Allocation when supply is short

Sometimes there simply isn’t enough: a raw material shortage, a plant outage or a demand spike. With 5,000 units available and orders for 9,000, someone gets less than they asked for. Allocation decides who, using an explicit rule rather than whoever shouts loudest:

• First come, first served: simple, but rewards early ordering, not importance.
• Fair share: everyone gets the same percentage, here 5,000 ÷ 9,000, about 56% of each order.
• Priority tiers: key accounts and contractual commitments first, then others.
• Economic: favour orders with the highest margin or the biggest penalty for missing them.
• Optimized: a model balancing margin, penalties, service targets and strategic value.

Customers A, B and C asking for 3,000, 2,000 and 4,000 might, under fair share, get about 1,700, 1,100 and 2,200. Under a priority rule where A has a contract, A might get its full 3,000 and B and C split the rest.

The choice changes behaviour. If customers learn that allocation is proportional to what they order, they inflate their orders to get more, a form of shortage gaming. That sends false demand signals upstream and is one of the classic causes of the bullwhip effect. Allocating on past sales history or on forecasts agreed in advance avoids rewarding inflated orders.

One shortage, four rules

A fourth approach, economic allocation, ranks orders by margin or by the penalty for missing them. It maximizes short-term profit but can damage relationships with customers who are less profitable today and important tomorrow.

Allocation quotas set in advance

Allocation works best as a planned process, not an emergency decision:

1. Forecast the shortage: S&OP shows supply below demand for certain products and weeks.
2. Set quotas per customer, channel or region, based on agreed forecasts, history or contracts.
3. Load the quotas into order promising, so each order consumes its group’s allocation.
4. Release unused quotas at an agreed point before shipment, so stock isn’t stranded.
5. Review weekly as supply and demand change.

Allocation as an optimization

With many products, customers and constraints, allocation can be modelled: maximize margin minus penalties for unmet commitments, subject to supply, minimum service levels for strategic customers, contractual quantities and fairness rules such as no customer below a certain share of its forecast. The model makes trade-offs explicit, and planners can see what each rule costs.

Turn-and-earn and gaming

Cachon and Lariviere (1999) studied allocation based on past sales, known in the car industry as turn-and-earn: dealers that sell more today earn a larger share of scarce models later. It discourages inflated orders and rewards real sales, but it can push retailers to sell harder (with discounts) in slow periods to secure future allocation. Their broader work shows that allocation rules shape the orders customers place: rules that reward order size invite inflation, which feeds the bullwhip effect.

Contract terms that matter

• Allocation clauses: how the supplier will share scarce output among customers.
• Firm commitments: volumes the supplier must deliver regardless, often in return for the buyer’s commitment.
• Force majeure: what counts, and how obligations change when it applies.
• Price during shortages: whether prices can rise and how.

Knowing these terms before a shortage avoids disputes during one.

Communicating a shortage

Customers cope better with a shortage they hear about early and clearly:

• Tell them as soon as the shortage is known, with the expected duration.
• Explain the rule and apply it consistently.
• Give each customer their quantity and dates, and update them when anything changes.
• Offer alternatives: substitutes, other pack sizes, later dates.

Customers who trust the process are less likely to inflate orders, which keeps the demand signal honest.