SUNIL Chopra_Meindl_SCM

3 {{{ Supply Chain Drivers and Metrics LEARNING OBJECTIVES After reading this chapter, you will be able to 1. Describe key financial measures of firm performance. 2. Identify the major drivers of supply chain performance. 3. Discuss the role of each driver in creating strategic fit between the supply chain strategy and the competitive strategy. 4. Define the key metrics that track the performance of the supply chain in terms of each driver. In this chapter, our goal is to link key financial measures of firm performance to supply chain performance. We introduce the three logistical drivers—facilities, inventory, and transportation—and the three cross-functional drivers—information, sourcing, and pricing—that determine the performance of any supply chain. We discuss how these drivers are used in the design, planning, and operation of the supply chain. We define several metrics that can be used to gauge the performance of each driver and its impact on financial performance. 3.1 FINANCIAL MEASURES OF PERFORMANCE In Chapter 1,we discussed how growing the supply chain surplus is the ultimate goal of a supply chain. Our premise was that growing the surplus allows for a growth of supply chain profitability, which facilitates an improvement in the financial performance of each member of the supply chain. In this section, we define important financial measures that are reported by a firm and impacted by supply chain performance. In later sections, we then link supply chain drivers and associated metrics to the various financial measures. The definitions of financial measures in this section are taken from Dyckman, Magee, and Pfeiffer (2011). From a shareholder perspective, return on equity (ROE) is the main summary measure of a firm’s performance. ROE = Net Income Average Shareholder Equity 38

Chapter 3 • Supply Chain Drivers and Metrics 39 Whereas ROE measures the return on investment made by a firm’s shareholders, return on assets (ROA) measures the return earned on each dollar invested by the firm in assets. Earnings before interest Net income + [Interest expense * (1 - Tax rate)] ROA = = Average total assets Average total assets Consider Amazon.com’s financial performance shown in Table 3-1. In 2009, Amazon achieved ROE ϭ 902ր5,257 ϭ 17.2 percent {1,152ր6,864 ϭ 16.8 percent in 2010} and ROA ϭ [902ϩ34*(1Ϫ.35)]ր13,813 ϭ 6.7 percent {[1,152 ϩ 39*(1 – .35)]ր18,797 ϭ 6.3 percent in 2010}. The difference between ROE and ROA is referred to as return on financial leverage (ROFL). In 2009, Amazon had ROFL ϭ 17.2 Ϫ 6.7 ϭ 10.5 percent {16.8Ϫ6.3 ϭ 10.5 percent in 2010}. ROFL captures the amount of ROE that can be attributed to financial leverage (accounts payable, debt, etc.). In Amazon’s case, a significant portion of the financial leverage in 2009 and 2010 came from accounts payable rather than debt. Thus, an important ratio that defines financial leverage is accounts payable turnover (APT). Cost of goods sold APT = Accounts payable Table 3-1 Selected Financial Data for Amazon.com Inc. Year ended December 31 ($ millions) 2010 2009 2008 Net operating revenues 34,204 24,509 19,166 Cost of goods sold 26,561 18,978 14,896 Gross profit Selling, general, and administrative expense 7,643 5,531 4,270 Operating income 6,237 4,402 3,428 Interest expense 1,406 1,129 Other income (loss) – net 842 Income before income taxes 39 34 71 Income taxes 130 66 Net income 1,497 1,161 130 352 253 901 Assets 1,152 902 247 645 Cash and cash equivalents 3,777 3,444 Short-term investments 4,985 2,922 2,769 Net receivables 1,783 1,260 958 Inventories 3,202 2,171 Total current assets 13,747 9,797 1,031 Property, plant and equipment 2,414 1,290 1,399 Goodwill 1,349 1,234 6,157 Other assets 1,265 1,492 Total assets 18,797 13,813 854 438 Liabilities and Stockholder Equity 10,372 7,364 705 8,314 Accounts payable 10,372 7,364 Short-term debt 109 4,687 Total current liability 1,561 59 Long-term debt 11,933 1,083 Other liabilities 6,864 8,556 4,746 Total liabilities 5,257 533 Stockholder equity 363 5,642 2,672

40 Chapter 3 • Supply Chain Drivers and Metrics In Amazon’s case, in 2009 APT ϭ 18,978ր7,364 ϭ 2.58 {26,561/10,372 ϭ 2.56 in 2010}. A small APT indicates that Amazon was able to use the money it owed suppliers to finance a considerable fraction of its operations. In 2009, Amazon effectively financed its own operations for about 52ր2.58 ϭ 20.18 {52ր2.56 ϭ 20.31 in 2010} weeks with its suppliers’ money. A low value of APT helps Amazon improve its financial performance. ROA can be written as the product of two ratios—profit margin and asset turnover—as shown below: ROA = Earnings before interest * Sales revenue 1Asset turnover2 1Profit margin2 Sales revenue Total assets Thus, a firm can increase ROA by growing the profit margin and/or increasing the asset turnover. In 2009, Amazon achieved a profit margin of [902 ϩ 34*(1Ϫ.35)]/24,509 ϭ 3.8 per- cent {[1,152 ϩ 39*(1 Ϫ .35)]ր34,204 ϭ 3.4 percent in 2010}. Profit margin can be improved by getting better prices or by reducing the various expenses incurred. A responsive supply chain can allow a firm to provide high value to a customer, thus potentially getting higher prices. Good supply chain management can also allow a firm to decrease the expenses incurred to serve customer demand. In Amazon’s case, a significant expense is outbound shipping cost. In its 2009 annual report, the company reported outbound shipping costs of $1.77 billion. After accounting for shipping revenue, the net loss on outbound shipping was reported to be $849 million, about the same order of magnitude as net income. Clearly, a reduction in outbound shipping costs can have a significant impact on Amazon’s profit margin. The key components of asset turnover are accounts receivable turnover (ART); inventory turnover (INVT); and property, plant and equipment turnover (PPET). These are defined as follows: ART = Sales revenue ; INVT = Cost of goods sold = Sales revenue ; PPET Accounts receivable Inventories PP&E Amazon achieved accounts receivable turnover of 24,509ր1,260 ϭ 19.45 {34,204ր1,783 ϭ 19.18 in 2010} in 2009. Amazon collected its money from sales relatively quickly (in about 52ր19.45 ϭ 2.7 weeks on average in 2009) after it made a sale. Amazon turned its inventory about 18,978/2,171 ϭ 8.74 {26,561/3,202 ϭ 8.30 in 2010} times and had PPET ϭ 24,509ր1,290 ϭ 19.00 {34,204/2,414 ϭ 14.17 in 2010} in 2009. Thus, inventory sat with Amazon in 2009 for about 52/8.74 ϭ 5.95 {52/8.30 ϭ 6.27 in 2010} weeks on average, and each dollar invested in property, plant and equipment supported about $19 {$14.17 in 2010} of sales in 2009. Observe that Amazon saw its inventory turns and PPET decrease in 2010 relative to 2009. Amazon can improve its asset turnover by turning its inventory more quickly or using its existing ware- housing and technology infrastructure to support a higher level of sales (or decreasing the warehousing and technology infrastructure needed to support the existing level of sales). Another useful metric is the cash-to-cash (C2C) cycle, which roughly measures the average amount of time from when cash enters the process as cost to when it returns as collected revenue. C2C = - weeks payable (1/APT) + weeks in inventory (1/INVT) + weeks receivable (1/ART) In Amazon’s case, we obtain C2C ϭ –20.18 ϩ 5.95 ϩ 2.70 ϭ –11.53 {–20.31 ϩ 6.27 ϩ 2.71 ϭ –11.33 in 2010} in 2009. In 2009 and 2010, Amazon was collecting its money from the sale of products more than 11 weeks before it had to pay its suppliers. As we discussed earlier, this allowed Amazon to achieve significant financial leverage without having to take on debt. There are two important measures, however, that are not explicitly part of a firm’s financial statements. They are markdowns and lost sales. Markdowns represent the discounts required to convince customers to buy excess inventory. Financial statements show only the revenue received from sales, not the revenue that “could” have been received. For General Motors (GM), one of the biggest problems in the early part of the 21st century was the discounts required to move excess inventory from dealer lots. These discounts significantly hurt financial performance. In 2010, one

Chapter 3 • Supply Chain Drivers and Metrics 41 of the biggest improvements in financial performance for GM was its ability to sell its cars with much smaller discounts because the supply chain had far less excess inventory. Lost sales repre- sent customer sales that did not materialize because of the absence of products the customer want- ed to buy. Every lost sale corresponds to product margin that is lost. Both markdowns and lost sales reduce net income and arguably represent the biggest impact of supply chain performance on the financial performance of a firm. Firms like Amazon, Wal-Mart, and Zara that achieve strong financial performance do so in large part because their supply chains allow them to better match supply and demand, thereby reducing markdowns and lost sales. From our brief discussion of Amazon’s financial statements, supply chain management activities such as planning, transportation, inventory, and warehousing clearly have a significant impact on financial performance. In the next section, we identify key drivers of supply chain performance that influence the financial performance of a firm. 3.2 DRIVERS OF SUPPLY CHAIN PERFORMANCE The strategic fit discussed in Chapter 2 requires that a company’s supply chain achieve the balance between responsiveness and efficiency that best supports the company’s competitive strategy. To understand how a company can improve supply chain performance in terms of responsiveness and efficiency, we must examine the logistical and cross–functional drivers of supply chain perform- ance: facilities, inventory, transportation, information, sourcing, and pricing. These drivers interact to determine the supply chain’s performance in terms of responsiveness and efficiency. These driv- ers also impact the financial measures discussed in Section 3.1. The goal is to structure the drivers to achieve the desired level of responsiveness at the lowest possible cost, thus improving the supply chain surplus and the firm’s financial performance. First we define each driver and discuss its impact on the performance of the supply chain. 1. Facilities are the actual physical locations in the supply chain network where prod- uct is stored, assembled, or fabricated. The two major types of facilities are production sites and storage sites. Decisions regarding the role, location, capacity, and flexibility of facilities have a significant impact on the supply chain’s performance. For example, in 2009, Amazon increased the number of warehousing facilities (observe increase in Property, plant and equipment, in Table 3-1) located close to customers to improve its responsiveness. In contrast, Blockbuster tried to improve its efficiency in 2010 by shutting down many facilities even though it reduced responsiveness. Facility costs show up under property, plant and equipment, if facilities are owned by the firm or under selling, general, and administrative if they are leased. 2. Inventory encompasses all raw materials, work in process, and finished goods within a supply chain. The inventory belonging to a firm is reported under assets. Changing inventory policies can dramatically alter the supply chain’s efficiency and responsiveness. For example, W.W. Grainger makes itself responsive by stocking large amounts of inventory and satisfying customer demand from stock even though the high inventory levels reduce efficiency. Such a practice makes sense for Grainger because its products hold their value for a long time. A strate- gy using high inventory levels can be dangerous in the fashion apparel business where inventory loses value relatively quickly with changing seasons and trends. Rather than hold high levels of inventory, Spanish apparel retailer Zara has worked hard to shorten new product and replenish- ment lead times. As a result, the company is very responsive but carries low levels of inventory. Zara thus provides responsiveness at low cost. 3. Transportation entails moving inventory from point to point in the supply chain. Transportation can take the form of many combinations of modes and routes, each with its own performance characteristics. Transportation choices have a large impact on supply chain responsiveness and efficiency. For example, a mail-order catalog company can use a faster mode of transportation such as FedEx to ship products, thus making its supply chain more responsive, but also less efficient given the high costs associated with using FedEx. McMaster-Carr and

42 Chapter 3 • Supply Chain Drivers and Metrics W.W. Grainger, however, have structured their supply chain to provide next-day service to most of their customers using ground transportation. They are providing a high level of responsiveness at lower cost. Outbound transportation costs of shipping to the customer are typically included in selling, general, and administrative expense, while inbound transportation costs are typically included in the cost of goods sold. 4. Information consists of data and analysis concerning facilities, inventory, transporta- tion, costs, prices, and customers throughout the supply chain. Information is potentially the biggest driver of performance in the supply chain because it directly affects each of the other drivers. Information presents management with the opportunity to make supply chains more responsive and more efficient. For example, Seven-Eleven Japan has used information to better match supply and demand while achieving production and distribution economies. The result is a high level of responsiveness to customer demand while production and replenishment costs are lowered. Information technology–related expenses are typically included under either operating expense (typically under selling, general, and administrative expense) or assets. For example, in 2009, Amazon included $1.24 billion in technology expense under operating expense and another $551 million under fixed assets to be depreciated. 5. Sourcing is the choice of who will perform a particular supply chain activity such as production, storage, transportation, or the management of information. At the strategic level, these decisions determine what functions a firm performs and what functions the firm outsources. Sourcing decisions affect both the responsiveness and efficiency of a supply chain. After Motorola outsourced much of its production to contract manufacturers in China, it saw its efficiency improve but its responsiveness suffer because of the long distances. To make up for the drop in responsiveness, Motorola started flying in some of its cell phones from China even though this choice increased transportation cost. Flextronics, an electronics contract manufactur- er, is hoping to offer both responsive and efficient sourcing options to its customers. It is trying to make its production facilities in high-cost locations very responsive while keeping its facilities in low-cost countries efficient. Flextronics hopes to become an effective source for all customers using this combination of facilities. Sourcing costs show up in the cost of goods sold, and monies owed to suppliers are recorded under accounts payable. 6. Pricing determines how much a firm will charge for the goods and services that it makes available in the supply chain. Pricing affects the behavior of the buyer of the good or serv- ice, thus affecting supply chain performance. For example, if a transportation company varies its charges based on the lead time provided by the customers, it is likely that customers who value efficiency will order early and customers who value responsiveness will be willing to wait and order just before they need a product transported. Differential pricing provides responsiveness to customers that value it and low cost to customers that do not value responsiveness as much. Any change in pricing impacts revenues directly but could also affect costs based on the impact of this change on the other drivers. Our definitions of these drivers attempt to delineate logistics and supply chain management. Supply chain management includes the use of logistical and cross-functional drivers to increase the supply chain surplus. Cross-functional drivers have become increasingly important in raising the supply chain surplus in recent years. While logistics remains a major part, supply chain man- agement is increasingly becoming focused on the three cross-functional drivers. It is important to realize that these drivers do not act independently but interact to deter- mine the overall supply chain performance. Good supply chain design and operation recognize this interaction and make the appropriate trade-offs to deliver the desired level of responsiveness. Consider, for example, the furniture industry in the United States. Low-cost furniture sourced from Asia is available at many discount retailers. The primary goal of this supply chain is to deliver a low price and acceptable quality. Variety is typically low and retailers such as Wal-Mart stock inventory of finished goods. The low variety and stable replenishment orders allow furni- ture manufacturers in Asia to focus on efficiency. Given the available inventory, low-cost modes

