269. Bullwhip SD

The quantitative analysis of supply-chain order amplification, from the system-dynamics perspective. Closed-form result by Chen, Drezner, Ryan, Simchi-Levi (2000) for the order-up-to (OUT) policy with AR(1) demand.

269.1. Chen-Drezner-Ryan-Simchi-Levi formula

For demand following an AR(1) process and an OUT policy with smoothing parameter and lead time :

The order variance is always at least the demand variance, and grows quadratically with ratio. For two-echelon chains, this multiplies stage by stage.

269.2. Key insights

269.3. Sources of bullwhip (Lee-Padmanabhan-Whang 1997)

Four causes, each contributing variance amplification:

  1. Demand signal processing: each echelon uses smoothed orders (not actual demand) — like the Chen-DRSL formula
  2. Order batching: large infrequent orders smooth retailer’s demand into spikes upstream
  3. Price fluctuations: forward-buying during promotions creates spikes
  4. Rationing & shortage gaming: customers over-order during shortages, creating ghost demand

Each cause is amplified by long lead times.

269.4. SD beer-game replication

The Sterman beer-game model implements this dynamically:

This is the quantitative validation of the SD bullwhip explanation.

269.5. Mitigations

What reduces bullwhip:

269.6. See also