SLOANSELECT COLLECTION FALL 2009 A SPECIAL COLLECTION OF BUSINESS OPERATIONS INSIGHTS FROM MIT SLOAN MANAGEMENT REVIEW Lessons in Operations Management CONTENTS SLOANSELECT COLLECTION FALL 2009 Lessons in Operations Management 1 The Bullwhip Effect in Supply Chains Spring 1997 11 7 Deadly Sins of Performance Measurement Spring 2007 21 Sharing Global Supply Chain Knowledge Summer 2008 28 Evolving From Value Chain to Value Grid Summer 2006 37 Taking the Measure of Outsourcing Providers Spring 2005 45 Rethinking Procurement in the Era of Globalization Fall 2008 REPRINT NUMBER OPS1109 LESSONS IN OPERATIONS MANAGEMENT • MIT SLOAN MANAGEMENT REVIEW i The Bullwhip Effect in Supply Chains Hau L. Lee • V. Padmanabhan • Seungjin Whang Distorted information from one end over time. However, when they examined the orders from the reseller, they observed much bigger swings. of a supply chain to the other can Also, to their surprise, they discovered that the orders from the printer division to the company’s integrated lead to tremendous inefficiencies: circuit division had even greater fluctuations. excessive inventory investment, poor What happens when a supply chain is plagued with a bullwhip effect that distorts its demand information customer service, lost revenues, as it is transmitted up the chain? In the past, without misguided capacity plans, ineffective being able to see the sales of its products at the distri- bution channel stage, HP had to rely on the sales or- transportation, and missed ders from the resellers to make product forecasts, plan capacity, control inventory, and schedule production. production schedules. How do Big variations in demand were a major problem for exaggerated order swings occur? What HP’s management. The common symptoms of such variations could be excessive inventory, poor product can companies do to mitigate them? forecasts, insufficient or excessive capacities, poor cus- tomer service due to unavailable products or long back- ot long ago, logistics executives at Procter & logs, uncertain production planning (i.e., excessive revi- Gamble (P&G) examined the order pat- sions), and high costs for corrections, such as for expe- Nterns for one of their best-selling products, dited shipments and overtime. HP’s product division Pampers. Its sales at retail stores were fluctuating, but was a victim of order swings that were exaggerated by the variabilities were certainly not excessive. However, the resellers relative to their sales; it, in turn, created as they examined the distributors’ orders, the execu- additional exaggerations of order swings to suppliers. tives were surprised by the degree of variability. When In the past few years, the Efficient Consumer Re- they looked at P&G’s orders of materials to their sup- sponse (ECR) initiative has tried to redefine how the pliers, such as 3M, they discovered that the swings grocery supply chain should work.1 One motivation were even greater. At first glance, the variabilities did for the initiative was the excessive amount of invento- not make sense. While the consumers, in this case, ry in the supply chain. Various industry studies found the babies, consumed diapers at a steady rate, the de- that the total supply chain, from when products leave mand order variabilities in the supply chain were am- the manufacturers’ production lines to when they ar- plified as they moved up the supply chain. P&G rive on the retailers’ shelves, has more than 100 days of called this phenomenon the “bullwhip” effect. (In some industries, it is known as the “whiplash” or the Hau L. Lee is the Kleiner Perkins, Mayfield, Sequoia Capital Professor “whipsaw” effect.) in Industrial Engineering and Engineering Management, and professor When Hewlett-Packard (HP) executives examined of operations management at the Graduate School of Business, Stanford University. V. Padmanabhan is an associate professor of marketing, and the sales of one of its printers at a major reseller, they Seungjin Whang is an associate professor of operations information and found that there were, as expected, some fluctuations technology, also at Stanford. SLOAN MANAGEMENT REVIEW/SPRING 1997 LEE ET AL. 93 LESSONS IN OPERATIONS MANAGEMENT • MIT SLOAN MANAGEMENT REVIEW 1 Figure 1 Increasing Variability of Orders up the Supply Chain Consumer Sales Retailer's Orders to Manufacturer 20 20 15 15 10 10 Order Quantity 5 Order Quantity 5 0 0 Time Time Wholesaler's Orders to Manufacturer Manufacturer's Orders to Supplier 20 20 15 15 10 10 Order Quantity 5 Order Quantity 5 0 0 Time Time inventory supply. Distorted information has led every distributors’ warehouses, and store warehouses along entity in the supply chain — the plant warehouse, a the distribution channel have inventory stockpiles. manufacturer’s shuttle warehouse, a manufacturer’s And in the pharmaceutical industry, there are duplicat- market warehouse, a distributor’s central warehouse, ed inventories in a supply chain of manufacturers such the distributor’s regional warehouses, and the retail as Eli Lilly or Bristol-Myers Squibb, distributors