Chapter 3 • Supply Chain Drivers and Metrics 43 of transportation from Asia are used. In this instance, relatively low-cost inventory at the retailer allows the supply chain to become efficient by lowering transportation and production costs. In contrast, some U.S. furniture makers have chosen to focus on providing variety. Given the high variety and high prices, keeping inventory of all variants at a retailer would be very expensive. In this case, the supply chain has been designed so that the retailer carries little inventory. Customers place their orders with the retailer by seeing one variant of the furniture and selecting among the various options. The supply chain is made responsive by using information technology to convey order information effectively, structuring flexible manufacturing facilities to be able to produce in small lots, and using responsive transportation to deliver the furniture to the customer. In this instance, responsive facilities, transportation, and information are used to lower inventory costs. As the rest of this chapter will illustrate, the key to achieving strategic fit and strong financial performance across the supply chain is to structure the supply chain drivers appropriately to provide the desired level of responsiveness at the lowest possible cost. Doheny et al. (2010) point out that supply chain performance affects nearly 35 percent of the financial performance of apparel retailers. As a percentage of sales, they state that mark- downs, representing 10–30 percent of sales, and lost sales, representing 5–10 percent of sales, are the dominant drivers of retailers’ financial performance. They further state that transportation represents 2–5 percent, warehousing 1–3 percent, store product handling 3–5 percent, and inven- tory costs 2–5 percent of sales. While the precise fraction will vary for different supply chains, it is evident that supply chain performance along the six drivers has a significant influence on a firm’s financial performance. Before we discuss each of the six drivers in detail, we put these drivers into a framework that helps clarify the role of each in improving supply chain performance. 3.3 FRAMEWORK FOR STRUCTURING DRIVERS Recall from Chapter 2 that the goal of a supply chain strategy is to strike the balance between responsiveness and efficiency that fits with the competitive strategy. To reach this goal, a com- pany must structure the right combination of the three logistical and three cross-functional drivers. The combined impact of these drivers then determines the responsiveness and the profits of the entire supply chain. We provide a visual framework for supply chain decision making in Figure 3-1. Most com- panies begin with a competitive strategy and then decide what their supply chain strategy ought to be. The supply chain strategy determines how the supply chain should perform with respect to efficiency and responsiveness. The supply chain must then use the three logistical and three cross-functional drivers to reach the performance level the supply chain strategy dictates and maximize the supply chain profits. Although this framework is generally viewed from the top down, in many instances, a study of the six drivers may indicate the need to change the supply chain strategy and potentially even the competitive strategy. Consider this framework using Wal-Mart as an example. Wal-Mart’s competitive strategy is to be a reliable, low-cost retailer for a wide variety of mass-consumption goods. This strategy dic- tates that the ideal supply chain will emphasize efficiency but also maintain an adequate level of responsiveness in terms of product availability. Wal-Mart uses the three logistical and three cross- functional drivers effectively to achieve this type of supply chain performance. With the inventory driver, Wal-Mart maintains an efficient supply chain by keeping low levels of inventory. For instance, Wal-Mart pioneered cross-docking, a system in which inventory is not stocked in a ware- house but rather is shipped to stores from the manufacturer with a brief stop at a distribution center (DCs), where product is transferred from inbound trucks from the supplier to outbound trucks to the retail store. This significantly lowers inventory because products are stocked only at stores, not at both stores and warehouses. With respect to inventory, Wal-Mart favors efficiency over respon- siveness. On the transportation front, Wal-Mart runs its own fleet, to keep responsiveness high. This increases transportation cost, but the benefits in terms of reduced inventory and improved product

44 Chapter 3 • Supply Chain Drivers and Metrics Competitive Strategy Efficiency Supply Chain Responsiveness Strategy Supply Chain Structure Logistical Drivers Facilities Inventory Transportation Information Sourcing Pricing Cross-Functional Drivers FIGURE 3-1 Supply Chain Decision-Making Framework availability justify this cost in Wal-Mart’s case. In the case of facilities, Wal-Mart uses centrally located DCs within its network of stores to decrease the number of facilities and increase efficien- cy at each DC. Wal-Mart builds retail stores only where the demand is sufficient to justify having several of them supported by a DC, thereby increasing efficiency of its transportation assets. Wal-Mart has invested significantly more than its competitors in information technology, allowing the company to feed demand information across the supply chain to suppliers who manufacture only what is being demanded. As a result, Wal-Mart is a leader in its use of the information driver to improve responsiveness and decrease inventory investment. With regard to the sourcing driver, Wal-Mart identifies efficient sources for each product it sells. Wal-Mart feeds them large orders, allowing them to be efficient by exploiting economies of scale. Finally, for the pricing driver, Wal-Mart practices “every day low pricing” (EDLP) for its products. This ensures that customer demand stays steady and does not fluctuate with price variations. The entire supply chain then focuses on meeting this demand in an efficient manner. Wal-Mart uses all the supply chain drivers to achieve the right balance between responsiveness and efficiency so that its competitive strategy and supply chain strategy are in harmony. We devote the next six sections to a detailed discussion of each of the three logistical and three cross-functional drivers, their roles in the supply chain, and their impact on financial performance. 3.4 FACILITIES In this section, we discuss the role that facilities play in the supply chain and critical facility-related decisions that supply chain managers need to make.

Chapter 3 • Supply Chain Drivers and Metrics 45 Role in the Supply Chain If we think of inventory as what is being passed along the supply chain and transportation as how it is passed along, then facilities are the where of the supply chain. They are the locations to or from which the inventory is transported. Within a facility, inventory is either transformed into another state (manufacturing) or it is stored (warehousing). Role in the Competitive Strategy Facilities are a key driver of supply chain performance in terms of responsiveness and efficiency. For example, companies can gain economies of scale when a product is manufactured or stored in only one location; this centralization increases efficiency. The cost reduction, however, comes at the expense of responsiveness, as many of a company’s customers may be located far from the production facility. The opposite is also true. Locating facilities close to customers increases the number of facilities needed and consequently reduces efficiency. If the customer demands and is willing to pay for the responsiveness that having numerous facilities adds, however, then this facilities decision helps meet the company’s competitive strategy goals. EXAMPLE 3-1 Toyota and Honda Both Toyota and Honda use facilities decisions to be more responsive to their customers. These companies have an end goal of opening manufacturing facilities in every major market that they enter. While there are other benefits to opening local facilities, such as protection from currency fluctuation and trade barriers, the increase in responsiveness plays a large role in Toyota’s and Honda’s decision to place facilities in their local markets. The flexibility of Honda facilities to assemble both SUVs and cars in the same plant allowed the company to keep costs down in the downturn of 2008. While competitors’ SUV production facilities were idle, Honda facilities maintained a high level of utilization. Components of Facilities Decisions Decisions regarding facilities are a crucial part of supply chain design. We now identify components of facilities decisions that companies must analyze. ROLE Firms must decide whether production facilities will be flexible, dedicated, or a combi- nation of the two. Flexible capacity can be used for many types of products but is often less efficient, whereas dedicated capacity can be used for only a limited number of products but is more efficient. Firms must also decide whether to design a facility with a product focus or a functional focus. A product-focused facility performs all functions (e.g., fabrication and assembly) needed for producing a single type of product. A functional-focused facility performs a given set of functions (e.g., fabrication or assembly) on many types of products. A product focus tends to result in more expertise about a particular type of product at the expense of the functional expertise that comes from a functional methodology. For warehouses and DCs, firms must decide whether they will be primarily cross-docking facilities or storage facilities. At cross-docking facilities, inbound trucks from suppliers are unloaded; the product is broken into smaller lots and is quickly loaded onto store-bound trucks. Each store-bound truck carries a variety of products, some from each inbound truck. For storage facilities, firms must decide on the products to be stored at each facility. LOCATION Deciding where a company will locate its facilities constitutes a large part of the design of a supply chain. A basic trade-off here is whether to centralize in order to gain economies of scale or to decentralize to become more responsive by being closer to the customer.

46 Chapter 3 • Supply Chain Drivers and Metrics Companies must also consider a host of issues related to the various characteristics of the local area in which the facility is situated. These include macroeconomic factors, quality of workers, cost of workers, cost of facility, availability of infrastructure, proximity to customers, the location of that firm’s other facilities, tax effects, and other strategic factors. CAPACITY Companies must also determine a facility’s capacity to perform its intended func- tion or functions. A large amount of excess capacity allows the facility to respond to wide swings in the demands placed on it. Excess capacity, however, costs money and therefore can decrease efficiency. A facility with little excess capacity will likely be more efficient per unit of product it produces than one with a lot of unused capacity. The high-utilization facility, however, will have difficulty responding to demand fluctuations. Therefore, a company must make a trade-off to determine the right amount of capacity to have at each of its facilities. FACILITY-RELATED METRICS Facility-related decisions impact both the financial performance of the firm and the supply chain’s responsiveness to customers. On the financial side, facilities decisions impact the cost of goods sold and the assets in property plant and equipment. A manag- er should track the following facility-related metrics that influence supply chain performance: • Capacity measures the maximum amount a facility can process. • Utilization measures the fraction of capacity that is currently being used in the facility. Utilization affects both the unit cost of processing and the associated delays. Unit costs tend to decline (PPET increases) and delays increase with increasing utilization. • Processing/setup/down/idle time measure the fraction of time that the facility was processing units, being set up to process units, unavailable because it was down, or idle because it had no units to process. Ideally, utilization should be limited by demand and not setup or downtime. • Production cost per unit measures the average cost to produce a unit of output. These costs may be measured per unit, per case, or per pound depending on the product. • Quality losses measure the fraction of production lost due to defects. Quality losses hurt both financial performance and responsiveness. • Theoretical flow/cycle time of production measures the time required to process a unit if there are absolutely no delays at any stage. • Actual average flow/cycle time measures the average actual time taken for all units processed over a specified duration such as a week or month. The actual flow/cycle time includes the the- oretical time and any delays. This metric should be used when setting due dates for orders. • Flow time efficiency is the ratio of the theoretical flow time to the actual average flow time. Low values for flow time efficiency indicate that a large fraction of time is spent waiting. • Product variety measures the number of products/product families processed in a facility. Processing costs and flow times are likely to increase with product variety. • Volume contribution of top 20 percent SKUs and customers measures the fraction of total volume processed by a facility that comes from the top 20 percent SKUs or customers. An 80/20 outcome in which the top 20 percent contribute 80 percent of volume indicates likely benefits from focusing the facility where separate processes are used to process the top 20 percent and the remaining 80 percent. • Average production batch size measures the average amount produced in each produc- tion batch. Large batch sizes will decrease production cost but increase inventories. • Production service level measures the fraction of production orders completed on time and in full. OVERALL TRADE-OFF: RESPONSIVENESS VERSUS EFFICIENCY The fundamental trade-off that managers face when making facilities decisions is between the cost of the number, location, capacity, and type of facilities (efficiency) and the level of responsiveness that these facilities

Chapter 3 • Supply Chain Drivers and Metrics 47 provide the company’s customers. Increasing the number of facilities increases facility and inventory costs but decreases transportation costs and reduces response time. Increasing the flexibility or capacity of a facility increases facility costs but decreases inventory costs and response time. 3.5 INVENTORY In this section, we discuss the role that inventory plays in the supply chain and how managers use inventory to drive supply chain performance. Role in the Supply Chain Inventory exists in the supply chain because of a mismatch between supply and demand. This mismatch is intentional at a steel manufacturer, where it is economical to manufacture in large lots that are then stored for future sales. The mismatch is also intentional at a retail store where inventory is held in anticipation of future demand. An important role that inventory plays in the supply chain is to increase the amount of demand that can be satisfied by having the prod- uct ready and available when the customer wants it. Another significant role that inventory plays is to reduce cost by exploiting economies of scale that may exist during production and distribution. Inventory impacts the assets held, the costs incurred, and responsiveness provided in the supply chain. High levels of inventory in an apparel supply chain improve responsiveness but also leave the supply chain vulnerable to the need for markdowns, lowering profit margins. Low levels of inventory improve inventory turns but may result in lost sales if customers are unable to find products they are ready to buy. Inventory also has a significant impact on the material flow time in a supply chain. Material flow time is the time that elapses between the point at which material enters the supply chain to the point at which it exits. For a supply chain, throughput is the rate at which sales occur. If inventory is represented by I, flow time by T, and throughput by D, the three can be related using Little’s law as follows: I = DT (3.1) For example, if the flow time of an auto assembly process is 10 hours and the throughput is 60 units an hour, Little’s law tells us that the inventory is 60 ϫ 10 ϭ 600 units. If we were able to reduce inventory to 300 units while holding throughput constant, we would reduce our flow time to 5 hours (300/60). We note that in this relationship, inventory and throughput must have consistent units. The logical conclusion here is that inventory and flow time are synonymous in a supply chain because throughput is often determined by customer demand. Managers should use actions that lower the amount of inventory needed without increasing cost or reducing responsiveness, because reduced flow time can be a significant advantage in a supply chain. Role in the Competitive Strategy The form, location, and quantity of inventory allow a supply chain to range from being very low cost to very responsive. Large amounts of finished goods inventory close to customers allow a supply chain to be responsive but at a high cost. Centralized inventory in raw material form allows a supply chain to lower cost but at the expense of responsiveness. The goal of good sup- ply chain design is to find the right form, location, and quantity of inventory that provides the right level of responsiveness at the lowest possible cost.

48 Chapter 3 • Supply Chain Drivers and Metrics EXAMPLE 3-2 Amazon.com Amazon attempts to provide a wide variety of books (among other products) to its customers. Best-selling books are stocked in many regional warehouses close to customers for high responsiveness. Slower moving books are stocked at fewer warehouses to lower the cost of inventory at the expense of some responsiveness. Some of the slowest moving books are not held in inventory but are obtained from the publisher/distributor or printed on demand when requested by a customer. Amazon changes the form, location, and quantity of inventory it holds by the level of sales of a book to provide the right balance of responsiveness and efficiency. Components of Inventory Decisions We now identify major inventory-related decisions that supply chain managers must make to effectively create more responsive and more efficient supply chains. CYCLE INVENTORY Cycle inventory is the average amount of inventory used to satisfy demand between receipts of supplier shipments. The size of the cycle inventory is a result of the produc- tion, transportation, or purchase of material in large lots. Companies produce or purchase in large lots to exploit economies of scale in the production, transportation, or purchasing process. With the increase in lot size, however, comes an increase in carrying costs. As an example of a cycle stock decision, consider an online book retailer. This retailer’s sales average around 10 truckloads of books a month. The cycle inventory decisions the retailer must make are how much to order for replenishment and how often to place these orders. The e-retailer could order 10 truckloads once each month or it could order one truckload every three days. The basic trade-off supply chain managers face is the cost of holding larger lots of inventory (when cycle inventory is high) versus the cost of ordering product frequently (when cycle inventory is low). SAFETY INVENTORY Safety inventory is inventory held in case demand exceeds expectation; it is held to counter uncertainty. If the world were perfectly predictable, only cycle inventory would be needed. Because demand is uncertain and may exceed expectations, however, companies hold safety inventory to satisfy an unexpectedly high demand. Managers face a key decision when determining how much safety inventory to hold. For example, a toy retailer such as Toys “R” Us must calculate its safety inventory for the holiday buying season. If it has too much safety inven- tory, toys go unsold and may have to be discounted after the holidays. If the company has too little safety inventory, however, then Toys “R” Us loses sales, along with the margin those sales would have brought. Therefore, choosing safety inventory involves making a trade-off between the costs of having too much inventory and the costs of losing sales due to not having enough inventory. SEASONAL INVENTORY Seasonal inventory is built up to counter predictable seasonal variabil- ity in demand. Companies using seasonal inventory build up inventory in periods of low demand and store it for periods of high demand when they will not have the capacity to produce all that is demanded. Managers face key decisions in determining whether to build seasonal inventory, and if they do build it, in deciding how much to build. If a company can rapidly change the rate of its production system at very low cost, then it may not need seasonal inventory, because the produc- tion system can adjust to a period of high demand without incurring large costs. However, if changing the rate of production is expensive (e.g., when workers must be hired or fired), then a company would be wise to establish a smooth production rate and build up its inventory during periods of low demand. Therefore, the basic trade-off supply chain managers face in determining how much seasonal inventory to build is the cost of carrying the additional seasonal inventory versus the cost of having a more flexible production rate.

