Bullwhip Effect

Gerard P. Cachon (Fred R. Sullivan professor of operations and information management-The Wharton School) defines bullwhip effect as –

 "The phenomenon of increasing demand variability in the supply chain from down stream echelons (retail) to upstream echelons (manufacturing)."

 Bullwhip effect is more common in forecast driven supply chain and suggests that the variation in demand increases up the supply chain from consumer to supplier. Small changes in customer demand lead to large swings in orders placed upstream.

On January 27, 2010, Wall Street Journal published an article on bullwhip effect and defined-

 “This phenomenon occurs when companies significantly cut or add inventories. Economists call it a bullwhip effect because even small increase in demand can cause a big snap in the need for parts and materials further down the supply chain.”

 
The more echelons exist between the customer and the supplier, the bigger the variation is:

In 1997, Hau L. Lee, V. Padmanabhan and Seungjin Whang published a research paper- "The bullwhip effect in Supply Chain" and explained-

 Tremendous variability in orders along the supply chain can plague companies trying to eliminate excess inventory, forecast product demand, and simply make their supply chain more efficient. What causes the bullwhip effect that distorts information as it is transmitted up the chain? The authors identify four major causes:

1. Demand forecast updating. As each entity along the chain places an order, it replenishes stock and includes some safety stock. With long lead times, there may be weeks of safety stocks, which make the fluctuation in demand more significant.

2. Order batching. Companies may place orders in batches, often to avoid the cost of processing orders more frequently or the high transportation costs for less-than-truckload orders. Suppliers, in turn, face erratic streams of orders, and the bullwhip effect occurs. When order cycles overlap, the effect is even more pronounced.

3. Price fluctuation. Special promotions and price discounts result in customers buying in large quantities and stocking up. When prices return to normal, customers stop buying. As a result, their buying pattern does not reflect their consumption pattern.

4. Rationing and shortage gaming. If product demand exceeds supply, a manufacturer may ration its products. Customers, in turn, may exaggerate their orders to counteract the rationing. Eventually, orders will disappear and cancellations pour in, making it impossible for the manufacturer to determine the real demand for its product.

The authors suggest several ways in which companies can counteract the bullwhip effect:

1. Avoid multiple demand forecast updates. Companies can make demand data from downstream available upstream. Or they can bypass the downstream site by selling directly to the consumer. Also, they can improve operational efficiency to reduce highly variable demand and long resupply lead times.

2. Break order batches. Companies can use electronic data interchange to reduce the cost of placing orders and place orders more frequently. And they can ship assortments of products in a truckload to counter high transportation costs or use third-party logistics companies to handle shipping.

3. Stabilize prices. Manufacturers can reduce the frequency and level of wholesale price discounting to prevent customers from stockpiling. They can also use activity-based costing systems so they can recognize when companies are buying in bulk.

4. Eliminate gaming in shortage situations. In shortages, suppliers can allocate product based on past sales records, rather than on orders, so customers don’t exaggerate their orders. They can also eliminate their generous return policies, so retailers are less likely to cancel orders.

Only by thoroughly understanding the underlying causes of the bullwhip effect, say the authors, can companies counteract and control it.


Causes of bullwhip effect:

Primary Cause: 

·         Lead time of information and material 

Secondary Causes:

Demand Forecast Updating 

·         Forecasting is based on order history from company’s immediate customer

·         Often, forecasts are made at whims & fancies of managers

·         Upstream manager updates his/her demand forecasts based on customer demand variations, longer lead times, price fluctuations etc.

·         Techniques like exponential smoothing creates bigger swings at the supplier’s end

Order Batching 

Periodic Ordering:

·         Weekly, fortnightly, monthly etc.

·         Creates spikes in order sizes, disrupting supplier’s demand forecasts

·         Benefits from transportation & distribution side

 Push Ordering: 

·         Orders are pushed by sales personnel

·         Done usually at monthly/quarterly sales review and demand estimates from sales team

·         This results in uneven spread of customer orders resulting in the bullwhip effect

 Price Fluctuation 

·         Forward buys; discount sales; offer merchandise; coupons; rebates; end of season sales

·         Customers buy in bulk

·         But the buying pattern never matches the consumption pattern

·         This results in overstocking at the far ends of the supply chain and also results in idle capacity 

Rationing and Shortage Gaming

·         When demand exceeds supply, manufacture ration supplies to distributors

·         This results in distributors ordering more than they need, to fulfill the demand

·         When the market cools down, orders start getting cancelled; excess inventory piles up, leading to the bullwhip effect

·         Real demand is never known in such market conditions.

·         Most commonly affected is the IT hardware & telecom industry 

(Reference: http://sinaslogisticsblog.blogspot.com)

Behavioral Causes 

·         Misuse of base-stock policies

·         Misperceptions of feedback and time delays

·         Panic ordering reactions after unmet demand

·         Perceived risk of other players' bounded rationality

Operational Causes 

·         Dependent demand processing

• Forecast Errors
• Adjustment of inventory control parameters with each demand observation

·         Lead time variability (forecast error during replenishment lead time) 

·         Lot-sizing/ order synchronization

• Consolidation of demands
• Transaction motive
• Quantity discount

·         Trade promotion and forward buying

·         Anticipation of shortages

• Allocation rule of suppliers
• Shortage gaming
• Lean and JIT style management of inventories and a chase production strategy

Counter measures

Investment in information technology, creation of a corporate culture of flexibility, focus on customer demand, trustful collaboration and information sharing are the keys and in order to achieve the objective the following methods are proven to be effective:

·         Vendor Managed Inventory (VMI)

·         Just in Time replenishment (JIT)

·         Demand Driven Material Requirements Planning (DDMRP)

·         Strategic partnership

·         Information sharing

·         Smooth the flow of products

·         Coordinate with retailers to spread deliveries evenly

• Reduce minimum batch sizes
• Smaller and more frequent replenishment

·         Eliminate pathological incentives

• Every day low price policy
• Restrict returns and order cancellations
• Order allocation based on past sales instead of current size in case of shortage

(Reference: http://en.wikipedia.org/wiki/Bullwhip_effect)

Sina's blog explains –

First of all long lead times should be reduced if possible.

Demand Forecast Updating

One way to avoid inaccuracies is by reducing the lack of demand visibility by providing access to point of sale (POS) data. Also it is possible to overcome exaggerated demand forecasts by single control of replenishment or vendor Management Inventory (VMI).

Order Batching 

·         One countermeasure would be Electronic Data Interchange (EDI) and computer aided ordering (CAO).

·         Random or correlated ordering is countered with regular delivery appointments

·         More frequent ordering results in smaller orders and smaller variance

 Price Fluctuation

High-low pricing can be replaced with everyday low prices. This way, special purchase contracts can be implemented in order to specify ordering at regular intervals to better synchronize delivery and purchase.

Rationing and Shortage Gaming 

One solution is to allocate units based on past sales. Unrestricted ordering capability can be addressed by reducing the order size flexibility and implementing capacity reservations. For example, one can reserve a fixed quantity for a given year and specify the quantity of each order shortly before it is needed, as long as the sum of the order quantities equals to the reserved quantity.