From Supply Chain Management Encyclopedia
Russian: Эффект хлыста
The bullwhip effect (also known as demand amplification, whip-saw, whiplash effect, or Forrester effect) refers to the phenomenon of demand variability amplification as moving up in the supply chain: from the point of actual (final) demand to the point of origin. It means that variability at the "end" of supply chain (closer to consumption, e. g. retailer) is much less, than at the other "end", where it begins (far from consumer, e. g. producer or supplier). Moving up in the supply chain from consumer to supplier increases demand variability. The more actors exist in a particular supply chain and the greater is lead time, the greater demand variability would be. The first record of this phenomena belongs, probably, to J. Forrester (that is why it is possible to find in the literature references to "Forrester effect", although J. Forrester had never named it neither Forrester, nor bullwhip effect). The term "Forrester effect" used to denote what is now called demand signal processing, as Forrester was the first to demonstrate this phenomenon through dynamo simulation. In 1997, the phenomenon of bullwhip effect was popularized by Lee et al. The bullwhip effect has quite negative impact on supply chain efficiency. It leads to excessive safety stock, higher production costs and overheads, potential final product quality distortions, poor customer service, lost sales, higher logistics costs, and so on.
Causes and Consequences of the Bullwhip Effect
Lee et al.: identified four major causes of the bullwhip effect
- Demand forecast updating. Demand forecasting and production planning are often based on the history of direct customers` orders of the company. These customers, however, rarely make exactly the same orders in a short period of time as they receive from their own direct customers due to various reasons, including sales expectations, risk aversion, personal factors, industry characteristics, etc. As soon as a manager observes a change in downstream orders, he or she readjusts its own plans, however this order might not (and often not) reflect the real demand. As a result, each member of the supply chain makes slightly different order from the order it receives itself and, finally, the supplier may receives an order that is totally different from the real market dynamics. This situation is very common, because if the lead time is more than 0 (and it is 0 only in electronic commerce), it is a wise rule to keep safety stocks. The higher lead time the larger safety stock should be. On the one hand, the safety stocks are the cause of inefficiency in the supply chain, because they require extra operational costs, and on the other hand, they increase the bullwhip effect.
- Order batching. Orders usually are accumulated in batches: in time (daily, weekly, monthly, etc.) or in volume (palette, container, etc). Batching also increase the bullwhip effect. There are various reasons for batching: from order processing costs (how much does the company spend on managing the orders) to transportation issues. Sometimes this effect is referred as the Burbidge effect. Burbidge outlined particular problems that my be caused by this effect unless duly watched.
- Price fluctuation. Manufacturers or retailers often launch various promotion programs (discounts, flexible prices, etc.). As a result, customers observe different prices and may artificially boost or cut their buying to get benefits from temporary advantages. This leads to additional order fluctuations. For more information see everyday low pricing
- Rationing and shortage gaming. If the producer is not able to fulfill the excessive demand in a short period of time, and the retailer (wholesaler or distributor) knows about this, they will act to increase their orders to get at least something. For example, if the retailer really needs 100 pieces of a product and he knows that the producer will fulfill only about 50% of the order, he will order 200. However, "200 pieces" is very often a real practice of suppliers who make their strategic decisions basing on this information, however in the next period there might be only 100 (real) pieces ordered by retailers. Behavioral psychology often resorts to the term "bounded rationality" implying a sub-optimal but borderline rational decision making by actors.. Rationing and gaming are sometimes referred to as the Houlihan effect after John Houlihan. This effect suggests that missed deliveries lead to higher safety stock levels and thus inflated orders. As more orders are made, the chain becomes more vulnerable to unreliable sources as reliable ones lack capacity to increase production instantly. All this leads to the bullwhip effect going up the supply chain with increasing magnitude. Houlihan described this process as a "flywheel effect". Olsmats et al. (1988) demonstrated this phenomenon in action in the automotive sector. Price variation describes offering goods and services to consumers at lower prices through various promotions in order to boost immediate demand.
Some of researchers try to find origins of the bullwhip effect in the psychology of decision makers. Using modeling of bullwhip effect with the Beer Game, they prove that the manager uses one of two basic strategies: ‘safe harbor’ or ‘panic’, both having negative impact on the supply chain efficiency. However, as soon as there is a chance for negotiation, the results of simulation become much better .
Particularly negative impacts of the bullwhip effect for the supply chain are:
- Inefficient inventory management. The varying demand leads to variation in inventory levels at each tier of the supply chain. As supplier receives order, which is higher than the order on previous period, the company has to increase inventory level. On the other hand, if the order level is lower, it is not always possible to decrease inventory level in short period of time. The higher variability in demand (in orders), the higher safety stocks should be. Safety stock have trend to increase, as moving away from point of consumption.
- Backlogged orders and poor service to product outlets. The safety stock that is required to ensure a sufficient service level increases with the variation in the demand, however, it is not always enough to fulfill excessive demand (orders). Hence, sometimes companies might face absence of goods on the shelves of the retailer.