Chapter 3 • Supply Chain Drivers and Metrics 49 LEVEL OF PRODUCT AVAILABILITY Level of product availability is the fraction of demand that is served on time from product held in inventory. A high level of product availability provides a high level of responsiveness but increases cost because much inventory is held but rarely used. In contrast, a low level of product availability lowers inventory holding cost but results in a higher fraction of customers who are not served on time. The basic trade-off when determining the level of product availability is between the cost of inventory to increase product availability and the loss from not serving customers on time. INVENTORY-RELATED METRICS Inventory-related decisions affect the cost of goods sold, the cash- to-cash cycle, and the assets held by the supply chain and its responsiveness to customers. A manager should track the following inventory-related metrics that influence supply chain performance: • Cash-to-cash cycle time is a high-level metric that includes inventories, accounts payable, and receivables. • Average inventory measures the average amount of inventory carried. Average inventory should be measured in units, days of demand, and financial value. • Inventory turns measure the number of times inventory turns over in a year. It is the ratio of average inventory to either the cost of goods sold or sales. • Products with more than a specified number of days of inventory identifies the products for which the firm is carrying a high level of inventory. This metric can be used to identify products that are in oversupply or to identify reasons that justify the high inventory, such as price discounts or being a very slow mover. • Average replenishment batch size measures the average amount in each replenishment order. The batch size should be measured by SKU in terms of both units and days of demand. It can be estimated by averaging over time the difference between the maximum and the minimum inventory (measured in each replenishment cycle) on hand. • Average safety inventory measures the average amount of inventory on hand when a replenishment order arrives. Average safety inventory should be measured by SKU in both units and days of demand. It can be estimated by averaging over time the minimum inventory on hand in each replenishment cycle. • Seasonal inventory measures the difference between the inflow of product (beyond cycle and safety inventory) and its sales that is purchased solely to deal with anticipated spikes in demand. • Fill rate (order/case) measures the fraction of orders/demand that were met on time from inventory. Fill rate should not be averaged over time but over a specified number of units of demand (say, every thousand, million, etc.). • Fraction of time out of stock measures the fraction of time that a particular SKU had zero inventory. This fraction can be used to estimate the lost sales during the stock out period. • Obsolete inventory measures the fraction of inventory older than a specified obsoles- cence date. OVERALL TRADE-OFF: RESPONSIVENESS VERSUS EFFICIENCY The fundamental trade-off that managers face when making inventory decisions is between responsiveness and efficiency. Increasing inventory generally makes the supply chain more responsive to the customer. A higher level of inventory also facilitates a reduction in production and transportation costs because of improved economies of scale in both functions. This choice, however, increases inventory holding cost. 3.6 TRANSPORTATION In this section, we discuss the role that transportation plays in the supply chain and key transportation-related decisions that supply chain managers must make.

50 Chapter 3 • Supply Chain Drivers and Metrics Role in the Supply Chain Transportation moves product between different stages in a supply chain and impacts both responsiveness and efficiency. Faster transportation allows a supply chain to be more responsive but reduces its efficiency. The type of transportation a company uses also affects the inventory and facility locations in the supply chain. Dell, for example, flies some components from Asia because doing so allows the company to lower the level of inventory it holds. Clearly, such a practice also increases responsiveness but decreases transportation efficiency because it is more costly than transporting parts by ship. Role in the Competitive Strategy Transportation allows a firm to adjust the location of its facilities and inventory to find the right balance between responsiveness and efficiency. A firm selling high-value items such as pacemakers may use rapid transportation to be responsive while centralizing its facilities and inventory to lower cost. In contrast, a firm selling low-value, high-demand items like light bulbs may carry a fair amount of inventory close to the customer but then use low-cost trans- portation like sea, rail, and full trucks to replenish this inventory from plants located in low-cost countries. EXAMPLE 3-3 Blue Nile Blue Nile is an online retailer of diamonds that has used responsive transportation with FedEx to ship diamonds to customers in the United States, Canada, and several countries in Europe and Asia. Given the high value of diamonds, Blue Nile offers free shipping for the overnight delivery. Responsive shipping, however, allows Blue Nile to centralize its inventory of diamonds and also eliminate the need for expensive storefronts. In spite of the high transportation costs, Blue Nile has very low costs compared to bricks-and-mortar retailers because of the low facility and inven- tory expenses. Blue Nile is thus able to offer significantly lower prices than its bricks-and-mortar competition. Components of Transportation Decisions We now identify key components of transportation that companies must analyze when designing and operating a supply chain. DESIGN OF TRANSPORTATION NETWORK The transportation network is the collection of transportation modes, locations, and routes along which product can be shipped. A company must decide whether transportation from a supply source will be direct to the demand point or will go through intermediate consolidation points. Design decisions also include whether or not multiple supply or demand points will be included in a single run. CHOICE OF TRANSPORTATION MODE The mode of transportation is the manner in which a product is moved from one location in the supply chain network to another. Companies can choose among air, truck, rail, sea, and pipeline as modes of transport for products. Today, information goods can also be sent via the Internet. Each mode has different characteristics with respect to the speed, size of shipments (individual parcels to pallets to full trucks to entire ships), cost of shipping, and flexibility that lead companies to choose one particular mode over the others. TRANSPORTATION-RELATED METRICS Inbound transportation decisions impact the cost of goods sold while outbound transportation costs are part of the selling, general, and administrative

Chapter 3 • Supply Chain Drivers and Metrics 51 expenses. Thus, transportation costs affect the profit margin. A manager should track the follow- ing transportation-related metrics that influence supply chain performance: • Average inbound transportation cost typically measures the cost of bringing product into a facility as a percentage of sales or cost of goods sold (COGS). Ideally, this cost should be measured per unit brought in, but this can be difficult. The inbound transportation cost is generally included in COGS. It is useful to separate this cost by supplier. • Average incoming shipment size measures the average number of units or dollars in each incoming shipment at a facility. • Average inbound transportation cost per shipment measures the average transportation cost of each incoming delivery. Along with the incoming shipment size, this metric identifies opportunities for greater economies of scale in inbound transportation. • Average outbound transportation cost measures the cost of sending product out of a facility to the customer. Ideally, this cost should be measured per unit shipped, but it is often measured as a percentage of sales. It is useful to separate this metric by customer. • Average outbound shipment size measures the average number of units or dollars on each outbound shipment at a facility. • Average outbound transportation cost per shipment measures the average transporta- tion cost of each outgoing delivery. Along with the outgoing shipment size, this metric identifies opportunities for greater economies of scale in outbound transportation. • Fraction transported by mode measures the fraction of transportation (in units or dollars) using each mode of transportation. This metric can be used to estimate if certain modes are overused or underutilized. OVERALL TRADE-OFF: RESPONSIVENESS VERSUS EFFICIENCY The fundamental trade-off for transportation is between the cost of transporting a given product (efficiency) and the speed with which that product is transported (responsiveness). Using fast modes of transport raises responsiveness and transportation cost but lowers the inventory holding cost. 3.7 INFORMATION In this section, we discuss the role that information plays in the supply chain, as well as key information-related decisions that supply chain managers must make. Role in the Supply Chain Good information can help improve the utilization of supply chain assets and the coordination of supply chain flows to increase responsiveness and reduce costs. Seven-Eleven Japan uses infor- mation to improve product availability while decreasing inventories. Wal-Mart uses information on shipments from suppliers to facilitate cross-docking and lower inventory and transportation expense. Li & Fung, a global trading group supplying time-sensitive consumer goods such as apparel, uses information on its third party manufacturers to source each order from the most appropriate supplier. Airlines routinely use information to offer the right number of seats at a discount price, leaving sufficient seats for business customers making reservations at the last minute and willing to pay a higher price. Each of these examples illustrates the importance of information as a key driver that can be used to provide higher responsiveness while simultane- ously improving efficiency. Role in the Competitive Strategy The right information can help a supply chain better meet customer needs at lower cost. The appropriate investment in information technology improves visibility of transactions and coordination of decisions across the supply chain. Coordination is essential if all stages of the

52 Chapter 3 • Supply Chain Drivers and Metrics supply chain are to work together toward a common goal. The goal in general should be to share the minimum amount of information required to achieve coordination because, beyond a certain point, the marginal cost of handling additional information increases, whereas the marginal benefit from the additional information decreases. The following examples illus- trate how information can be used to provide customized products and improve supply chain performance. EXAMPLE 3-4 Andersen Windows Andersen Windows, a major manufacturer of residential wood windows located in Bayport, Minnesota, has invested in an information system that enables the company to bring customized products to the market rapidly. This system, called “Window of Knowledge,” allows distributors and customers to design windows to custom-fit their needs. Users can select from a library of more than 50,000 components that can be combined in any number of ways. The system imme- diately gives the customer price quotes and automatically sends the order to the factory if the customer decides to buy. This information investment not only gives the customer a much wider variety of products, it also allows Andersen to be much more responsive to the customer, as it gets the customer’s order to the factory as soon as the order is placed. EXAMPLE 3-5 Sunsweet Growers Sunsweet Growers, a California-based dried fruit producer, implemented a supply chain sales and operations planning (S&OP) suite to replace its Excel-based planning system. The company has a highly seasonal supply with harvest taking place primarily during September and October. Demand is also seasonal with peak times occurring during the Christmas period. Good planning thus can be very valuable. Sunsweet’s goal when implementing the suite was twofold: Each function should operate with the same data and an early warning capability should alert planners and managers about any potential mismatches in supply and demand. After the implementation, production overruns at Sunsweet dropped from 30 percent to under 15 percent. Forecast accura- cy improved by 15 to 20 percent. The early warning system alerts allowed planners to react as much as two to three weeks earlier than before the implementation. Components of Information Decisions We now consider key components of information that a company must analyze to increase efficiency and improve responsiveness within its supply chain. PUSH VERSUS PULL When designing processes of the supply chain, managers must determine whether these processes are part of the push or pull phase in the chain. We discussed this distinc- tion in Chapter 1, but we mention it again because different types of systems require different types of information. Push systems start with forecasts that are used to build the master production schedule and roll it back, creating schedules for suppliers with part types, quantities, and delivery dates. Pull systems require information on actual demand to be transmitted extremely quickly throughout the entire chain so that production and distribution of products can reflect the real demand accurately. COORDINATION AND INFORMATION SHARING Supply chain coordination occurs when all stages of a supply chain work toward the objective of maximizing total supply chain profitability based on shared information. Lack of coordination can result in a significant loss of supply chain sur- plus. Coordination among different stages in a supply chain requires each stage to share appropriate

Chapter 3 • Supply Chain Drivers and Metrics 53 information with other stages. For example, if a supplier is to produce the right parts in a timely manner for a manufacturer in a pull system, the manufacturer must share demand and production information with the supplier. Information sharing is thus crucial to the success of a supply chain. SALES AND OPERATIONS PLANNING Sales and operations planning (S&OP) is the process of creating an overall supply plan (production and inventories) to meet the anticipated level of demand (sales). The S&OP process starts with sales and marketing communicating their needs to the supply chain, which in turn communicates to sales and marketing whether the needs can be met and at what cost. The goal of S&OP is to come up with an agreed-upon sales, production, and inventory plan that can be used to plan supply chain needs and project revenues and profits. The sales and operations plan becomes a critical piece of information to be shared across the sup- ply chain because it affects both the demand on a firm’s suppliers and the supply to its customers. ENABLING TECHNOLOGIES Many technologies exist to share and analyze information in the supply chain. Managers must decide which technologies to use and how to integrate them into their supply chain. Some of these technologies include the following: 1. Electronic data interchange (EDI) was developed in the 1970s to facilitate the placement of instantaneous, paperless purchase orders with suppliers. Its proprietary nature, however, required significant upfront investment and often some translation between the communicating parties. It did make transactions faster and more accurate than when they were paper based. 2. Relative to EDI, the Internet conveys much more information using a standard infra- structure allowing supply chains to improve both efficiency and responsiveness. The beginning of the 21st century has seen the Internet become the dominant medium of communication across all the macro processes (CRM, ISCM, and SRM discussed in Chapter 1) that link the supply chain from suppliers to customers. 3. Enterprise resource planning (ERP) systems provide the transactional tracking and global visibility of information from within a company and across its supply chain. This real-time information helps a supply chain improve the quality of its operational decisions. ERP systems keep track of the information, whereas the Internet provides one method with which to view this information. A more detailed discussion of ERP systems is in Chapter 17. 4. Supply chain management (SCM) software uses the information in ERP systems to provide analytical decision support in addition to the visibility of information. ERP systems show a company what is going on, while SCM systems help a company decide what it should do. A more detailed discussion of SCM systems is in Chapter 17. 5. Radio frequency identification (RFID) consists of an active or passive radio frequency (RF) tag applied to the item being tracked and an RF reader/emitter. A passive tag draws energy from the reader, whereas an active tag has its own battery and draws power from it. RFID has many potential uses. It can be used in manufacturing to check availability of the entire bill of materials. The technology can make the receiving of a truck much faster and cheaper. Full imple- mentation of RFID could eliminate the need for manual counting and bar-code scanning at the receiving dock. It can also be used to get an exact count of incoming items and items in storage. RFID technology, however, has yet to reach 100 percent accuracy, and its cost per unit is still high enough to make global acceptance difficult, even at the case level. INFORMATION-RELATED METRICS A manager should track the following information-related metrics that influence supply chain performance: • Forecast horizon identifies how far in advance of the actual event a forecast is made. The forecast horizon must be greater than or equal to the lead time of the decision that is driven by the forecast.