such store’s storage space — to stockpile because of the as McKesson, and retailers such as Longs Drug Stores. high degree of demand uncertainties and variabili- Again, information distortion can cause the total in- ventory in this supply chain to exceed 100 days of sup- ply. With inventories of raw materials, such as integrat- he ordering patterns share a ed circuits and printed circuit boards in the computer common, recurring theme: the industry and antibodies and vial manufacturing in the variabilities of an upstream pharmaceutical industry, the total chain may contain T more than one year’s supply. site are always greater than those In a supply chain for a typical consumer product, of the downstream site. even when consumer sales do not seem to vary much, there is pronounced variability in the retailers’ orders to the wholesalers (see Figure 1). Orders to the manu- ties. It’s no wonder that the ECR reports estimated a facturer and to the manufacturers’ supplier spike even potential $30 billion opportunity from streamlining more. To solve the problem of distorted information, the inefficiencies of the grocery supply chain.2 companies need to first understand what creates the Other industries are in a similar position. Computer bullwhip effect so they can counteract it. Innovative factories and manufacturers’ distribution centers, the companies in different industries have found that they 94 LEE ET AL. SLOAN MANAGEMENT REVIEW/SPRING 1997 LESSONS IN OPERATIONS MANAGEMENT • MIT SLOAN MANAGEMENT REVIEW 2 can control the bullwhip effect and improve their sup- The outcomes of the beer game are the conse- ply chain performance by coordinating information quence of many behavioral factors, such as the players’ and planning along the supply chain. perceptions and mistrust. An important factor is each player’s thought process in projecting the demand pat- Causes of the Bullwhip Effect tern based on what he or she observes. When a down- stream operation places an order, the upstream man- Perhaps the best illustration of the bullwhip effect is ager processes that piece of information as a signal the well-known “beer game.”3 In the game, partici- about future product demand. Based on this signal, pants (students, managers, analysts, and so on) play the upstream manager readjusts his or her demand the roles of customers, retailers, wholesalers, and sup- forecasts and, in turn, the orders placed with the sup- pliers of a popular brand of beer. The participants pliers of the upstream operation. We contend that de- cannot communicate with each other and must make mand signal processing is a major contributor to the order decisions based only on orders from the next bullwhip effect. downstream player. The ordering patterns share a For example, if you are a manager who has to de- common, recurring theme: the variabilities of an up- termine how much to order from a supplier, you use a stream site are always greater than those of the down- simple method to do demand forecasting, such as ex- stream site, a simple, yet powerful illustration of the ponential smoothing. With exponential smoothing, bullwhip effect. This amplified order variability may future demands are continuously updated as the new be attributed to the players’ irrational decision making. daily demand data become available. The order you Indeed, Sterman’s experiments showed that human be- send to the supplier reflects the amount you need to havior, such as misconceptions about inventory and replenish the stocks to meet the requirements of future demand information, may cause the bullwhip effect.4 demands, as well as the necessary safety stocks. The fu- In contrast, we show that the bullwhip effect is a ture demands and the associated safety stocks are up- consequence of the players’ rational behavior within dated using the smoothing technique. With long lead the supply chain’s infrastructure. This important dis- times, it is not uncommon to have weeks of safety tinction implies that companies wanting to control the stocks. The result is that the fluctuations in the order bullwhip effect have to focus on modifying the chain’s quantities over time can be much greater than those in infrastructure and related processes rather than the de- the demand data. cision makers’ behavior. Now, one site up the supply chain, if you are the We have identified four major causes of the bull- manager of the supplier, the daily orders from the man- whip effect: ager of the previous site constitute your demand. If you 1. Demand forecast updating are also using exponential smoothing to update your 2. Order batching forecasts and safety stocks, the orders that you place 3. Price fluctuation with your supplier will have even bigger swings. For an 4. Rationing and shortage gaming example of such fluctuations in demand, see Figure 2. Each of the four forces in concert with the chain’s As we can see from the figure, the orders placed by the infrastructure and the order managers’ rational deci- dealer to the manufacturer have much greater variabili- sion making create the bullwhip effect.