Logo
Beer Game SimulationOperations Management

BullwhipSim

Experience the Bullwhip Effect firsthand. Play as a Retailer in a 4-tier supply chain and discover how small demand changes create massive upstream volatility.

Based on the MIT Beer Distribution Game — Sterman (1989), Lee, Padmanabhan & Whang (1997)

20 Rounds

~30–45 min

Single Player

AI opponents

4 Tiers

Supply chain

Academic

Debrief included

Supply Chain Structure

Information and material flows — orders travel upstream (right), shipments travel downstream (left)

2-week pipeline delay at each link

Customer

Generates demand

Places orders to Retailer

ORDER
SHIP

Retailer

← YOU PLAY THIS

Orders from Wholesaler

ORDER
SHIP

Wholesaler

AI controlled

Orders from Distributor

ORDER
SHIP

Distributor

AI controlled

Orders from Factory

ORDER
SHIP

Factory

AI controlled

Produces to order

Order flow (upstream, with info delay)
Shipment flow (downstream, 2-week pipeline)
Your role (Retailer)

Supply Chain Roles

Each tier only sees orders from the tier immediately downstream — never actual customer demand.

Retailer

Tier 1

YOU

Faces actual customer demand directly. Must balance holding costs against stockout risk. Your ordering decisions are the primary focus of this simulation.

  • Observe weekly customer demand
  • Check current inventory & backlog
  • Review pipeline (units in transit)
  • Decide order quantity to Wholesaler

💡 Your orders are the starting point of the Bullwhip Effect. Small overreactions here cascade dramatically upstream.

12

4+4

Wholesaler

Tier 2

AI

Receives orders from the Retailer. Has no visibility into actual customer demand — only sees Retailer's orders. AI uses a demand-signal-plus-safety-stock heuristic.

  • Receives Retailer orders as its "demand"
  • Cannot see actual customer demand
  • Adds safety stock based on perceived risk
  • Orders from Distributor with amplification

💡 Wholesaler sees Retailer orders, not customer demand. Any amplification you create here doubles upstream.

12

4+4

Distributor

Tier 3

AI

Two tiers removed from customer demand. Receives only Wholesaler orders. The information distortion at this level is typically 3–5× actual demand variance.

  • Receives Wholesaler orders as its "demand"
  • No visibility into Retailer or customer
  • Amplifies orders further upstream
  • Orders from Factory

💡 At this distance from real demand, AI ordering is heavily influenced by backlog panic and safety stock accumulation.

12

4+4

Factory

Tier 4

AI

The most upstream node. Sees only Distributor orders, which by this point may be 5–10× actual customer demand. Produces to order with a production lead time.

  • Receives Distributor orders as its "demand"
  • Must produce (cannot simply reorder)
  • Highest demand variance of all tiers
  • Experiences the worst cost consequences

💡 Factory is the final amplifier. In real supply chains, factory production swings are often catastrophic — idle capacity followed by overtime.

12

4+4

Game Mechanics & Rules

How each round works — read carefully before starting

Holding Cost

₹0.50

per unit per week in inventory

Backlog Cost

₹1.00

per unit per week unfulfilled

  1. 1📦 Receive shipment from pipeline (units ordered 2 weeks ago arrive)
  2. 2👁️ Observe incoming customer demand for this week
  3. 3🚚 Ship what you can from inventory; remainder becomes backlog
  4. 4💰 Calculate holding cost (inventory × ₹0.50) + backlog cost (backlog × ₹1.00)
  5. 5✍️ Decide your order quantity and submit to Wholesaler
  6. 6🤖 AI tiers (Wholesaler, Distributor, Factory) process their turns simultaneously

The 2-Week Pipeline Delay

Orders placed today arrive in 2 weeks. This delay is the root cause of the Bullwhip Effect — you cannot see what is already in transit, so you tend to order more than needed, causing inventory to overshoot when the pipeline arrives.

A≥ 85%Near-optimal ordering — minimal bullwhip
B70–84%Good control — some amplification
C55–69%Moderate bullwhip — room to improve
D40–54%Significant amplification — high costs
F< 40%Severe bullwhip effect observed

The Bullwhip Effect — Academic Primer

Lee, Padmanabhan & Whang (1997) · MIT Sloan Management Review

Core Concept

The Bullwhip Effect describes the phenomenon where small fluctuations in consumer demand cause progressively larger fluctuations in orders placed by each upstream supply chain tier — like the tip of a whip amplifying a small wrist flick into a massive crack.

01

Demand Signal Processing

Each tier uses observed orders (not real demand) to forecast. They add safety stock, creating order inflation at every link.

02

Rationing & Shortage Gaming

When supply is scarce, buyers order more than needed to secure allocation — inflating apparent demand further.

03

Order Batching

Ordering in large periodic batches (e.g. weekly) rather than continuously introduces artificial demand spikes.

04

Price Fluctuations

Promotions and discounts cause forward buying — buyers stock up beyond current needs, distorting the demand signal.

Var(Orders) / Var(Demand)

A ratio > 1 indicates Bullwhip Effect. Typical real-world values:

Retailer: 1.2–2× · Wholesaler: 2–4× · Distributor: 4–8× · Factory: 8–15×

HUL (Hindustan Unilever): Distributor orders for detergents spike 4–6× during promotional periods vs. actual offtake.
Auto ancillary supply chains: Tier-2 and Tier-3 component suppliers face 8–12× demand swings vs. final vehicle sales.
FMCG modern trade: Seasonal stocking by large-format retailers creates phantom demand spikes upstream.

Start Simulation

No login required — your results are shown at the end of the session

Used to personalise your debrief report. Leave blank to play as "Student".

Controls how aggressively the AI tiers (Wholesaler, Distributor, Factory) respond to demand signals.