54 Chapter 3 • Supply Chain Drivers and Metrics • Frequency of update identifies how frequently each forecast is updated. The forecast should be updated somewhat more frequently than a decision will be revisited, so that large changes can be flagged and corrective action taken. • Forecast error measures the difference between the forecast and actual demand. The forecast error is a measure of uncertainty and drives all responses to uncertainty such as safety inventory or excess capacity. • Seasonal factors measure the extent to which the average demand in a season is above or below the average in the year. • Variance from plan identifies the difference between the planned production/inventories and the actual values. These variances can be used to raise flags that identify shortages and surpluses. • Ratio of demand variability to order variability measures the standard deviation of incoming demand and supply orders placed. A ratio less than one potentially indicates the existence of the bullwhip effect, which is discussed in Chapter 10. OVERALL TRADE-OFF: COMPLEXITY VERSUS VALUE Good information clearly helps a firm improve both its efficiency and responsiveness. There is a danger, however, in the assumption that more information is always better. As more information is shared across a supply chain, the complexity and cost of both the required infrastructure and the follow-up analysis grow exponen- tially. The marginal value provided by the information shared, however, diminishes as more and more information is available. It is thus important to evaluate the minimum information required to accomplish the desired objectives. For example, it may often be enough if aggregate sales are shared between a retailer and a manufacturer instead of detailed point-of-sale data. Aggregate information is cheaper to share and provides most of the value with regard to better production planning. The trade-off between complexity and value is important to consider when setting up the information infrastructure. 3.8 SOURCING In this section, we discuss the role that sourcing plays in the supply chain and key sourcing-related decisions that managers need to make. Role in the Supply Chain Sourcing is the set of business processes required to purchase goods and services. Managers must first decide whether each task will be performed by a responsive or efficient source and then whether the source will be internal to the company or a third party. Sourcing from low-cost countries allows a company like IKEA to provide the basic modules for the furniture it sells at low cost. Sourcing some of its PCs sold at Wal-Mart from China has allowed Dell to lower their cost. Meanwhile, Dell continues to produce in-house those machines for which responsiveness is required. As supply chains have globalized, many more sourcing options now offer both con- siderable opportunity and potential risks. Thus, sourcing decisions have a significant impact on supply chain performance. Role in the Competitive Strategy Sourcing decisions are crucial because they affect the level of efficiency and responsiveness the supply chain can achieve. In some instance, firms outsource to responsive third parties if it is too expensive for them to develop this responsiveness on their own. An example is the out- sourcing of next-day package delivery by all firms to a few package carriers because it is too expensive for a firm to develop next-day delivery capability on its own. In other instances, firms have kept the responsive process in-house to maintain control. An example is Zara,

Chapter 3 • Supply Chain Drivers and Metrics 55 which keeps responsive capacity in-house so it can respond quickly to orders as they arrive. Firms also outsource for efficiency if the third party can achieve significant economies of scale or has a lower underlying cost structure for other reasons. The following example illustrates how Cisco has sourced appropriately to be efficient for low-end products and responsive for high-end products. EXAMPLE 3-6 Cisco Cisco has outsourced almost all of its manufacturing. It does, however, have a sourcing strategy that varies by product type. For low-end products such as routers for home networks, Cisco aims for efficiency. These routers are produced and packed in China and shipped in bulk for sale in the United States. Cisco aims for the lowest cost manufacturing location and economies of scale in transportation because the targeted market segment values low cost. For high-end products, in contrast, Cisco outsources to contract manufacturers in the United States. These manufacturers are not low cost, but they are responsive and can serve the rapidly evolving needs of the high-end market. Components of Sourcing Decisions We now consider key sourcing decisions that are made within a firm. IN-HOUSE OR OUTSOURCE The most significant sourcing decision for a firm is whether to perform a task in-house or outsource it to a third party. Within a task such as transportation, man- agers must decide whether to outsource all of it, outsource only the responsive component, or outsource only the efficient component. This decision should be driven in part by its impact on the total supply chain surplus. It is best to outsource if the growth in total supply chain surplus is significant with little additional risk. SUPPLIER SELECTION Managers must decide on the number of suppliers they will have for a particular activity. They must then identify the criteria along which suppliers will be evaluated and how they will be selected. For the selection process, managers must decide whether they will use direct negotiations or resort to an auction. If an auction is used, it must be structured to ensure the desired outcome. PROCUREMENT Procurement is the process of obtaining goods and services within a supply chain. Managers must structure procurement with a goal of increasing supply chain surplus. For example, a firm should set up procurement for direct materials to ensure good coordination between the supplier and buyer. In contrast, the procurement of MRO products should be structured to ensure that transaction costs are low. SOURCING-RELATED METRICS Sourcing decisions directly impact the cost of goods sold and accounts payable. The performance of the source also impacts quality, inventories, and inbound transportation costs. A manager should track the following sourcing-related metrics that influence supply chain performance: • Days payable outstanding measures the number of days between when a supplier performed a supply chain task and when it was paid. • Average purchase price measures the average price at which a good or service was pur- chased during the year. The average price should be weighted by the quantity purchased at each price. • Range of purchase price measures the fluctuation in purchase price during a specified period. The goal is to identify if the quantity purchased correlated with the price.

56 Chapter 3 • Supply Chain Drivers and Metrics • Average purchase quantity measures the average amount purchased per order. The goal is to identify whether a sufficient level of aggregation is occurring across locations when placing an order. • Supply quality measures the quality of product supplied. • Supply lead time measures the average time between when an order is placed and when the product arrives. Long lead times reduce responsiveness and add to the inventory the supply chain must carry. • Fraction of on-time deliveries measures the fraction of deliveries from the supplier that were on time. • Supplier reliability measures the variability of the supplier’s lead time as well as the delivered quantity relative to plan. Poor supplier reliability hurts responsiveness and adds to the amount of inventory the supply chain must carry. OVERALL TRADE-OFF: INCREASE THE SUPPLY CHAIN SURPLUS Sourcing decisions should be made to increase the size of the total surplus to be shared across the supply chain. The total surplus is affected by the impact of sourcing on sales, service, production costs, inventory costs, transportation costs, and information costs. Outsourcing to a third party is meaningful if the third party raises the supply chain surplus more than the firm can on its own. In contrast, a firm should keep a supply chain function in-house if the third party cannot increase the supply chain surplus or if the risk associated with outsourcing is significant. 3.9 PRICING In this section, we discuss the role that pricing plays in the supply chain. Role in the Supply Chain Pricing is the process by which a firm decides how much to charge customers for its goods and services. Pricing affects the customer segments that choose to buy the product, as well as the customer’s expectations. This directly affects the supply chain in terms of the level of responsiveness required as well as the demand profile that the supply chain attempts to serve. Pricing is also a lever that can be used to match supply and demand especially when the supply chain is not very flexible. Short-term discounts can be used to eliminate supply surpluses or decrease seasonal demand spikes by moving some of the demand forward. In short, pricing is one of the most significant factors that affect the level and type of demand that the supply chain will face. Role in the Competitive Strategy Pricing is a significant attribute through which a firm executes its competitive strategy. For example, Costco, a membership-based wholesaler in the United States, has a policy that prices are kept steady but low. Customers expect low prices but are comfortable with a lower level of product availability. The steady prices also ensure that demand stays relatively stable. Costco serves a well-defined segment, and it can thus design an appropriate supply chain. The Costco supply chain aims to be efficient, at the expense of some responsiveness. In contrast, some man- ufacturing and transportation firms use pricing that varies with the response time desired by the customer. Through their pricing, these firms are targeting a broader set of customers, some of whom need responsiveness while others need efficiency. In this case, it becomes important for these firms to structure a supply chain that can meet the two divergent needs. Amazon uses a menu of shipping options and prices to identify customers who value responsiveness and those who value low cost. This identification allows the company to serve both effectively, as shown in the following example.

Chapter 3 • Supply Chain Drivers and Metrics 57 EXAMPLE 3-7 Amazon.com Amazon offers its customers a large menu of prices for products that are purchased from the com- pany. For example, in July 2008, a person purchasing two books worth $30 could use standard shipping (ships in 3–5 business days) at a cost of $4.98, two-day shipping (ships in 2 business days) at a cost of $13.97, one-day shipping (ships in 1 business day) at a cost of $22.97 or use free shipping (ships in 7–14 business days). The pricing menu allows Amazon to attract customers with varying levels of desired responsiveness. Whereas customers paying for one-day shipping impose a high degree of uncertainty on Amazon, customers opting for free shipping can be used to level out the workload at the warehouse over time. Amazon can thus use its pricing to provide responsiveness to those who value it while using customers who want a low price to help it improve its efficiency. Components of Pricing Decisions We now describe key components of pricing decisions that affect supply chain performance. PRICING AND ECONOMIES OF SCALE Most supply chain activities display economies of scale. Changeovers make small production runs more expensive per unit than large production runs. Loading and unloading costs make it cheaper to deliver a truckload to one location than four. In each case, the provider of the supply chain activity must decide how to price it appropri- ately to reflect these economies of scale. A commonly used approach is to offer quantity discounts. Care must be taken to ensure that quantity discounts offered are consistent with the economies of scale in the underlying process. Otherwise, there is a danger of customer orders being driven primarily by the quantity discounts even though the underlying process does not have significant economies of scale. EVERYDAY LOW PRICING VERSUS HIGH–LOW PRICING A firm such as Costco practices everyday low pricing at its warehouse stores, keeping prices steady over time. Costco will go to the extent of not offering any discount on damaged books to ensure its everyday low-pricing strategy. In contrast, most supermarkets practice high–low pricing and offer steep discounts on a subset of their product every week. The Costco pricing strategy results in relatively stable demand. The high–low pricing strategy results in a peak during the discount week, often followed by a steep drop in demand during the following weeks. The two pricing strategies lead to different demand profiles that the supply chain must serve. FIXED PRICE VERSUS MENU PRICING A firm must decide whether it will charge a fixed price for its supply chain activities or have a menu with prices that vary with some other attribute, such as the response time or location of delivery. If marginal supply chain costs or the value to the customer vary significantly along some attribute, it is often effective to have a pricing menu. We have already discussed Amazon as an example of a firm offering a menu that is somewhat consistent with the cost of providing the particular supply chain service. An example of when the pricing menu is somewhat inconsistent is seen at many MRO suppliers. They often allow customers to have their order shipped to them or to be picked up in person. A customer pays an additional shipping fee for home delivery but pays nothing for a personal pickup. The pick, pack, and deliver cost at the warehouse, however, is higher in the case of a personal pickup compared to home delivery. The pricing policy thus can lead to customer behavior that has a negative impact on profits. PRICING-RELATED METRICS Pricing directly affects revenues but can also affect production costs and inventories depending upon its impact on consumer demand. A manager should track

58 Chapter 3 • Supply Chain Drivers and Metrics the following pricing-related metrics. With menu pricing, each metric should be tracked sepa- rately for each segment in the menu: • Profit margin measures profit as a percentage of revenue. A firm needs to examine a wide variety of profit margin metrics to optimize its pricing, including dimensions such as type of margin (gross, net, etc.), scope (SKU, product line, division, firm), customer type, and others. • Days sales outstanding measures the average time between when a sale is made and when the cash is collected. • Incremental fixed cost per order measures the incremental costs that are independent of the size of the order. These include changeover costs at a manufacturing plant or order processing or transportation costs that are incurred independent of shipment size at a mail-order firm. • Incremental variable cost per unit measures the incremental costs that vary with the size of the order. These include picking costs at a mail-order firm or variable production costs at a manufacturing plant. • Average sale price measures the average price at which a supply chain activity was performed in a given period. The average should be obtained by weighting the price with the quantity sold at that price. • Average order size measures the average quantity per order. The average sale price, order size, incremental fixed cost per order, and incremental variable cost per unit help estimate the contribution from performing the supply chain activity. • Range of sale price measures the maximum and the minimum of sale price per unit over a specified time horizon. • Range of periodic sales measures the maximum and minimum of the quantity sold per period (day/week/month) during a specified time horizon. The goal is to understand any correlation between sales and price and any potential opportunity to shift sales by changing price over time. OVERALL TRADE-OFF: INCREASE FIRM PROFITS All pricing decisions should be made with the objective of increasing firm profits. This requires an understanding of the cost struc- ture of performing a supply chain activity and the value this activity brings to the supply chain. Strategies such as everyday low pricing may foster stable demand that allows for efficiency in the supply chain. Other pricing strategies may lower supply chain costs, defend market share, or even steal market share. Differential pricing may be used to attract customers with varying needs, as long as this strategy helps either increase revenues or shrink costs, preferably both. 3.10 SUMMARY OF LEARNING OBJECTIVES 1. Describe key financial measures of firm performance. The key financial measures of firm performance include return on equity; return on assets; accounts payable turnover; prof- it margin; asset turnover, accounts receivable turnover; inventory turns; property, plant and equipment turns; and cash-to-cash cycle. 2. Identify the major drivers of supply chain performance. The major drivers of supply chain performance are facilities, inventory, transportation, information, sourcing, and pricing. 3. Discuss the role of each driver in creating strategic fit between the supply chain strategy and the competitive strategy. A company achieving strategic fit has found the right balance between responsiveness and efficiency. Each driver affects this balance. Having more facilities generally makes a chain more responsive, while having fewer, central facilities creates higher efficiency. Holding higher levels of inventory increases the responsiveness of a supply

Chapter 3 • Supply Chain Drivers and Metrics 59 chain, while keeping inventory low increases the chain’s efficiency. Using faster modes of transportation increases a chain’s responsiveness, while using slower modes generally increas- es efficiency. Investing in information can vastly improve the supply chain performance on both dimensions. This investment, however, must be made based on the strategic position supported by the other drivers. Appropriate sourcing decisions raise supply chain profits by assigning sup- ply chain functions to the right party, which brings higher economies of scale or a higher level of aggregation of uncertainty. Pricing can be used to attract the right target customer segment. Differential pricing can be used to attract customers who value responsiveness as well as customers who want efficiency. The supply chain can then be structured to provide responsive- ness to some customers while improving overall efficiency. 4. Define the key metrics that track the performance of the supply chain in terms of each driver. Facility-related metrics are capacity, utilization, theoretical flow/cycle time of production, actual flow/cycle time, flow time efficiency, product variety, volume contribution of top 20 percent SKUs/customers, processing/setup/down/idle time, and average production batch size. Inventory-related metrics are average inventory, products with more than a specified num- ber of days of inventory, average replenishment batch size, average safety inventory, seasonal inventory, fill rate, and fraction of time out of stock. Transportation-related metrics are average inbound transportation cost, average incoming shipment size, average inbound transportation cost per shipment, average outbound transportation cost, average outbound shipment size, average outbound transportation cost per shipment, and fraction transported by mode. Information-related metrics are forecast horizon, forecast error, seasonal factors, variance from plan, and ratio of demand variability to order variability. Sourcing-related metrics are days payable outstanding, average purchase price, range of purchase price, average purchase quantity, fraction on-time deliveries, supply quality, and supply lead time. Pricing-related metrics are profit margin, days sales outstanding, incremental fixed cost per order, incremental variable cost per unit, average sale price, average order size, range of sale price, and range of periodic sales. Each of these metrics directly or indirectly impacts the financial metrics and the responsiveness to customers. Discussion Questions 1. How could a grocery retailer use inventory to increase the 7. What are some industries in which products have proliferated responsiveness of the company’s supply chain? and life cycles have shortened? How have the supply chains in these industries adapted? 2. How could an auto manufacturer use transportation to increase the efficiency of its supply chain? 8. How can the full set of logistical and cross-functional drivers be used to create strategic fit for a PC manufacturer targeting 3. How could a bicycle manufacturer increase responsiveness both time-sensitive and price-conscious customers? through its facilities? 9. On which supply chain drivers should a firm trying to shrink 4. How could an industrial supplies distributor use information its cash-to-cash cycle focus? to increase its responsiveness? 10. Would you expect a brick-and-mortar retailer or an online 5. Motorola has gone from manufacturing all its cell phones retailer to have a higher asset turnover? Which supply chain in-house to almost completely outsourcing the manufacturing. drivers impact asset turnover? What are the pros and cons of the two approaches? 6. How can a home-delivery company like Peapod use pricing of its delivery services to improve its profitability? Bibliography Doheny, Mike, Karl-Hendrik Magnus, Paulo Marchesan, Brian Dyckman, Thomas R., Robert P. Magee, and Glenn M. Pfeiffer. Ruwadi, Chris Turner, and Nursen Ulker. “Driving Financial Accounting. Westmont, IL: Cambridge Business Productivity in the Apparel Supply Chain.” December 17, Publishers, 2011. 2010. Available at https://operations-extranet.mckinsey.com/ html/knowledge/article/20101213_apparel_supply_chain.asp. Hofman, Debra. “The Hierarchy of Supply Chain Metrics.” Supply Chain Management Review (September 2004): 28–37.

