diversifying

geographically can help companies avoid being affected by regional regulatory
changes.
5.TECHNOLOGY AND INNOVATION UNCERTAINTY:

Impact:

Rapid technological advancements or failures can change the landscape of network

operations, necessitating redesign or upgrades.

Mitigation:

Incorporating modularity into the design allows networks to integrate new

technologies more seamlessly. Future-proofing strategies, like anticipating
automation, can provide long-term value.
6.GEOPOLITICAL AND ECONOMIC UNCERTAINTY:

Impact:

Political instability, trade tensions, or currency fluctuations can affect cross-border

networks and global supply chains.

Mitigation:

Creating multiple routing options or nearshoring facilities closer to customers can

minimize geopolitical risks. Hedging against currency risk can also reduce financial
impacts.
7.NATURAL DISASTERS AND CLIMATE CHANGE:

Impact:

Extreme weather events or long-term climate changes can disrupt logistics, damage
infrastructure, and cause delays.

Mitigation:

Designing resilient and redundant networks with backup routes, distributed

warehousing, and climate-proof infrastructure is essential.
8.MARKET UNCERTAINTY:

Impact:

Changing market conditions, customer preferences, and new competitors can

affect the relevance of a network configuration.

Mitigation:

Agile design allows rapid adaptation to new market trends by enabling fast
switching between suppliers, distributors, or transportation modes.
NETWORK DESIGN DECISIONS USING DECISION TREES

INTRODUCTION TO NETWORK DESIGN

Network design involves planning a computer network to meet business or

technical goals, considering various factors such as scalability, reliability, cost,
and performance.

Key Decisions:

 Topology (structure) of the network

 Selection of hardware (routers, switches, etc.)

 Routing protocols and security measures

WHAT IS A DECISION TREE?

Definition: A decision tree is a tree-like structure that helps in decision-making by

representing choices and their possible consequences, including chance event
outcomes, resource costs, and utility.

Components:

 Nodes: Represent decisions or events.

 Branches: Represent outcomes or next steps.

 Leaves: Final outcomes of the decision process.

DECISION TREES IN NETWORK DESIGN

Simplifies complex decisions by breaking them down into smaller, more

manageable parts.

Helps evaluate multiple factors simultaneously, like performance vs. cost.

Assists in optimizing design by visually comparing different decision paths.

DECISION 1 : Choosing a Network Topology

Factors to Consider:

 Network size, physical layout, performance requirements, cost

Topology Options:

 Star, mesh, bus, ring, hybrid

Decision Tree Application: Helps identify the most suitable topology by asking key
questions such as:

 Is scalability a priority? (Yes → Mesh, Hybrid)

 Is cost a major concern? (Yes → Bus, Star)

DECISION 2: Bandwidth Allocation

Factors to Consider:

• Application requirements, user demand, future growth

Options:

• Allocate more bandwidth to critical applications, shared bandwidth across all users,
or tiered allocation

Decision Tree Application: Determines bandwidth allocation based on:

• Number of users

• Type of traffic (e.g., VoIP, video, data)

• Cost constraints
DECISION 3: Selecting Routing Protocols

Factors to Consider:

• Network size, topology, speed of convergence, security

Protocol Options:

• OSPF, EIGRP, BGP, RIP

Decision Tree Application:

• Large network? (Yes → OSPF, BGP)

• High security needs? (Yes → BGP)

• Simple network with static routes? (Yes → RIP)

DECISION 4: Implementing Network Security

Factors to Consider:

• Sensitivity of data, regulatory compliance, user access

Options:

• Firewalls, VPNs, encryption protocols, intrusion detection systems

Decision Tree Application:

• Do you need remote access? (Yes → VPN)

• Sensitive data being transmitted? (Yes → Encryption protocols)

• Regulatory compliance? (Yes → Specific encryption/security measures)

Benefits of Using Decision Trees in Network Design

 Structured Decision-Making:

Breaks down complex problems into simpler, step-by-step solutions.

 Visualization:

Provides a clear picture of the decision-making process and possible outcomes.

 Comparison:

Helps compare different network design approaches based on key parameters like
cost, performance, and scalability.
Challenges of Decision Trees in Network Design

 Scalability Issues:

Decision trees can become large and complex when many variables are involved.

 Pruning for Simplicity:

Large trees need pruning to avoid overfitting or overcomplication in design choices.

 Handling Continuous Data:

Traditional decision trees work best with discrete data; special techniques may be
needed for continuous variables like bandwidth.
DISTRIBUTION CENTER LOCATION MODELS:

Distribution center location models help determine the best location for
distribution centers to meet customer needs while minimizing costs.

When choosing a distribution center location, it's important to consider the

following:

Distance

The distribution center should be roughly the same distance from all the retail
locations it serves.

Access

The distribution center should have easy access to roads, rails, ports, and airports
to receive products and distribute them to customers.
1. CENTER OF GRAVITY MODEL:

Purpose: To minimize transportation costs by identifying a central location that

minimizes the distance to all customers.

Use Case: Common for single-facility location decisions where transportation cost
is a key factor.

2. FACILITY LOCATION PROBLEM (FLP):

Purpose: To select one or more facility locations to minimize the total cost of
transportation and facility operation.

Use Case: Suitable for both single and multiple facilities. There are different
variations like capacitated and uncapacitated FLP.
3. NETWORK OPTIMIZATION MODELS:

Purpose: To optimize the supply chain network, considering multiple distribution

centers, suppliers, and customer locations.

Use Case: For complex distribution networks with multiple echelons (suppliers,
manufacturers, DCs, and customers).

4. K- MEANS CLUSTERING :

Purpose: To group customer demand points into clusters and place a distribution
center in the center of each cluster.

Use Case: Effective for large-scale location problems where demand is scattered
across a wide geographic area.
5. P- MEDIAN PROBLEM :

Purpose: To minimize the total distance or travel time between customers and
the nearest distribution center.

Use Case: When the number of distribution centers is fixed, and the objective is
to minimize distance or cost.

6. HUB-AND-SPOKE MODEL:

Purpose: To centralize distribution at hubs, reducing the number of routes and

optimizing transportation.

Use Case: Suitable for companies with large volumes of shipments spread
over wide geographic areas.
7. HEURISTIC METHODS:

Purpose: To find good enough solutions for very complex or large-scale

location problems where exact optimization methods may be computationally
expensive.

Use Case: Used in cases where the location problem is too complex for
traditional optimization methods.

8. GRAVITY AND FLOW MODELS :

Purpose: To optimize the flow of goods through a network by selecting

facility locations that balance supply and demand.

Use Case: Useful for balancing inventory and transportation across multiple
distribution centers.
SUPPLY CHAIN NETWORK OPTIMIZATION MODELS: