SRCSourcing & procurementStop 4 of 9
Single, dual or multi-sourcing
How many suppliers to use for an item, and how to split the volume once capacity, lead time and risk are counted.
Read atChange any time
Single sourcing gives a supplier all your volume. You get better prices, simpler coordination and a closer relationship, but if that supplier has a fire, a strike or a quality problem, you have nothing.
Dual or multi-sourcing spreads volume across suppliers. You get a backup that’s already qualified and some competition, but smaller volumes per supplier, more management work, and sometimes slight differences in quality between sources.
Once you use several suppliers, you have to decide how to split the volume, and the cheapest supplier doesn’t simply get everything. Suppose you need 50,000 units within 10 days:
| Supplier | Price | Capacity | Lead time |
|---|---|---|---|
| A | ₹100 | 30,000 | 7 days |
| B | ₹104 | 40,000 | 5 days |
| C | ₹98 | 20,000 | 15 days |
C is cheapest but can’t deliver in time, so it’s out for this order. A is next cheapest but can only make 30,000. So the best split is A 30,000 and B 20,000, at a total cost of ₹50.8 lakh. With hundreds of items, suppliers and constraints, this becomes a linear program: minimize total cost, subject to meeting demand, capacities, lead times and minimum order quantities, and often rules such as “no supplier gets more than 70%”.
Companies frequently choose dual sourcing for high-spend or critical items, with an uneven split such as 70/30, so the second supplier stays qualified and ready to scale up if needed.
Words to use
- Dual sourcing
- Buying the same item from two qualified suppliers, often in an uneven split.
- Quota arrangement
- A rule that splits volume between suppliers by percentage.
- Volume allocation
- Deciding how much each supplier gets, given cost and constraints.
The allocation model
- Decision: , the quantity ordered from supplier .
- Objective: minimize , the purchase cost (or landed cost, if freight differs).
- Constraints: total quantity meets demand; each stays within capacity; suppliers that can’t meet the deadline get zero; optional rules such as a maximum share per supplier or minimum order quantities.
Minimum order quantities and quantity discounts need yes-or-no variables, which makes it a mixed-integer program, still easy for a solver at this size.
Suppliers: A at ₹100 (30,000 capacity, 7 days), B at ₹104 (40,000, 5 days), C at ₹98 (20,000, 15 days).
- Within 10 days: C is excluded. A 30,000 + B 20,000 = ₹50.8 lakh.
- Within 20 days: C is allowed. C 20,000 + A 30,000 = ₹49.6 lakh, ₹1.2 lakh cheaper.
The deadline is worth ₹1.2 lakh here; if the business can wait ten more days, it saves that much.
Putting a price on single sourcing
Annual volume 1 lakh units. Supplier X charges ₹100; supplier Y ₹103. Assume a 5% yearly chance that X is disrupted, which would cost ₹2 crore in lost production if there were no backup.
- Single source (X): purchases ₹1 crore, expected disruption cost 5% × ₹2 crore = ₹10 lakh. Total ₹1.10 crore.
- Dual source (70% X, 30% Y): purchases ₹70 lakh + ₹30.9 lakh = ₹1.009 crore. Y can ramp up if X fails, so the loss falls to, say, ₹50 lakh: an expected ₹2.5 lakh. Total about ₹1.034 crore.
Dual sourcing costs ₹90,000 more in purchases but saves about ₹6.6 lakh in expected disruption cost.
The numbers depend on judgement about probabilities and impacts, but writing them down turns “is a second source worth it?” into a discussion of specific assumptions.
Mitigation or contingency?
Tomlin (2006) compared ways of handling supplier disruptions: mitigation decided in advance (holding inventory, sourcing from two suppliers) versus contingency actions taken when the disruption happens (rerouting demand to a backup with spare capacity). The best choice depends on the disruption’s character: frequent, short disruptions favour inventory; rare, long ones favour dual sourcing or contingent backups, because no affordable buffer lasts long enough.
When single sourcing works
Single sourcing works best when the relationship is close and the network can respond. When a fire in 1997 destroyed the only plant making a critical brake valve for Toyota, the Toyota group’s suppliers organized emergency production across dozens of firms and restored supply within days (Nishiguchi and Beaudet, 1998). The protection came from relationships, shared know-how and slack capacity in the network rather than from a second contract.
Parallel sourcing
In parallel sourcing, each part has one supplier, but similar parts are spread across several suppliers. Each supplier gets the scale and commitment of single sourcing, while the buyer keeps competition (performance on one part affects the award of the next) and capable alternatives for emergencies. It is common in automotive.
Keeping a second source real
- Order regularly from the second source, so its process stays proven and its people practised.
- Keep tooling, specifications and quality approvals current.
- Agree in advance how quickly it can scale up and at what price.
- Test it occasionally with larger orders.
A second source qualified years ago and never used is a document, not a backup.
Splits in ERP
ERP systems implement splits with quota arrangements: each source gets a target share, and the system assigns new requisitions to the source furthest below its quota. The quotas should come from the allocation analysis and be reviewed when prices, capacity or risks change.
Interview kit
Practise it out loud
Answer each question yourself before opening the model answer.
Single sourcing gives better prices and simpler coordination but concentrates risk; dual or multi-sourcing adds resilience and competition at the cost of volume leverage and management effort. When allocating across suppliers I’d optimize total cost subject to demand, capacity, lead time and MOQs, plus policy rules like a maximum share per supplier. For critical items, an uneven dual-source split keeps a qualified backup ready.
You need 50,000 units in 10 days. A: ₹100, 30,000 capacity, 7 days. B: ₹104, 40,000, 5 days. C: ₹98, 20,000, 15 days. How do you split it?
C is cheapest but its 15-day lead time misses the date, so it’s excluded. Fill the next cheapest, A, to its 30,000 capacity, and buy the remaining 20,000 from B. Total cost: ₹30 lakh + ₹20.8 lakh = ₹50.8 lakh.
When is single sourcing the right choice?
When the item is low-risk or the supplier relationship is strategic and deep, when volumes are too small to split economically, or when qualification is very costly. Even then, I’d keep a contingency plan: buffer stock, a pre-qualified alternative, or contract clauses on capacity and continuity.
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. Chapter on sourcing decisions.
- Sodhi, M. S. & Tang, C. S. (2012). Managing Supply Chain Risk. Springer.
- Tomlin, B. (2006). On the value of mitigation and contingency strategies for managing supply chain disruption risks. Management Science, 52(5), 639–657.
- Nishiguchi, T., & Beaudet, A. (1998). The Toyota group and the Aisin fire. Sloan Management Review, 40(1), 49–59.