60 Chapter 3 • Supply Chain Drivers and Metrics Marien, Edward J. “The Four Supply Chain Enablers.” Presutti, William D., Jr., and John R. Mawhinney. “The Supply Supply Chain Management Review (March–April 2000): Chain–Finance Link.” Supply Chain Management Review 60–68. (September 2007): 32–38. O’Marah, Kevin. “The Top Twenty Five Supply Chains.” Slone, Reuben E., J. Paul Dittman, and John T. Mentzer. The New Supply Chain Management Review (September 2007): Supply Chain Agenda: The Five Steps that Drive Real Value. 16–22. Boston: Harvard Business Press, 2010. Case Study Seven-Eleven Japan Co. convenience stores in Japan. After rejecting his initial request, Southland agreed in 1973 to a licensing agree- Established by Ito Yokado in 1973, Seven-Eleven Japan ment. In exchange for 0.6 percent of total sales, Southland set up its first store in Koto-ku, Tokyo, in May 1974. The gave Ito exclusive rights throughout Japan. In May 1974, company was first listed on the Tokyo Stock Exchange in the first Seven-Eleven convenience store opened in Tokyo. October 1979. On September 1, 2005, Seven & i Holdings Co. Ltd., was established as the holding company for This new concept was an immediate hit in Japan, Seven-Eleven Japan, Ito-Yokado, and Denny’s Japan. As and Seven-Eleven Japan experienced tremendous a result, detailed financial results for Seven-Eleven Japan growth. By 1979, there were already 591 Seven-Eleven have not been available since then and are only reported stores in Japan; by 1984, there were 2,001. Rapid growth as the convenience store portion of Seven & i Holdings. continued (Table 3-2), resulting in 12,753 stores by 2009. Seven-Eleven Japan realized a phenomenal growth between 1985 and 2009. During that period, the number On October 24, 1990, the Southland Corporation of stores increased from 2,299 to 12,753 and annual sales entered into bankruptcy protection. Southland asked for increased from 386 billion to 2,785 billion yen in Japan. Ito-Yokado’s help, and on March 5, 1991, IYG Holding Globally, the firm had over 40,000 convenience stores by was formed by Seven-Eleven Japan (48 percent) and January 2011 and was the world’s largest chain in terms of Ito-Yokado (52 percent). IYG acquired 70 percent of retail outlets. Global revenues for Seven & i from conven- Southland’s common stock for a total price of $430 million. ience store operations were 1,968 billion yen in 2009 with an operating income of 183.8 billion yen. The firm was Table 3-2 Stores and Annual Sales present in 38 of Japan’s 47 prefectures and opened 966 for Seven-Eleven Japan stores in Japan while closing 511 stores in 2009. Customer visits to Seven-Eleven outlets totaled 4.1 billion in 2007, Year Number of Annual Sales averaging almost 35 visits to a Seven-Eleven annually for Stores (billion yen) every person in Japan. 1974 15 0.7 Company History and Profile 1979 801 109.8 1984 2,299 386.7 Both Ito-Yokado and Seven-Eleven Japan were founded 1989 3,954 780.3 by Masatoshi Ito. He started his retail empire after 1994 5,905 1,392.3 World War II, when he joined his mother and elder 1999 8,153 1,963.9 brother and began to work in a small clothing store in 2000 8,602 2,046.6 Tokyo. By 1960, he was in sole control, and the single 2001 9,060 2,114.0 store had grown into a $3 million company. After a trip 2002 9,690 2,213.2 to the United States in 1961, Ito became convinced that 2003 10,303 2,343.2 superstores were the wave of the future. At that time, 2004 10,826 2,440.8 Japan was still dominated by mom-and-pop stores. Ito’s 2005 11,310 2,498.7 chain of superstores in the Tokyo area was instantly pop- 2006 11,735 2,533.5 ular and soon constituted the core of Ito-Yokado’s retail 2007 12,034 2,574.3 operations. 2008 12,298 2,762.5 2009 12,753 2,784.9 In 1972, Ito first approached the Southland Corporation about the possibility of opening Seven-Eleven

Chapter 3 • Supply Chain Drivers and Metrics 61 Table 3-3 Financial Figures for Seven & i (2008–2010) For Fiscal Years Ending February 28/29 2008 2009 2010 Total revenues (billion yen) 5,752.4 5,649.9 5,111.0 Total operating income (billion yen) 281.9 281.9 226.7 Convenience store revenues (billion yen) Convenience store operating income (billion yen) 2,395.7 2,308.7 1,968.6 201.0 213.4 183.8 In 2005, Seven & i Holdings was established effective in terms of same-store sales. In 2004, average through a stock transfer combining Seven-Eleven Japan, daily sales at the four major convenience store chains Ito-Yokado, and Denny’s Japan. In 2009, convenience excluding Seven-Eleven Japan totaled 484,000 yen. store operations from Seven-Eleven Japan and other Seven-Eleven stores, in contrast, had daily sales of subsidiaries in North America and China contributed 647,000 yen—more than 30 percent higher than the 38.3 percent of total revenues from operations and 80.9 competition put together. By 2009, average daily sales at percent of operating income for the Seven & i Holdings Seven-Eleven Japan stores had declined somewhat to Company (see Table 3-3 for details). The relative 616,000 yen. In 2004, Seven-Eleven’s operating income performance of convenience stores within Japanese of 165.7 billion yen positioned it as a leader not only of operations was even more dominant. The drop in finan- the convenience store sector but also of Japan’s retail cial performance in 2009 relative to 2008 was attributed industry as a whole. In terms of growth, its performance largely to the drop of gasoline prices in North America was even more impressive. In 2004, Seven-Eleven and the stronger yen. The discrepancy between Tables 3-2 accounted for 60 percent of the total net increase in the and 3-3 results because Table 3-2 reports sales (at both number of stores among the top 10 convenience store company-owned and franchised stores), whereas Table chains in Japan. This growth had been carefully 3-3 reports revenues for only Seven & i. planned, exploiting the core strengths that Seven-Eleven Japan had developed in the areas of information systems The Convenience Store Industry and Seven- and distribution systems. Eleven in Japan The Seven-Eleven Japan Franchise System The convenience store sector was one of the few business areas that continued to grow during the prolonged slow- Seven-Eleven Japan developed an extensive franchise net- down in Japan toward the end of the 20th century and the work and performed a key role in the daily operations of start of the 21st century. From 1991 to 2002, the number of this network. The Seven-Eleven Japan network included convenience stores in Japan increased from 19,603 to both company-owned stores and third-party-owned fran- almost 42,000. As a percentage of all retail stores in Japan, chises. In 2004, franchise commissions accounted for more this represented an increase from 1.2 percent to 3.2 than 68 percent of revenue from operations. To ensure effi- percent. During that period, annual sales at convenience ciency, Seven-Eleven Japan based its fundamental network stores more than doubled, from just over 3 trillion to 6.7 expansion policy on a market-dominance strategy. Entry trillion yen. As a percentage of all retail sales in Japan, this into any new market was built around a cluster of 50 to 60 represented an increase from 2.2 percent to 5.0 percent. stores supported by a distribution center. Such clustering gave Seven-Eleven Japan a high-density market presence Japan’s convenience store sector gradually and allowed it to operate an efficient distribution system. consolidated, with larger players growing and smaller Seven-Eleven Japan, in its 1994 annual report, listed the fol- operators shutting down. In 2004, the top 10 convenience lowing six advantages of the market-dominance strategy: store chains accounted for approximately 90 percent of Japan’s convenience stores. As the chains improved their • Boosted distribution efficiency operating structures and better leveraged economies of • Improved brand awareness scale, smaller operators found it hard to compete. • Increased system efficiency • Enhanced efficiency of franchise support services Seven-Eleven Japan had increased its share of the • Improved advertising effectiveness convenience store market since it opened. In 2008, • Prevented competitors’ entrance into the dominant Seven-Eleven was Japan’s leading convenience store operator, accounting for 34.3 percent market share in the area convenience store segment. Seven-Eleven was very

62 Chapter 3 • Supply Chain Drivers and Metrics Adhering to its dominant strategy, Seven-Eleven Table 3-4 Sales by Product Category in 2009 Japan opened the majority of its new stores in areas with existing clusters of stores. For example, the Aichi Percentage of Total Sales prefecture, where Seven-Eleven began opening stores in 2002, saw a large increase in 2004, with 108 new store Processed foods 28.3 openings. This represented more than 15 percent of the Fast foods 27.0 new Seven-Eleven stores opened in Japan that year. Fresh/daily foods 12.1 Nonfoods 32.6 Seven-Eleven had a limited geographic presence in Japan. In 2009, the company had stores in about 80 150 square meters, still significantly smaller than the percent (37 of 47) of the prefectures within Japan. size of most U.S. 7-Eleven stores. In 2009, daily sales at However, within prefectures where they were present, a store averaged 613,000 yen (about $7,558 in March stores tended to be dense. As the 2004 annual report 2011 at an exchange rate of about 81 yen to a U.S. dol- stated, “Filling in the entire map of Japan is not our lar), which was almost twice the average at a U.S. store. priority. Instead, we look for demand where Seven- Eleven stores already exist, based on our fundamental Seven-Eleven Japan offered its stores a choice area-dominance strategy of concentrating stores in from a set of 5,000 SKUs. Each store carried on average specific areas.” about 3,000 SKUs depending on local customer demand. Seven-Eleven Japan emphasized regional mer- With Seven-Eleven franchises being highly chandizing to cater precisely to local preferences. Each sought after, fewer than one of 100 applicants was store carried food items, beverages, magazines, and awarded a franchise (a testament to store profitability). consumer items such as soaps and detergents. The rela- The franchise owner was required to put a significant tive sales across product categories in 2009 for Seven amount of money up front. Half of this amount was Eleven Japan are given in Table 3-4. used to prepare the store and train the owner. The rest was used for purchasing the initial stock for the store. The food items were classified in four broad cate- In 1994, 45 percent of total gross profits at a store went gories: (1) chilled-temperature items including sandwich- to Seven-Eleven Japan, and the rest went to the store es, delicatessen products, and milk; (2) warm-temperature owner. The responsibilities of the two parties were as items including box lunches, rice balls, and fresh bread; follows. (3) frozen items including ice cream, frozen foods, and ice cubes; (4) and room-temperature items including canned Seven-Eleven Japan responsibilities: food, instant noodles, and seasonings. Processed food and fast-food items were big sellers for the stores. In 2009, • Develop supply and merchandise processed and fast foods contributed about 55 percent of • Provide the ordering system the total sales at each store. More than 1 billion rice balls • Pay for the system operation were sold in 2004; this amounted to each Japanese citizen • Supply accounting services eating approximately eight Seven-Eleven rice balls a year. • Provide advertising The top-selling products in the fast-food category were • Install and remodel facilities lunch boxes, rice balls, bread-based products, and pasta. • Pay 80 percent of utility costs As of February 2004, Seven-Eleven Japan had 290 dedi- cated manufacturing plants that produced only fast food Franchise owner responsibilities: for their stores. • Operate and manage store Other products sold at Seven-Eleven stores includ- • Hire and pay staff ed soft drinks, nutritional drinks, alcoholic beverages • Order supplies such as beer and wine, game software, music CDs, and • Maintain store appearance magazines. • Provide customer service Seven-Eleven was focused on increasing the num- Store Information and Contents ber of original items that were available only at their stores. In 2004, original items accounted for roughly 52 Seven-Eleven had 12,753 stores in Japan as of 2009 (see percent of total store sales. In 2007, Seven & i launched Table 3-2). In 2004, Seven-Eleven Japan changed the Seven Premium private brand products for sale at its standard size of new stores from 125 square meters to

Chapter 3 • Supply Chain Drivers and Metrics 63 stores. By February 2010, Seven Premium offered customers. A survey by eSBook (a joint venture among 1,035 SKUs, and this number was expected to grow in Softbank, Seven-Eleven Japan, Yahoo!Japan, and the future. Private brand products were sold across all Tohan, a publisher) discovered that 92 percent of its cus- store formats and were viewed by the company as an tomers preferred to pick up their online purchases at the important part of the expansion of synergies across its local convenience store, rather than have them delivered various retail formats. to their homes. This was understandable given the frequency with which Japanese customers visit their Store Services local convenience store; 7dream hoped to build on this preference along with the synergies from the existing Besides products, Seven-Eleven Japan gradually added a distribution system. variety of services that customers could obtain at its stores. The first service, added in October 1987, was the In March 2007, Seven-Eleven Japan introduced in-store payment of Tokyo Electric Power bills. The “Otoriyose-bin” or Internet shopping. The service company later expanded the set of utilities for which enabled customers to buy products that were typically customers could pay their bills in the stores to include not available at the retail stores. Customers were allowed gas, insurance premiums, and telephone. With more to order on the Web with both pick-up and payment at convenient operating hours and locations than banks or Seven-Eleven stores. There was no shipping fee charged other financial institutions, the bill payment service for this service. The company built Seven Net Shopping, attracted millions of additional customers every year. In its Internet site aimed at combining the group’s stores April 1994, Seven-Eleven Japan began accepting install- and Internet services. In April 2007, “nanaco” electronic ment payments on behalf of credit companies. It started money was offered in Seven-Eleven stores. The service selling ski-lift pass vouchers in November 1994. In allowed customers to prepay and use a card or cell phone 1995, it began to accept payment for mail-order purchas- to make payments. The service was offered as a conven- es. This was expanded to include payment for Internet ience to customers making small purchases and was also shopping in November 1999. In August 2000, a meal a reward system offering one yen worth of points for delivery service company, Seven-Meal Service Co. Ltd., every 100 yen spent by the customer. By the end of was established to serve the aging Japanese population. 2007, nanaco was used by customers to make more than Seven Bank was set up as the core operating company 30 million payments each month. for Seven & i in financial services. By 2009, virtually every Seven Eleven Japan store had an ATM installed Given Japan’s aging population and an increase in with Seven Bank having more than 14,000 ATMs. The the number of women working outside the home (Seven company averaged 114 transactions per ATM per day. Eleven estimated that in 2009 more than 70 percent of women in their 40s worked outside the home), Seven Other services offered at stores include photo- Eleven wanted to exploit its “close by convenient stores” copying, ticket sales (including baseball games, express to better serve its customers. The company attempted to buses, and music concerts) using multifunctional do this by offering “meal solutions” that speeded up copiers, and being a pick-up location for parcel delivery cooking at home and services like “home meal delivery.” companies that typically do not leave the parcel outside if the customer is not at home. In 2010, the convenience Seven-Eleven Japan’s Integrated Store stores also started offering some government services Information System such as providing certificates of residence. The major thrust for offering these services was to take advantage From its start, Seven-Eleven Japan sought to simplify its of the convenient locations of Seven-Eleven stores in operations by using advanced information technology. Japan. Besides providing additional revenue, the servic- Seven-Eleven Japan attributed a significant part of its es also got customers to visit the stores more frequently. success to the Total Information System installed in Several of these services exploited the existing Total every outlet and linked to headquarters, suppliers, and Information System (see text following) in the store. the Seven-Eleven distribution centers. The first online network linking the head office, stores, and vendors was In February 2000, Seven-Eleven Japan established established in 1979, though the company did not collect 7dream.com, an e-commerce company. The goal was to point-of-sales (POS) information at that time. In 1982, exploit the existing distribution system and the fact that Seven-Eleven became the first company in Japan to stores were easily accessible to most Japanese. Stores introduce a POS system comprising POS cash registers served as drop-off and collection points for Japanese and terminal control equipment. In 1985, the company

