SRCSourcing & procurementStop 5 of 9
Supplier network design
Which supplier sites should feed which plants, weighing landed cost, lead time and risk, and where local sourcing beats global.
Read atChange any time
Supplier network design applies network thinking to the supply side: which supplier sites should feed which of your plants, and how much should be sourced locally versus globally?
Take a two-wheeler maker with plants in Pune and Chennai, buying brake assemblies, wiring harnesses and plastic parts. For each part it weighs:
- Landed cost: price plus freight, duties, packaging and inventory.
- Lead time and reliability: a supplier 50 km away can deliver daily in small lots; one overseas needs weeks on the water and a big buffer of stock.
- Capacity and capability: can the supplier grow with you, and does it have the technology?
- Risk: geography, currency, tariffs, single-site dependence, and financial health.
Bulky, fast-moving or frequently changing parts usually favour local suppliers near the plant, sometimes in a supplier park right next to it. High-tech, low-weight parts with big economies of scale are often bought globally from a few specialist plants.
The network also has depth. Your direct (tier 1) suppliers depend on their own suppliers (tier 2, tier 3). Mapping those tiers often reveals that several “independent” tier 1 suppliers all rely on the same tier 2 plant, which is a hidden single point of failure.
Like warehouse network design, the decision can be modelled as an optimization: choose supplier sites and flows to minimize landed cost, subject to capacity, lead-time and risk constraints, then stress-test it against disruptions and tariff changes.
Words to use
- Tier 1 / tier 2
- Your direct suppliers, and their suppliers.
- Nearshoring
- Moving supply closer to the point of use to cut lead time and risk.
- Supplier park
- A cluster of suppliers located next to a plant for frequent, small deliveries.
Landed cost by pairing
For each supplier and each plant it could serve, add the price, freight, duties and packaging, and the inventory cost of the lead time.
Annual demand: Pune plant 60,000 harnesses, Chennai plant 40,000.
| Supplier | Price | Capacity | Landed at Pune | Landed at Chennai |
|---|---|---|---|---|
| S1, near Pune | ₹1,000 | 70,000 | ₹1,010 | ₹1,060 |
| S2, near Chennai | ₹1,020 | 50,000 | ₹1,080 | ₹1,030 |
| S3, overseas | ₹900 | 100,000 | ₹1,040 | ₹1,040 |
S3’s landed cost includes ₹120 of freight and duty and ₹20 of extra inventory for its long lead time.
Cheapest assignment: S1 supplies Pune (60,000 at ₹1,010) and S2 supplies Chennai (40,000 at ₹1,030), a total of ₹10.18 crore a year. S3 has the lowest quote and supplies nothing.
S3 is still useful: if S2 were disrupted, S3 could supply Chennai for only ₹10 a unit more, which makes it a sensible qualified backup.
Local suppliers and inbound flow
Close suppliers enable practices distant ones can’t:
- Frequent small deliveries (several a day), keeping line-side stock low.
- Just-in-sequence delivery, where parts arrive in the exact order of the vehicles on the line (seats, dashboards, wiring).
- Milk runs: one truck collects from several nearby suppliers on a fixed loop, giving frequent deliveries without many part-empty trucks.
- Supplier parks: key suppliers located next to the assembly plant.
Modelling it
At scale this is an optimization like distribution network design, applied upstream: choose supplier sites and flows to minimize landed cost, within capacities, lead-time limits and rules such as a minimum share for a second source, and then test it against scenarios like a tariff change or a supplier outage.
Concentration in the lower tiers
A network that looks diverse at tier 1 can depend on a handful of tier 2 or tier 3 sources: one resin producer, one chip foundry, one specialty chemical plant. Network design that stops at tier 1 misses the true single points of failure. For critical categories, map deeper, and treat a shared sub-supplier as a risk to model, not a detail.
Regionalization and local-content rules
Several forces push supply networks towards regions:
- Trade policy: tariffs and export controls change landed costs and access.
- Local-content incentives: schemes such as India’s production-linked incentives and phased manufacturing programmes reward local value addition.
- “China plus one”: many companies have added suppliers in other countries, including India, to reduce concentration.
- Speed: shorter supply lines support faster product changes and lower inventory.
Regional networks usually cost more per unit than the single cheapest global source, so the business case rests on lower risk, inventory and lead time, and sometimes on access to incentives.
Design decisions shape the supply base
Engineering choices decide how many suppliers a company needs and how replaceable they are. Common parts across products and plants concentrate volume and simplify sourcing; standard specifications allow more suppliers to compete; proprietary components lock in one source. Involving procurement early in design avoids accidental single sources.
Resilience as a design goal
Resilient supply networks are designed, not hoped for:
- qualified alternatives for critical parts, ideally in different regions,
- flexibility to move volume between suppliers and plants,
- buffer stocks placed where recovery would be slowest,
- visibility of supplier capacity and status.
Each costs money in normal times, so the design should be explicit about which risks it protects against and at what price.
Interview kit
Practise it out loud
Answer each question yourself before opening the model answer.
Supplier network design decides which supplier sites serve which plants and how much to source locally versus globally. I’d compare landed cost, lead time and reliability, capacity and capability, and risk such as tariffs, currency and single-site dependence. Bulky or fast-changing parts often favour local suppliers, while high-tech parts with scale economies are sourced globally. I’d also map tier 2 and tier 3 to find shared dependencies, and stress-test the design.
Why might a manufacturer pay more to source a part locally?
Shorter lead times mean less inventory and faster reaction to design or demand changes, freight is lower for bulky parts, and the supply is less exposed to ports, currency and tariffs. The total landed and risk-adjusted cost can be lower even when the unit price is higher.
What does mapping tier 2 suppliers usually reveal?
Hidden concentration: several tier 1 suppliers depending on the same sub-supplier, region or raw material. That’s a single point of failure that tier 1 diversification doesn’t fix.
Connected stops
Where this idea feeds in, or gets its inputs from.
Sources and further reading
- Chopra, S. & Meindl, P. Supply Chain Management: Strategy, Planning, and Operation. Pearson. Chapters on sourcing decisions and global network design.
- Choi, T. Y. & Hong, Y. (2002). Unveiling the structure of supply networks: case studies in Honda, Acura, and DaimlerChrysler. Journal of Operations Management, 20(5), 469–493.
- Simchi-Levi, D., Schmidt, W., & Wei, Y. (2014). From superstorms to factory fires: Managing unpredictable supply-chain disruptions. Harvard Business Review, 92(1/2), 96–101.