What is the bullwhip effect
The bullwhip effect is the amplification of demand variability as orders move up the supply chain: a small swing at the consumer becomes a large swing at the manufacturer and its suppliers.
Jay Forrester described the pattern in 1961, and Lee, Padmanabhan and Whang explained its causes in 1997. It occurs even when end-customer demand is stable, and the result is excess inventory, stockouts, overtime and idle capacity along the chain.
Bullwhip effect visual map
The map shows the four echelons, the four causes and the same numerical example used in this article.
The 4 classic causes
Lee, Padmanabhan and Whang identified four operational causes. All of them come from rational decisions made by each echelon without seeing real end demand.
- Demand signal processing: each echelon forecasts from the orders it receives, not from consumer sales. A small increase pushes safety stock and the target inventory up.
- Order batching: to spread freight and ordering cost, the buyer orders in large, infrequent lots. For the supplier, demand looks lumpy.
- Price fluctuation: promotions and volume discounts lead to forward buying. Demand rises during the promotion and drops after it.
- Rationing and shortage gaming: when product is short, customers order more than they need to secure their share, then cancel when the shortage ends.
Long replenishment lead times make all four worse, because each echelon has to cover more weeks of uncertain demand.
Numerical example: 4 echelons, one 10% change
Illustrative example (a simulation, not real data). Consumer demand is 100 units a week, rises to 110 for 6 weeks and returns to 100. Every echelon (retailer, wholesaler, factory) follows the same rule: 3-week moving-average forecast, 2-week replenishment lead time, and an order-up-to level equal to 3 weeks of forecast (lead time + 1). The weekly order is the demand received plus the change in the order-up-to level. Each echelon's order is the next echelon's demand.
Order(t) = Demand(t) + [Order-up-to level(t) − Order-up-to level(t−1)], with Order-up-to level = 3 × 3-week moving average
When demand goes from 100 to 110, the moving average rises by 3.33 in the first week, the order-up-to level rises by 10 and the retailer orders 120. Orders by echelon, by week range:
| Echelon | Wk 1-6 | Wk 7-9 | Wk 10-12 | Wk 13-15 | Wk 16-18 |
|---|---|---|---|---|---|
| Consumer (demand) | 100 | 110 | 110 | 100 | 100 |
| Retailer (order) | 100 | 120 | 110 | 90 | 100 |
| Wholesaler (order) | 100 | 140 | 100 | 70 | 110 |
| Factory (order) | 100 | 180 | 60 | 40 | 150 |
The consumer moves between 100 and 110. The factory receives orders of 180 and then 40, although consumption never changed by more than 10%.
How to measure the bullwhip effect
The usual measure is the ratio between the variance of the orders an echelon places and the variance of the demand it receives. Above 1, there is amplification. Using the 18 weeks of the example:
Variance ratio = Var(orders) ÷ Var(consumer demand)
| Echelon | Variance (18 wks) | Ratio to consumer | Peak / trough vs. 100 |
|---|---|---|---|
| Consumer | 22.2 | — | — |
| Retailer | 88.9 | 4.0 | +20% / −10% |
| Wholesaler | 422.2 | 19.0 | +40% / −30% |
| Factory | 2,325.0 | 104.6 | +80% / −60% |
Factory calculation: 2,325.0 ÷ 22.2 = 104.6. Variance grows at an accelerating rate at each echelon. Measure with real data covering at least 26 weeks, weekly or monthly, never with the annual average.
How to reduce the bullwhip effect
| Cause | Countermeasure | How to apply |
|---|---|---|
| Forecast from orders | Share sales data | Give the supplier real point-of-sale data and work from one forecast, through S&OP |
| Large batches | Cut ordering and freight cost | Smaller, more frequent orders, load consolidation, milk run |
| Price | Stable pricing | Everyday low price instead of one-off promotions; discounts tied to actual sell-through |
| Rationing | Predictable allocation | Allocate scarce product by sales history, not by order size |
| Long lead time | Shorten lead time | Work on lead time and use pull replenishment or VMI (vendor-managed inventory) |
Inside the plant, heijunka and pull production absorb variation before it becomes a supplier order. The base is always a sound demand forecast and a master production schedule that does not react to every single order.
Where it shows up in practice
- Launches and promotions: the order peak becomes capacity and stock left over weeks later.
- Imported or long-lead items: the buyer covers months of demand and amplifies every signal.
- Chains with many intermediaries: each layer adds its own forecast and stock.
- Suppliers to OEMs: the schedule sent to the supplier changes every week, and the supplier needs protective stock. See the stock policy in inventory management.
Common mistakes and limitations
- Blaming the customer. The cause is the decision rule of each echelon, including yours.
- Measuring with annual or aggregated monthly data: the amplification disappears in the average.
- Confusing the bullwhip effect with seasonality. Seasonality exists in end demand; the bullwhip is the extra amplification.
- Cutting safety stock without treating the cause: it creates shortages and more emergency orders.
- Using the numerical example as a forecast. It is a simulation that illustrates the mechanics, with one rule for all echelons.
Frequently asked questions
What is the bullwhip effect?
It is the amplification of demand variability along the supply chain: orders swing more at each echelon, from retailer to supplier.
What causes the bullwhip effect?
Demand signal processing, order batching, price fluctuation and rationing. Long lead times and lack of information make all four worse.
How is the bullwhip effect calculated?
As the ratio of the variance of the orders an echelon places to the variance of end demand. In the example, the factory reaches 104.6 times the consumer variance.
How do you reduce the bullwhip effect?
Share real sales data, reduce batch sizes and lead times, stabilize prices, allocate by history and plan jointly, for example in S&OP.
What is the difference between the bullwhip effect and seasonality?
Seasonality is a pattern in end demand. The bullwhip effect is the amplification of that pattern, or of any variation, in upstream orders.
Who discovered the bullwhip effect?
Jay Forrester observed it in 1961 in industrial dynamics models. Lee, Padmanabhan and Whang named it and explained its causes in 1997.
Sources
- FORRESTER, J. W. Industrial Dynamics. Cambridge, MA: MIT Press, 1961.
- LEE, H. L.; PADMANABHAN, V.; WHANG, S. The Bullwhip Effect in Supply Chains. Sloan Management Review, 1997.
- CHOPRA, S.; MEINDL, P. Supply Chain Management: Strategy, Planning, and Operation. Pearson.
- STERMAN, J. Modeling Managerial Behavior: Misperceptions of Feedback in a Dynamic Decision Making Experiment. Management Science, 1989.
Keep studying: trusted sources and links
- ILOS: Instituto de Logística e Supply Chain (in Portuguese)
- CSCMP: Council of Supply Chain Management Professionals
- Wikipedia: Demand forecasting
- Wikipedia: Supply chain management
Save this address: Manufatura Enxuta is your school for lean manufacturing. Come back whenever you need to study, with a complete article, a visual map and trusted sources. Follow one of the learning paths or check the A–Z glossary.
Want real pull production?
Download the free pull production e-book, with kanban, supermarkets and leveling.
Download the e-book
About the author
Lean Manufacturing & Behavioral Management Specialist
Over 20 years in the automotive and metalworking industries (GM and Dana), Lean Manufacturing practitioner since 2006. SENAI instructor and mentor in Brazil’s Brasil Mais Produtivo program, delivering consulting, training and audits for 50+ companies, combining quality, productivity and people development.