64 Chapter 3 • Supply Chain Drivers and Metrics developed, jointly with NEC, personal computers using delivery was checked. Once they were introduced, color graphics that were installed at each store and the driver simply dropped the delivery in the store, linked to the POS cash registers. These computers were and a store clerk received it at a suitable time when also on the network linking the store to the head office as there were few customers. The scanner terminals well as the vendors. Until July 1991, head office, stores, were also used when examining inventory at distribution centers, and suppliers were linked only by a stores. traditional analog network. At that time, an integrated • Store computer: This linked to the ISDN net- services digital network (ISDN) was installed. Linking work, the POS register, the graphic order terminal, more than 5,000 stores, it became one of the world’s and the scanner terminal. It communicated among largest ISDN systems at that time. the various input sources, tracked store inventory and sales, placed orders, provided detailed analy- The two-way, high-speed, online communication sis of POS data, and maintained and regulated capability of ISDN enabled Seven-Eleven Japan to col- store equipment. lect, process, and feed back POS data quickly. Sales data • POS register: To better understand the function- gathered in each store by 11:00 P.M. were processed and ing of this information network, one needs to con- ready for analysis the next morning. In 1997, Seven- sider a sampling of daily operations. As soon as a Eleven Japan introduced its fifth generation of the Total customer purchased an item and paid at the POS Information System, which was still in use in 2004. register, the item information was retrieved from the store computer and the time of sale was auto- The hardware system at a 1994 Seven-Eleven matically recorded. In addition, the cashier record- store included the following: ed the age and sex of the customer. To do this, the cashier used five register keys for the categories: • Graphic order terminal: This was a handheld under-13, 13–19, 20–29, 30–49, and 50+. This device with a wide-screen graphic display, used by POS data was automatically transmitted online the store owner or manager to place orders. The to a host computer. All sales data collected by items were recorded and brought up in the order in 11:00 P.M. were organized and ready for analysis which they were arranged on the shelves. The by the next morning. The data were evaluated on a store manager/owner walked down the aisles and company-wide, district, and store basis. placed orders by item. When placing an order, the store manager had access (from the store comput- The analyzed and updated data were then sent er) to detailed analysis of POS data related to the back to the Seven-Eleven Japan stores via the network. particular item. This included sales analysis of Each store computer automatically updated its product product categories and SKUs over time, analysis master file to analyze its recent sales and stock move- of waste, 10-week sales trends by SKU, 10-day ments. The main objective of the analysis was to sales trends by SKU, sales trends for new prod- improve the ordering process. All this information was ucts, sales analysis by day and time, list of slow- available on the graphic order terminal used for order moving items, analysis of sales and number of placement. customers over time, contribution of product to sections in store display, and sales growth by The information system allowed Seven-Eleven product categories. The store manager used this stores to better match supply with demand. Store staff information when placing an order, which was could adjust the merchandising mix on the shelves entered directly into the terminal. Once all the according to consumption patterns throughout the day. orders were placed, the terminal was returned to For example, popular breakfast items were stocked earlier its slot, at which point the orders were relayed by during the day, while popular dinner items were stocked the store computer to both the appropriate vendor later in the evening. The identification of slow and and the Seven-Eleven distribution center. nonmoving items allowed a store to convert shelf space to introduce new items. More than 50 percent of the • Scanner terminal: These scanners read bar items sold at a Seven-Eleven store changed in the course codes and recorded inventory. They were used to of a year. This was due partly to seasonal demand and receive products coming in from a distribution partly to new products. When a new product was center. This was automatically checked against a introduced, the decision whether to continue stocking it previously placed order, and the two were recon- was made within the first three weeks. Each item on the ciled. Before the scanner terminals were intro- duced, truck drivers waited in the store until the

Chapter 3 • Supply Chain Drivers and Metrics 65 shelf contributed to sales and margin and did not waste of each item was quite high. In 1974, 70 vehicles visited valuable shelf space. each store every day. By 2006, only 9 were necessary. This dramatically reduced delivery costs and enabled Seven-Eleven’s Distribution System rapid delivery of a variety of fresh foods. The Seven-Eleven distribution system tightly linked the As of February 2004, Seven-Eleven Japan had a entire supply chain for all product categories. The distri- total of 290 dedicated manufacturing plants throughout bution centers and the information network played a key the country that produced only fast food for Seven-Eleven role in that regard. The major objective was to carefully stores. These items were distributed through 293 dedicated track sales of items and offer short replenishment cycle distribution centers (DCs) that ensured rapid, reliable times. This allowed a store manager to forecast sales delivery. None of these DCs carried any inventory; they corresponding to each order accurately. merely transferred inventory from supplier trucks to Seven-Eleven distribution trucks. The transportation was From March 1987, Seven-Eleven offered three- provided by Transfleet Ltd., a company set up by Mitsui times-a-day store delivery of all rice dishes (which com- and Co. for the exclusive use of Seven-Eleven Japan. prised most of the fast-food items sold). Bread and other fresh food were delivered twice a day. The distribution 7-Eleven in the United States system was flexible enough to alter delivery schedules depending on customer demand. For example, ice cream Seven-Eleven had expanded rapidly around the world was delivered daily during the summer but only three (Table 3-5). The major growth was in Asia, although the times a week at other times. The replenishment cycle United States continued to be the second largest market time for fresh and fast-food items had been shortened for Seven-Eleven. Once Seven-Eleven Japan acquired to fewer than 12 hours. A store order for rice balls by Southland Corporation, it set about improving opera- 10:00 A.M. was delivered before the dinner rush. tions in the United States. In the initial years, several 7-Eleven stores in the United States were shut down. As discussed earlier, the store manager used a The number of stores started to grow beginning in 1998. graphic order terminal to place an order. All stores were given cutoff times for breakfast, lunch, and dinner order- Table 3-5 Global Store Distribution for ing. When a store placed an order, it was immediately Seven-Eleven in January 2011 transmitted to the supplier as well as the distribution cen- ter. The supplier received orders from all Seven-Eleven Country Stores stores and started production to fill the orders. The supplier then sent the orders by truck to the distribution center. Japan 13,049 Each store order was separated so the distribution center United States 6,726 could easily assign it to the appropriate store truck using Taiwan 4,790 the order information it already had. The key to store Thailand 5,840 delivery was what Seven-Eleven called the combined South Korea 3,150 delivery system. At the distribution center, delivery of like China 1,717 products from different suppliers (e.g., milk and sand- Malaysia 1,250 wiches) was directed into a single temperature-controlled Mexico 1,223 truck. There were four categories of temperature- Canada 465 controlled trucks: frozen foods, chilled foods, room- Australia 415 temperature processed foods, and warm foods. Each truck Singapore 549 made deliveries to multiple retail stores. The number of Philippines 567 stores per truck depended on the sales volume. All deliv- Norway 173 eries were made during off-peak hours and were received Sweden 189 using the scanner terminals. The system worked on trust Denmark 129 and did not require the delivery person to be present when Indonesia 23 the store personnel scanned in the delivery. That reduced Total the delivery time spent at each store. 40,255 This distribution system enabled Seven-Eleven to reduce the number of vehicles required for daily delivery service to each store, even though the delivery frequency

66 Chapter 3 • Supply Chain Drivers and Metrics Historically, the distribution structure in the United North American inventory turnover rate in 2004 was States was completely different from that in Japan. about 19, compared to more than 50 in Japan. This, how- Stores in the United States were replenished using direct ever, represented a significant improvement in North store delivery (DSD) by some manufacturers, with the American performance, where inventory turns in 1992 remaining products delivered by wholesalers. DSD were around 12. accounted for about half the total volume, with the rest coming from wholesalers. Study Questions With the goal of introducing “fresh” products in 1. A convenience store chain attempts to be responsive and the United States, 7-Eleven introduced the concept of provide customers with what they need, when they need it, combined distribution centers (CDCs) around 2000. By where they need it. What are some different ways that a 2003, 7-Eleven had 23 CDCs located throughout North convenience store supply chain can be responsive? What America supporting about 80 percent of the store net- are some risks in each case? work. CDCs delivered fresh items such as sandwiches, bakery products, bread, produce, and other perishables 2. Seven-Eleven’s supply chain strategy in Japan can be once a day. A variety of fresh-food suppliers sent prod- described as attempting to micro-match supply and uct to the CDC throughout the day, where they were demand using rapid replenishment. What are some risks sorted for delivery to stores at night. Requests from store associated with this choice? managers were sent to the nearest CDC, and by 10:00 P.M., the products were en route to the stores. Relative to 3. What has Seven-Eleven done in its choice of facility loca- Japan, a greater fraction of the food sold, especially hot tion, inventory management, transportation, and information food such as wings and pizza, was prepared in the store. infrastructure to develop capabilities that support its supply Fresh-food sales in North America exceeded $450 chain strategy in Japan? million in 2003. During this period, DSD by manufac- turers and wholesaler delivery to stores also continued. 4. Seven-Eleven does not allow direct store delivery in Japan but has all products flow through its distribution center. This was a period when 7-Eleven worked very hard What benefit does Seven-Eleven derive from this policy? to introduce new fresh-food items with a goal of compet- When is direct store delivery more appropriate? ing more directly with the likes of Starbucks than with traditional gas station food marts. 7-Eleven in the United 5. What do you think about the 7dream concept for Seven- States had more than 63 percent of its sales from non- Eleven Japan? From a supply chain perspective, is it gasoline products compared to the rest of the industry, for likely to be more successful in Japan or the United which this number was closer to 35 percent. The goal States? Why? was to continue to increase sales in the fresh-food and fast-food categories with a special focus on hot foods. 6. Seven-Eleven is attempting to duplicate the supply chain structure that has succeeded in Japan and the United States In 2009, revenue in the United States and Canada with the introduction of CDCs. What are the pros and cons totaled $16.0 billion, with about 63 percent coming from of this approach? Keep in mind that stores are also replen- merchandise and the rest from the sale of gasoline. The ished by wholesalers and DSD by manufacturers. 7. The United States has food service distributors that also replenish convenience stores. What are the pros and cons to having a distributor replenish convenience stores versus a company like Seven-Eleven managing its own distribu- tion function? Case Study Financial Statements for Wal-Mart Stores Inc. Table 3-6 contains the financial results for Wal-Mart for Wal-Mart perform better on? What supply chain drivers 2008 and 2009 (declared on January 31 of the following and metrics might explain this difference in performance? year). Evaluate Wal-Mart’s financial performance based on the various metrics discussed in Section 3.1, such as ROE, In 2010, Wal-Mart announced that it planned to ROA, profit margin, asset turns, APT, C2C, ART, INVT, move into urban areas in the United States by building and and PPET. Compare the metrics for Wal-Mart with simi- operating smaller format stores compared to the large lar metrics for Amazon from Table 3-1. Which metrics stores it had operated up to that point. Which supply chain does Amazon perform better on? Which metrics does metrics will be impacted by this move? How will this move impact the various financial metrics? Why?

Chapter 3 • Supply Chain Drivers and Metrics 67 Table 3-6 Selected Financial Data for Walmart Stores Inc. Year ended January 31 ($ millions) 2010 2009 408,214 404,374 Net operating revenues 304,657 304,056 Cost of goods sold 103,557 100,318 Gross profit 79,607 Selling, general, and administrative expense 23,950 77,520 Operating income 22,798 Interest expense 2,065 Other income (loss) – net 181 2,184 Income before income taxes 284 Income taxes 22,066 Other expenses 7,139 20,898 Net income 592 7,145 353 Assets 14,335 Cash and cash equivalents 13,400 Short-term investments 7,907 Net receivables - 7,275 Inventories 4,144 - Total current assets 33,160 3,905 Property, plant and equipment 48,331 34,511 Goodwill 102,307 48,949 Other assets 16,126 95,653 Total assets 3,942 15,260 170,706 3,567 Liabilities and Stockholder Equity 163,429 Accounts payable 50,550 Short-term debt 4,919 47,638 Other current liability 92 7,669 Total current liability 83 Long-term debt 55,561 Other liabilities 36,401 47,638 Total liabilities 34,549 Stockholder equity 7,688 99,650 7,808 97,747 5,257 2,672

4 {{{ Designing Distribution Networks and Applications to Online Sales LEARNING OBJECTIVES After reading this chapter, you will be able to 1. Identify the key factors to be considered when designing a distribution network. 2. Discuss the strengths and weaknesses of various distribution options. 3. Understand how online sales have affected the design of distribution networks in different industries. In this chapter, we provide an understanding of the role of distribution within a supply chain and identify factors that should be considered when designing a distribution network. We identify several potential designs for distribution networks and evaluate the strengths and weaknesses of each option. We apply these ideas to discuss the evolution of distribution networks in various industries since the advent of online sales. Our goal is to provide managers with a logical framework for selecting the appropriate distribution network given product, competitive, and market characteristics. 4.1 THE ROLE OF DISTRIBUTION IN THE SUPPLY CHAIN Distribution refers to the steps taken to move and store a product from the supplier stage to a customer stage in the supply chain. Distribution occurs between every pair of stages in the supply chain. Raw materials and components are moved from suppliers to manufacturers, whereas finished products are moved from the manufacturer to the end consumer. Distribution is a key driver of the overall profitability of a firm because it affects both the supply chain cost and the customer value directly. In the apparel retail industry, distribution impacts (including its influence on markdowns and lost sales) about 35 percent of the revenue. In India, the outbound distribution cost of cement is about 30 percent of the cost of producing and selling it. It would be no exaggeration to state that two of the world’s most profitable companies, Wal-Mart and Seven-Eleven Japan, have built the success of their entire business around outstanding distribution design and operation. In the case of Wal-Mart, distribution allows the company to provide high availability levels of relatively common products at a very low cost. In the case of Seven-Eleven Japan, effective distribution provides a very high level of customer responsiveness at a reasonable cost. The process of designing a distribution network has two broad phases. In the first phase, the broad structure of the supply chain network is visualized. This stage includes decisions such as whether the product will be sold directly or go through an intermediary. The second phase then takes the broad structure and converts it into specific locations and their capability, capacity, and demand allocation. This chapter focuses on issues that impact the design of the broad distribution network. Chapters 5 and 6 focus on the second phase that starts with the broad network and results in a specific supply chain network. 68

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 69 The appropriate distribution network can be used to achieve a variety of supply chain objectives ranging from low cost to high responsiveness. As a result, companies in the same industry often select different distribution networks. Next, we discuss industry examples that highlight the variety of distribution network choices and the issues that arise when selecting among these options. Until 2007, Dell distributed its PCs directly to end consumers, whereas companies such as HP distributed through resellers. Dell customers waited several days to get a PC, whereas customers could walk away with an HP computer from a reseller. Starting in June 2007, Dell also started selling its PCs through retailers such as Wal-Mart. In the late 1990s, Gateway opened Gateway Country stores, wherein customers could examine the products and have salespeople help them configure a PC that suited their needs. Gateway, however, chose to sell no products at the stores; all PCs were shipped directly from the factory to the customer. By April 2004, Gateway closed all its stores because of their poor financial performance. Apple Computer, in contrast, has opened many retail stores that sell computers. These PC companies have chosen different distribution models. How can we evaluate this wide range of distribution choices? Which ones serve the companies and their customers better? P&G has chosen to distribute directly to large supermarket chains while obligating smaller players to buy P&G products from distributors. Products move directly from P&G to the larger chains but move through an additional stage when going to smaller supermarkets. Texas Instruments, which once used only direct sales, now sells about 30 percent of its volume to 98 percent of its customers through distributors, while serving the remaining 2 percent of customers with 70 percent of the volume directly.1 What value do these distributors provide? When should a distribution network include an additional stage such as a distributor? Distributors play a much more significant role for consumer goods distribution in a country such as India compared to the United States. Why might this be the case? W.W. Grainger stocks about 200,000 SKUs that can be sent to customers within a day of order placement. The remaining slower moving products are not stocked but instead are shipped directly from the manufacturer when a customer places an order. It takes several days for the customer to receive the product in this case. Are these distribution choices appropriate? How can they be justified? As the preceding examples illustrate, firms can make many choices when designing their distribution network. An inappropriate network can have a significant negative effect on the profitability of the firm, as is evident in the failure of companies such as Blockbuster and Webvan. The appropriate choice of distribution network grows the supply chain surplus by satisfying customer needs at the lowest possible cost. In the next section, we identify performance measures that need to be considered when designing the distribution network. 4.2 FACTORS INFLUENCING DISTRIBUTION NETWORK DESIGN At the highest level, performance of a distribution network should be evaluated along two dimensions: 1. Customer needs that are met 2. Cost of meeting customer needs Thus, a firm must evaluate the impact on customer service and cost as it compares different distribution network options. The customer needs that are met influence the company’s revenues, which along with cost decide the profitability of the delivery network. Although customer value is impacted by many factors, we focus on those measures that are influenced by the structure of the distribution network: • Response time • Product variety • Product availability 1 A Tale of Two Electronic Component Distributors, Ananth Raman and Bharat P. Rao, Harvard Business School Case 9–697–064, 1997.