- Unpredictable production schedules. A variation in demand causes variation in capacity usage. During “high” period producer usually has to increase the number of shifts. During “low” period – to make extra safety stocks or leave workers without any work (both cases lead to financial losses).
- High prices for raw materials because of immediate need. In case of emergent need of producing the order, producer often face a situation of absence of raw materials (of some of raw materials). Ordering even small part of raw materials from supplier on emergence will cost to producer enormously high price (at least for unscheduled transportation) .
- Lost revenues. All these leads to financial losses: extra safety stocks (means more capital employed) or missed orders (missed sales).
Analyses of recent papers shows that researchers do not argue about the causes and consequences of bullwhip effect, but try to find remedies for negative impact on the supply chain performance.
Example of the Bullwhip Effect
Usually consumption of most FMCG goods is stable. For instance, consumption of diapers by babies – is constant; consumption of bread, salt, ketchup and other food – constant, etc. Retailers very often see smooth demand with minor fluctuations as shown on the figure below.
However, while making its own orders the retailer takes into account his own stock levels (from previous periods), sales expectations (including expectations of his own advertising and promotion), discounts from the manufacturer or the distributor, the price of transportation, order processing and other minor factors. Therefore, orders do not look that smooth any more.
Orders from the wholesaler to the distributor are even more volatile due to the same reasons.
At the end of the supply chain, orders to the manufacturer are even more variable. The manufacturer now has to solve problems of extra labor force or extra safety stock to fulfill all the orders. Extra costs and order failures are very common in this situation.
Analysis of the Bullwhip Effect
The bullwhip effect was analyzed by various researchers with different methods:
- Simulation approach
- Evolutionary least-mean-square algorithm
- Beer game simulation with different demand scenarios
- Multi-echelon supply chain system
- Analytical approach
Bullwhip Effect Simulation (Beer Game)
Bullwhip Effect Simulation Game (Beer Game, also known as beer distribution game) was developed by the Systems Dynamics Group at the Massachusetts Institute of Technology in the 1960s. It demonstrates the bullwhip effect by simulating a supply chain with four tiers: the retailer, the wholesaler, the distributor or the factory. It might be played in class or on-line and is a very effective mean of illustrating systems thinking. By enabling managers to experience the negative impact of the bullwhip effect on supply chain performance, the beer game makes them aware of the application of countermeasures in their companies. Each player takes the role (individually or in group of 2-3 players) one of the roles. An ultimate customer places orders at the retailer (buys beer). His demand is defined, but unknown to the participants. The ultimate demand is four units (bottles, packs of beer) during the first six periods (including “test” or “zero” period) and eight units during the following periods of the simulation. The game usually lasts for 50-70 period. It is enough to diagnose bullwhip effect. Each period represents one week. During this period participants have to make important decisions and activities in strict order:
- Each player (team) receives order from their customer. For retailer it is pre-defined order (demand). For the rest of players it is orders from previous players (eg order from wholesaler for distributor).
- Each player (team) makes a decision of how much to order. This decision is based on the received orders, on backlogged orders (all orders should be accomplished), on the previous orders, on the inventory left in stock and other factors.
- Each player (team) has to minimize its costs. A product on stock (safety stock) costs $0.50 per product per period. Backlogged orders costs $1.00 per product per period (penalty for out-of-stock situations). Thus participants have to take into account a trade-off between minimizing the costs of capital employed in stocks on the one hand and avoiding of out-of-stock situations, on the other hand.
Information flow (the information of how much to order) moves along supply chain with a delay of one week. It represents common situation in real companies. Good flow has a delay of two weeks due to transportation. Producer gets its orders from production after two weeks as well (to make it easier it is possible to say that one week is for production and one week is for quality control and packaging). Some important rules to remember:
- Do not try to look for your demand before it is time to.
- Do not change the sequence of steps.
- Do not mix the orders and finished products.
- It is possible to make an empty order.
- If you missed the round, don’t try to catch-up. Make sure that all other members did it correctly.
Remedies for the Bullwhip Effect
Lee et al. (1997) proposed a framework for supply chain initiatives to deal with the bullwhip effect: information sharing, channel alignment, operational efficiency. It was criticized for general approach and since then a lot of papers on this topic, trying to find more general or more specific solutions:
- Ordering policy
- Lot sizing rules
- Forecasting improvements
- Decreasing demand variability
- A multi-agent approach. Information sharing is one of the most important tools for minimizing the bullwhip effect. Most of contemporary tools and approaches, including VMI, CPFR, etc. and technical innovations, such as RFID use this principle. The importance of information in supply chains:
- Helps reduce variability in supply chains
- Help suppliers make better forecast
- Enables the coordination system of manufacturing and distribution systems and strategies
- Enables retailers to serve their customers better
- Enables retailers to react and adapt to supply chain problems more rapidly
- Enables lead time reductions
- ↑ 1.0 1.1 Lee H.L., Padmanabhan V. and Whang S. (1997) Information distortion in a supply chain: The bullwhip effect, Management Science; Apr 1997; 43, 4; pg. 546
- ↑ Forrester J.W., (1961) Industrial dynamics. New York: MIT Press and John Wiley & Sons.