70 Chapter 4 • Designing Distribution Networks and Applications to Online Sales • Customer experience • Time to market • Order visibility • Returnability Response time is the amount of time it takes for a customer to receive an order. Product vari- ety is the number of different products/configurations that are offered by the distribution network. Product availability is the probability of having a product in stock when a customer order arrives. Customer experience includes the ease with which customers can place and receive orders and the extent to which this experience is customized. It also includes purely experiential aspects, such as the possibility of getting a cup of coffee and the value that the sales staff provides. Time to market is the time it takes to bring a new product to the market. Order visibility is the ability of customers to track their orders from placement to delivery. Returnability is the ease with which a customer can return unsatisfactory merchandise and the ability of the network to handle such returns. It may seem at first that a customer always wants the highest level of performance along all these dimensions. In practice, however, this is not the case. Customers ordering a book at Amazon are willing to wait longer than those who drive to a nearby Barnes & Noble store to get the same book. In contrast, customers can find a much larger variety of books at Amazon compared to the Barnes & Noble store. Thus, Amazon customers trade off fast response times for high levels of variety. Firms that target customers who can tolerate a long response time require only a few locations that may be far from the customer. These companies can focus on increasing the capacity of each location. In contrast, firms that target customers who value short response times need to locate facilities close to them. These firms must have many facilities, each with a low capacity. Thus, a decrease in the response time customers desire increases the number of facilities required in the network, as shown in Figure 4-1. For example, Barnes & Noble provides its customers with books on the same day but requires hundreds of stores to achieve this goal for most of the United States. Amazon, in contrast, takes a few days to deliver a book to its U.S. customers, but it uses about 20 locations to store its books. Changing the distribution network design affects the following supply chain costs (notice that these are four of the six supply chain drivers we discussed earlier): • Inventories • Transportation • Facilities and handling • Information The other two drivers, sourcing and pricing, also affect the choice of the distribution system; the link will be discussed when relevant. As the number of facilities in a supply chain increases, the inventory and resulting inventory costs also increase (see Chapter 12), as shown in Figure 4-2. Required Number of Facilities Desired FIGURE 4-1 Relationship Response Between Desired Response Time Time and Number of Facilities

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 71 Inventory Costs Number of Facilities FIGURE 4-2 Relationship Between Number of Facilities and Inventory Costs To decrease inventory costs, firms try to consolidate and limit the number of facilities in their supply chain network. For example, with fewer facilities, Amazon is able to turn its inventory about 10 times a year, whereas Barnes & Noble, with hundreds of facilities, achieves only about 3 turns per year. Inbound transportation costs are the costs incurred in bringing material into a facility. Outbound transportation costs are the costs of sending material out of a facility. Outbound transportation costs per unit tend to be higher than inbound costs because inbound lot sizes are typically larger. For example, an Amazon warehouse receives full truckload shipments of books on the inbound side, but ships out small packages with only a few books per customer on the outbound side. Increasing the number of warehouse locations decreases the average outbound distance to the customer and makes outbound transportation distance a smaller fraction of the total distance traveled by the product. Thus, as long as inbound transportation economies of scale are maintained, increasing the number of facilities decreases total trans- portation cost, as shown in Figure 4-3. If the number of facilities is increased to a point where inbound lot sizes are also very small and result in a significant loss of economies of scale in inbound transportation, increasing the number of facilities increases total transportation cost, as shown in Figure 4-3. Transportation Cost Number of Facilities FIGURE 4-3 Relationship Between Number of Facilities and Transportation Cost

72 Chapter 4 • Designing Distribution Networks and Applications to Online Sales Facility Costs Number of Facilities FIGURE 4-4 Relationship Between Number of Facilities and Facility Costs Facility costs decrease as the number of facilities is reduced, as shown in Figure 4-4, because a consolidation of facilities allows a firm to exploit economies of scale. While Amazon achieved a property, plant, and equipment turnover (PPET) of 19 in 2009, Barnes & Noble had a PPET of just over 7. Total logistics costs are the sum of inventory, transportation, and facility costs for a supply chain network. As the number of facilities increases, total logistics costs first decrease and then increase as shown in Figure 4-5. Each firm should have at least the number of facilities that minimizes total logistics costs. Amazon has more than one warehouse primarily to reduce its logistics costs (and improve response time). If a firm wants to reduce the response time to its customers further, it may have to increase the number of facilities beyond the point that minimizes logistics costs. A firm should add facilities beyond the cost-minimizing point only if managers are confident that the increase in revenues because of better responsiveness is greater than the increase in costs because of the additional facilities. Response Time Total Logistics Cost Number of Facilities FIGURE 4-5 Variation in Logistics Cost and Response Time with Number of Facilities

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 73 The customer service and cost components listed earlier are the primary measures used to evaluate different delivery network designs. In general, no distribution network will outperform others along all dimensions. Thus, it is important to ensure that the strengths of the distribution network fit with the strategic position of the firm. In the next section, we discuss various distribution networks and their relative strengths and weaknesses. 4.3 DESIGN OPTIONS FOR A DISTRIBUTION NETWORK In this section, we discuss distribution network choices from the manufacturer to the end consumer. When considering distribution between any other pair of stages, such as supplier to manufacturer or even a service company serving its customers through a distribution network, many of the same options still apply. Managers must make two key decisions when designing a distribution network: 1. Will product be delivered to the customer location or picked up from a prearranged site? 2. Will product flow through an intermediary (or intermediate location)? Based on the firm’s industry and the answers to these two questions, one of six distinct distribution network designs may be used to move products from factory to customer. These designs are classified as follows: 1. Manufacturer storage with direct shipping 2. Manufacturer storage with direct shipping and in-transit merge 3. Distributor storage with carrier delivery 4. Distributor storage with last-mile delivery 5. Manufacturer/distributor storage with customer pickup 6. Retail storage with customer pickup Manufacturer Storage with Direct Shipping In this option, product is shipped directly from the manufacturer to the end customer, bypassing the retailer (who takes the order and initiates the delivery request). This option is also referred to as drop-shipping. The retailer carries no inventory. Information flows from the customer, via the retailer, to the manufacturer, and product is shipped directly from the manufacturer to customers as shown in Figure 4-6. Online retailers such as eBags and Nordstrom.com use drop-shipping to deliver goods to the end consumer. eBags holds few bags in inventory. Nordstrom carries some products in inventory and uses the drop-ship model for slow-moving footwear. W.W. Grainger also uses drop-shipping to deliver slow-moving items to customers. The biggest advantage of drop-shipping is the ability to centralize inventories at the manufac- turer who can aggregate demand across all retailers that it supplies. As a result, the supply chain is Manufacturers Retailer Customers Product Flow Information Flow FIGURE 4-6 Manufacturer Storage with Direct Shipping

74 Chapter 4 • Designing Distribution Networks and Applications to Online Sales able to provide a high level of product availability with lower levels of inventory. A key issue with regard to drop-shipping is the ownership structure of the inventory at the manufacturer. If specified portions of inventory at the manufacturer are allocated to individual retailers, there is little benefit of aggregation even though the inventory is physically aggregated. Benefit of aggregation is achieved only if the manufacturer can allocate at least a portion of the available inventory across retailers on an as-needed basis. The benefits from centralization are highest for high-value, low- demand items with unpredictable demand. The decision of Nordstrom to drop-ship low-demand shoes satisfies these criteria. Similarly, bags sold by eBags tend to have high value and relatively low demand per SKU. The inventory benefits of aggregation are small for items with predictable demand and low value. Thus, drop-shipping does not offer a significant inventory advantage to an online grocer selling a staple item such as detergent. For slow-moving items, inventory turns can increase by a factor of six or higher if drop-shipping is used instead of storage at retail stores. Drop-shipping also offers the manufacturer the opportunity to postpone customization until after a customer has placed an order. Postponement, if implemented, further lowers inventories by aggregating to the component level. For its customized machines, Dell holds inventories as common components and postpones product customization, thus lowering the level of inventory carried. Although inventory costs are typically low with drop-shipping, transportation costs are high because the average outbound distance to the end consumer is large, and package carriers are used to ship the product. Package carriers have high shipping costs per unit compared to truckload or less-than-truckload carriers. With drop-shipping, a customer order including items from several manufacturers will involve multiple shipments to the customer. This loss in aggregation of outbound transportation increases cost. Supply chains save on the fixed cost of facilities when using drop-shipping because all inventories are centralized at the manufacturer. This eliminates the need for other warehousing space in the supply chain. There can be some savings of handling costs as well, because the transfer from manufacturer to retailer no longer occurs. Handling cost savings must be evaluated carefully, however, because the manufacturer is now required to transfer items to the factory warehouse in full cases and then ship out from the warehouse in single units. The inability of a manufacturer to develop single-unit delivery capabilities can have a significant negative effect on handling cost and response time. Handling costs can be reduced significantly if the manufacturer has the capability to ship orders directly from the production line. A good information infrastructure is needed between the retailers and the manufacturer so that the retailer can provide product availability information to the customer, even though the inventory is located at the manufacturer. The customer should also have visibility into order processing at the manufacturer, even with the order being placed with the retailer. Drop-shipping generally requires significant investment in information infrastructure. The information infrastructure requirement is somewhat simpler for direct sellers like Dell because two stages (retailer and manufacturer) do not need to be integrated. Response times tend to be long when drop-shipping is used because the order has to be transmitted from the retailer to the manufacturer and shipping distances are generally longer from the manufacturer’s centralized site. eBags, for example, states that order processing may take from 1 to 5 days and ground transportation after that may take from 3 to 11 business days. This implies that customer response time at eBags will be 4 to 16 days using ground transportation and drop-shipping. Another issue is that the response time need not be identical for every manufacturer that is part of a customer order. Given an order containing products from several sources, the customer will receive multiple partial shipments over time, making receiving more complicated for the customer. Manufacturer storage allows a high level of product variety to be available to the customer. With a drop-shipping model, every product at the manufacturer can be made available to the customer without any limits imposed by shelf space. W.W. Grainger is able to offer hundreds of thousands of slow-moving items from thousands of manufacturers using drop-shipping.

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 75 This would be impossible if each product had to be stored by W.W. Grainger. Drop-shipping allows a new product to be available to the market the day the first unit is produced. Drop-shipping provides a good customer experience in the form of delivery to the customer location. The experience, however, suffers when a single order containing products from several manufacturers is delivered in partial shipments. Order visibility is important in the context of manufacturer storage, because two stages in the supply chain are involved in every customer order. Failure to provide this capability is likely to have a significant negative effect on customer satisfaction. Order tracking, however, becomes harder to implement in a drop-ship system because it requires complete integration of information systems at both the retailer and the manufacturer. For direct sellers such as Dell, order visibility is simpler to provide. A manufacturer storage network is likely to have difficulty handling returns, hurting customer satisfaction. The handling of returns is more expensive under drop-shipping because each order may involve shipments from more than one manufacturer. Returns can be handled in two ways. One is for the customer to return the product directly to the manufacturer. The second approach is for the retailer to set up a separate facility (across all manufacturers) to handle returns. The first approach incurs high transportation and coordination costs, whereas the second approach requires investment in a facility to handle returns. The performance characteristics of drop-shipping along various dimensions are summarized in Table 4-1. Table 4-1 Performance Characteristics of Manufacturer Storage with Direct Shipping Network Cost Factor Performance Inventory Lower costs because of aggregation. Benefits of aggregation are highest for low-demand, high-value items. Benefits are large Transportation if product customization can be postponed at the manufacturer. Facilities and handling Higher transportation costs because of increased distance and Information disaggregate shipping. Service Factor Lower facility costs because of aggregation. Some saving on handling costs if manufacturer can manage small shipments or ship from production line. Significant investment in information infrastructure to integrate manufacturer and retailer. Performance Response time Long response time of one to two weeks because of increased distance and two stages for order processing. Response time may Product variety vary by product, thus complicating receiving. Product availability Customer experience Easy to provide a high level of variety. Time to market Order visibility Easy to provide a high level of product availability because of Returnability aggregation at manufacturer. Good in terms of home delivery but can suffer if order from several manufacturers is sent as partial shipments. Fast, with the product available as soon as the first unit is produced. More difficult but also more important from a customer service perspective. Expensive and difficult to implement.