- ↑ Lee H.L., Padmanabhan V. and Whang S. (1997) The bullwhip effect in supply chains. Sloan Management Review 38(3) p93–102
- ↑ Burbidge J.L. (1991) Period Batch Control (PBC) with GT – the Way Forward from MRP, PBCIS Annual Conference, Birmingham, UK
- ↑ Sterman J.D. (1989) Modeling managerial behavior: misperceptions of feedback in a dynamic decision making experiments. Management Science, 35 (3), p321–339
- ↑ Houlihan J. B. (1988) International supply chains: a new approach. Management Decisions. Vol. 26. p13-19.
- ↑ Olsmats C. M., Edghill J. S. and Towill D. R. (1988) Industrial dynamics model building of a close-coupled production-distribution system. Engineering Costs & Production Economics, Vol. 13 Issue 4, p295-310, 16p
- ↑ Nienhaus J., Ziegenbein A. and Schoensleben P. (2006) How human behavior amplifies the bullwhip effect. A study based on the beer distribution game online Production Planning & Control, Vol. 17, No. 6, 547–557
- ↑ Buzzell R. D., J. A. Quelch and W. J. Salmon (1990) The costly bargain of trade promotion. Harvard Business Review, 68, p141–148
- ↑ Richard M. (1997) Quantifying the bullwhip effect in supply chains. Journal of Operations Management, Vol. 15 Issue 2, p89-100
- ↑ Kelly, K. 1995. Burned by busy signals: Why Motorola ramped up production way past demand. Business Week 6 36
- ↑ Holmstrom, J. 1997. Product range management: a case study of supply chain operations in the European grocery industry. Supply Chain Management 2(3) 107–115
- ↑ Dooley K., Yan T., Mohan S., Gopalakrishnan M. (2010) Inventory management and the bullwhip effect during the 2007–2009 recession: evidence from the manufacturing sector. Journal of Supply Chain Management, Vol. 46 Issue 1, p12-18
- ↑ Wangphanich P., Kara S. and Kayis B. (2010) Analysis of the bullwhip effect in multi-product, multi-stage supply chain systems-a simulation approach, International Journal of Production Research; Aug2010, Vol. 48 Issue 15, p4501-4517
- ↑ Tseng L-T., Tseng L-F., Chen H-C. (2011) Exploration of the bullwhip effect based on the evolutionary least-mean-square algorithm, International Journal of Electronic Business Management, Vol. 9 Issue 2, p160-168
- ↑ Matteo C., Chiara R., Tommaso R. and Fernanda S. (2010) Bullwhip effect and inventory oscillations analysis using the beer game model, International Journal of Production Research, Vol. 48 Issue 13, p3943-3956
- ↑ Xiao-Yuan, H. (2007) An H∞ control method of the bullwhip effect for a class of supply chain system. International Journal of Production Research, Vol. 45 Issue 1, p207-226
- ↑ Nienhaus J., Ziegenbein A. and Schoensleben P. (2006) How human behavior amplifies the bullwhip effect. A study based on the beer distribution game online, Production Planning & Control, Vol. 17, No. 6, p.547–557
- ↑ Disney S.M. and Towill D.R., (2003) On the bullwhip and inventory variance produced by an ordering policy. Omega, 31 (3), 157–167
- ↑ Kelle P. and Milne A. (1999) The effect of (s,S) ordering policy on the supply chain. International Journal of Production Economics, 59 (1–3), 113–122
- ↑ Pujawan I.N. (2004) The effect of lot sizing rules on order variability. European Journal of Operations Research, 159 (3), 617–635
- ↑ Zhang X. (2005) Delayed demand information and the dampened bullwhip effect. Operations Research Letters, 33 (3), 289–294
- ↑ Zhao X. and Xie J. (2002) Forecasting errors and the value of information sharing in a supply chain. International Journal of Production Research, 40 (2), 311–335
- ↑ Croson R. and Donohue K. (2005) Upstream versus downstream information and its impact on the bullwhip effect. System Dynamics Review, 21 (3), 249–260
- ↑ Ingalls R.G., Foote B.L. and Krishnamoorthy A. (2005) Reducing the bullwhip effect in supply chains with control-based forecasting. International Journal of Simulation & Process Modelling, 1–2 (1), 90–110
- ↑ Lin C. and Lin Y. (2006) Issues in the reduction of demand variance in the supply chain. International Journal of Production Research, 44 (9), 1821–1843
- ↑ Qing Cao and Leggio K. (2008) Alleviating the bullwhip effect in supply chain management using the multi-agent approach: an empirical study. International Journal of Computer Applications in Technology, Vol. 31 Issue 3/4, p225-237