76 Chapter 4 • Designing Distribution Networks and Applications to Online Sales Given its performance characteristics, manufacturer storage with direct shipping is best suited for a large variety of low-demand, high-value items for which customers are willing to wait for delivery and accept several partial shipments. Manufacturer storage is also suitable if it allows the manufacturer to postpone customization, thus reducing inventories. It is thus ideal for direct sellers that are able to build-to-order. For drop-shipping to be effective, there should be few sourcing locations per order. Manufacturer Storage with Direct Shipping and In-Transit Merge Unlike pure drop-shipping, under which each product in the order is sent directly from its manufacturer to the end customer, in-transit merge combines pieces of the order coming from different locations so that the customer gets a single delivery. Information and product flows for the in-transit merge network are shown in Figure 4-7. In-transit merge has been used by Dell and can be used by companies implementing drop-shipping. When a customer orders a PC from Dell along with a Sony monitor, the package carrier picks up the PC from the Dell factory and the monitor from the Sony factory; it then merges the two at a hub before making a single delivery to the customer. As with drop-shipping, the ability to aggregate inventories and postpone product customization is a significant advantage of in-transit merge. In-transit merge allows Dell and Sony to hold all their inventories at the factory. This approach has the greatest benefits for products with high value whose demand is difficult to forecast, particularly if product customization can be postponed. Although an increase in coordination is required, merge in transit decreases transportation costs relative to drop-shipping by aggregating the final delivery. Facility and processing costs for the manufacturer and the retailer are similar to those for drop-shipping. The party performing the in-transit merge has higher facility costs because of the merge capability required. Receiving costs at the customer are lower because a single delivery is received. Overall supply chain facility and handling costs are somewhat higher than with drop-shipping. A sophisticated information infrastructure is needed to allow in-transit merge. In addition to information, operations at the retailer, manufacturers, and the carrier must be coordinated. The investment in information infrastructure is higher than that for drop-shipping. Response times, product variety, availability, and time to market are similar to drop- shipping. Response times may be marginally higher because of the need to perform the merge. Customer experience is likely to be better than with drop-shipping, because the customer receives only one delivery for an order instead of many partial shipments. Order visibility is an important requirement. Although the initial setup is difficult because it requires integration of manufacturer, carrier, and retailer, tracking itself becomes easier given the merge that occurs at the carrier hub. Factories Retailer In-Transit Merge by Carrier Customers Product Flow Information Flow FIGURE 4-7 In-Transit Merge Network

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 77 Table 4-2 Performance Characteristics of In-Transit Merge Cost Factor Performance Inventory Similar to drop-shipping. Transportation Somewhat lower transportation costs than drop-shipping. Facilities and handling Handling costs higher than drop-shipping at carrier; receiving costs lower at customer. Information Investment is somewhat higher than for drop-shipping. Service Factor Performance Response time Similar to drop-shipping; may be marginally higher. Product variety Similar to drop-shipping. Product availability Similar to drop-shipping. Customer experience Better than drop-shipping because only a single delivery has to be received. Time to market Similar to drop-shipping Order visibility Similar to drop-shipping. Returnability Similar to drop-shipping. Returnability is similar to that with drop-shipping. Problems in handling returns are likely, and the reverse supply chain will continue to be expensive and difficult to implement, as with drop-shipping. The performance of factory storage with in-transit merge is compared with that of drop-shipping in Table 4-2. The main advantages of in-transit merge over drop-shipping are lower transportation cost and improved customer experience. The major disadvantage is the additional effort during the merge itself. Given its performance characteristics, manufacturer storage with in-transit merge is best suited for low- to medium-demand, high-value items the retailer is sourcing from a limited number of manufacturers. Compared to drop-shipping, in-transit merge requires a higher demand from each manufacturer (not necessarily each product) in order to be effective. When there are too many sources, in-transit merge can be difficult to coordinate and implement. In-transit merge is best implemented if there are no more than four or five sourcing locations. The in-transit merge of a Dell PC with a Sony monitor is appropriate because product variety is high, but there are few sourcing locations with relatively large total demand from each sourcing location. Distributor Storage with Carrier Delivery Under this option, inventory is not held by manufacturers at the factories but is held by distributors/retailers in intermediate warehouses, and package carriers are used to transport products from the intermediate location to the final customer. Amazon and industrial distributors such as W.W. Grainger and McMaster-Carr have used this approach combined with drop-shipping from a manufacturer (or distributor). Information and product flows when using distributor storage with delivery by a package carrier are shown in Figure 4-8. Relative to manufacturer storage, distributor storage requires a higher level of inventory because of a loss of aggregation. From an inventory perspective, distributor storage makes sense for products with somewhat higher demand. This is seen in the operations of both Amazon and W.W. Grainger. They stock only the slow- to fast-moving items at their warehouses, with very slow-moving items stocked farther upstream. In some instances, postponement of product differentiation can be implemented with distributor storage, but it does require that the warehouse develop some assembly capability. Distributor storage, however, requires much less

78 Chapter 4 • Designing Distribution Networks and Applications to Online Sales Factories Warehouse Storage by Distributor/ Retailer Customers Product Flow Information Flow FIGURE 4-8 Distributor Storage with Carrier Delivery inventory than a retail network. Amazon achieves about 10 turns of inventory annually using warehouse storage, whereas Barnes & Noble achieves about 3 turns using retail stores. Transportation costs are somewhat lower for distributor storage compared to manufacturer storage because an economic mode of transportation (e.g., truckloads) can be employed for inbound shipments to the warehouse, which is closer to the customer. Unlike manufacturer storage, under which multiple shipments may need to go out for a single customer order with multiple items, distributor storage allows outbound orders to the customer to be bundled into a single shipment, further reducing transportation cost. Distributor storage provides savings on the transportation of faster moving items relative to manufacturer storage. Compared to manufacturer storage, facility costs (of warehousing) are somewhat higher with distributor storage because of a loss of aggregation. Processing and handling costs are comparable to manufacturer storage unless the factory is able to ship to the end customer directly from the production line. In that case, distributor storage has higher processing costs. From a facility cost perspective, distributor storage is not appropriate for extremely slow-moving items. The information infrastructure needed with distributor storage is significantly less complex than that needed with manufacturer storage. The distributor warehouse serves as a buffer between the customer and the manufacturer, decreasing the need to coordinate the two completely. Real-time visibility between customers and the warehouse is needed, whereas real-time visibility between the customer and the manufacturer is not. Visibility between the distributor warehouse and manufacturer can be achieved at a much lower cost than real-time visibility between the customer and manufacturer. Response time under distributor storage is better than under manufacturer storage because distributor warehouses are, on average, closer to customers, and the entire order is aggregated at the warehouse before being shipped. Amazon, for example, processes most warehouse-stored items within a day and then it takes three to five business days using ground transportation for the order to reach the customer. W.W. Grainger processes customer orders on the same day and has enough warehouses to deliver most orders the next day using ground transport. Warehouse storage limits to some extent the variety of products that can be offered. W.W. Grainger does not store very low-demand items at its warehouse, relying on manufacturers to drop-ship those products to the customer. Customer convenience is high with distributor storage because a single shipment reaches the customer in response to an order. Time to market under distributor storage is somewhat higher than under manufacturer storage because of the need to stock another stage in the supply chain. Order visibility becomes easier than with manufacturer storage because there is a single shipment from the warehouse to the customer and only one stage of the supply chain is involved directly in filling the customer order. Returnability is better than with manufacturer storage because all returns can be processed at the warehouse itself. The customer also has to return only one package, even if the items are from several manufacturers. The performance of distributor storage with carrier delivery is summarized in Table 4-3. Distributor storage with carrier delivery is well suited for slow- to fast-moving items.

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 79 Table 4-3 Performance Characteristics of Distributor Storage with Carrier Delivery Cost Factor Performance Inventory Higher than manufacturer storage. Difference is not large for faster moving items but can be large for very slow-moving items. Transportation Lower than manufacturer storage. Reduction is highest for faster moving items. Facilities and handling Somewhat higher than manufacturer storage. The difference can be large for very slow-moving items. Information Simpler infrastructure compared to manufacturer storage. Service Factor Response time Performance Product variety Product availability Faster than manufacturer storage. Lower than manufacturer storage. Customer experience Higher cost to provide the same level of availability as manufacturer Time to market storage. Order visibility Better than manufacturer storage with drop-shipping. Returnability Higher than manufacturer storage. Easier than manufacturer storage. Easier than manufacturer storage. Distributor storage also makes sense when customers want delivery faster than is offered by manufacturer storage but do not need it immediately. Distributor storage can handle somewhat lower variety than manufacturer storage but can handle a much higher level of variety than a chain of retail stores. Distributor Storage with Last-Mile Delivery Last-mile delivery refers to the distributor/retailer delivering the product to the customer’s home instead of using a package carrier. Webvan, Peapod, and Albertsons have used last-mile delivery in the grocery industry. Amazon has launched “local express delivery” to provide same-day delivery to customers. Companies such as Kozmo and Urbanfetch tried to set up home-delivery networks for a variety of products but failed to survive. The automotive spare parts industry is one in which distributor storage with last-mile delivery is the dominant model. It is too expensive for dealers to carry all spare parts in inventory. Thus, original equipment manufacturers (OEMs) tend to carry most spare parts at a local distribution center typically located no more than a couple of hours’ drive from their dealers and often managed by a third party. The local distribution center is responsible for delivering needed parts to a set of dealers and makes multiple deliveries per day. Unlike package carrier delivery, last-mile delivery requires the distributor warehouse to be much closer to the customer. Given the limited radius that can be served with last-mile delivery, more warehouses are required compared to when package delivery is used. The warehouse storage with last-mile delivery network is as shown in Figure 4-9. Distributor storage with last-mile delivery requires higher levels of inventory than the other options (except for retail stores) because it has a lower level of aggregation. From an inventory perspective, warehouse storage with last-mile delivery is suitable for relatively fast-moving items that are needed quickly and for which some level of aggregation is beneficial. Auto parts required by car dealers fit this description. Among all the distribution networks, transportation costs are highest for last-mile delivery, especially when delivering to individuals. This is because package carriers aggregate delivery across many retailers and are able to obtain better economies of scale than are available to a distributor/retailer attempting last-mile delivery. Delivery costs

80 Chapter 4 • Designing Distribution Networks and Applications to Online Sales Factories Distributor/Retailer Warehouse Customers Product Flow Information Flow FIGURE 4-9 Distributor Storage with Last-Mile Delivery (including transportation and processing) can be more than $20 per home delivery in the grocery industry. Last-mile delivery may be somewhat less expensive in large, dense cities. Transportation costs may also be justifiable for bulky products for which the customer is willing to pay for home delivery. Home delivery of water and large bags of rice has proved quite successful in China, where the high population density has helped decrease delivery costs. Transportation costs of last-mile delivery are best justified in settings where the customer is purchasing in large quantities. This is rare for individual customers, but businesses such as auto dealerships purchase large quantities of spare parts on a daily basis and can thus justify daily delivery. Home delivery to individual customers can be justified for bulky items such as five-gallon jugs of water in the United States and large bags of rice in China. In each instance, last-mile delivery is cheaper and more convenient than customers picking up their own bottles or bags. Using this option, facility costs are somewhat lower than for a network with retail stores but much higher than for either manufacturer storage or distributor storage with package carrier delivery. Processing costs, however, are much higher than for a network of retail stores because all customer participation is eliminated. A grocery store using last-mile delivery performs all the processing until the product is delivered to the customer’s home, unlike a supermarket, where the customer does a lot more work. The information infrastructure with last-mile delivery is similar to that for distributor storage with package carrier delivery. However, it requires the additional capability of scheduling deliveries. Response times are faster than using package carriers. Kozmo and Urbanfetch tried to provide same-day delivery, whereas online grocers typically provide next-day delivery. Product variety is generally lower than for distributor storage with carrier delivery. The cost of providing product availability is higher than for every option other than retail stores. The customer experience can be good using this option, particularly for bulky, hard-to-carry items. Time to market is even higher than for distributor storage with package carrier delivery because the new product has to penetrate deeper before it is available to the customer. Order visibility is less of an issue given that deliveries are made within 24 hours. The order-tracking feature does become important to handle exceptions in case of incomplete or undelivered orders. Of all the options discussed, returnability is best with last-mile delivery, because trucks making deliveries can also pick up returns from customers. Returns are still more expensive to handle than at a retail store, where a customer can bring the product back. The performance characteristics of distributor storage with last-mile delivery are summarized in Table 4-4. In areas with high labor costs, it is hard to justify last-mile delivery to individual consumers on the basis of efficiency or improved margin. It can be justified only if there is a large enough customer segment willing to pay for this convenience. In that case, an effort should be made to couple last-mile delivery with an existing distribution network to exploit economies of scale and improve utilization. An example is Albertsons’ use of existing grocery

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 81 Table 4-4 Performance Characteristics of Distributor Storage with Last-Mile Delivery Cost Factor Performance Inventory Higher than distributor storage with package carrier delivery. Transportation Very high cost given minimal scale economies. Higher than any other Facilities and handling distribution option. Information Facility costs higher than manufacturer storage or distributor storage with package carrier delivery, but lower than a chain of retail stores. Similar to distributor storage with package carrier delivery. Service Factor Performance Response time Very quick. Same day to next-day delivery. Product variety Somewhat less than distributor storage with package carrier delivery Product availability but larger than retail stores. Customer experience More expensive to provide availability than any other option except Time to market retail stores. Order visibility Very good, particularly for bulky items. Returnability Slightly higher than distributor storage with package carrier delivery. Less of an issue and easier to implement than manufacturer storage or distributor storage with package carrier delivery. Easier to implement than other previous options. Harder and more expensive than a retail network. store facilities and labor to provide home delivery. A portion of the grocery store serves as a fulfillment center for online orders and as a replenishment center for the grocery store itself. This helps improve utilization and lower the cost of providing this service. Last-mile delivery may be justifiable if customer orders are large enough to provide some economies of scale, and customers are willing to pay for this convenience. Peapod has changed its pricing policies to reflect this idea. Minimum order size is $60 (with a delivery charge of $9.95), and delivery charges drop to $6.95 for orders totaling more than $100. Peapod offers discounts for deliveries during slower periods based on what its schedule looks like. Last-mile delivery is easier to justify when the customer is a business like an auto dealer purchasing large quantities. Manufacturer or Distributor Storage with Customer Pickup In this approach, inventory is stored at the manufacturer or distributor warehouse, but customers place their orders online or on the phone and then travel to designated pickup points to collect their merchandise. Orders are shipped from the storage site to the pickup points as needed. Examples include 7dream.com and Otoriyose-bin, operated by Seven-Eleven Japan, which allow customers to pick up online orders at a designated store. A business-to-business (B2B) example is W.W. Grainger, whose customers can pick up their orders at one of the W.W. Grainger retail outlets. Some items are stored at the pickup location, whereas others may come from a central location. In the case of 7dream.com, the order is delivered from a manufacturer or distributor warehouse to the pickup location. In 2007, Wal-Mart launched its “Site to Store” service that allows customers to order thousands of products online at Walmart.com and have them shipped free to a local Wal-Mart store. Items arrive in stores 7 to 10 business days after the order is processed, and customers receive an e-mail notification when their order is ready for pickup. The information and product flows shown in Figure 4-10 are similar to those in the Seven-Eleven Japan network. Seven-Eleven has distribution centers where product from

82 Chapter 4 • Designing Distribution Networks and Applications to Online Sales Factories Retailer Cross-Dock DC Pickup Sites Customers Customer Flow Product Flow Information Flow FIGURE 4-10 Manufacturer or Distributor Warehouse Storage with Consumer Pickup manufacturers is cross-docked and sent to retail outlets on a daily basis. An online retailer delivering an order through Seven-Eleven can be treated as one of the manufacturers, with deliveries cross-docked and sent to the appropriate Seven-Eleven outlet. Serving as an outlet for online orders allows Seven-Eleven to improve utilization of its existing logistical assets. Inventory costs using this approach can be kept low, with either manufacturer or distributor storage to exploit aggregation. W.W. Grainger keeps its inventory of fast-moving items at pickup locations, whereas slow-moving items are stocked at a central warehouse or in some cases at the manufacturer. Transportation cost is lower than for any solution using package carriers because significant aggregation is possible when delivering orders to a pickup site. This allows the use of truckload or less-than-truckload carriers to transport orders to the pickup site. For a company such as Seven-Eleven Japan, the marginal increase in transportation cost is small because trucks are already making deliveries to the stores, and their utilization can be improved by including online orders. As a result, Seven-Eleven Japan allows customers to pick up orders without a shipping fee. Facility costs are high if new pickup sites have to be built. A solution using existing sites can lower the additional facility costs. This, for example, is the case with 7dream.com, Wal-Mart, and W.W. Grainger, for which the stores already exist. Processing costs at the manufacturer or the warehouse are comparable to those of other solutions. Processing costs at the pickup site are high because each order must be matched with a specific customer when he or she arrives. Creating this capability can increase processing costs significantly if appropriate storage and information systems are not provided. Increased processing cost and potential errors at the pickup site are the biggest hurdle to the success of this approach. A significant information infrastructure is needed to provide visibility of the order until the customer picks it up. Good coordination is needed among the retailer, the storage location, and the pickup location. In this case, a response time comparable to that using package carriers can be achieved. Variety and availability comparable to any manufacturer or distributor storage option can be provided. There is some loss of customer experience, because unlike the other options discussed, customers must pick up their own orders. On the other hand, customers who do not want to pay online can pay by cash using this option. In countries like Japan, where Seven-Eleven has more than 10,000 outlets, it can be argued that the loss of customer convenience is small, because most customers are close to a pickup site and can collect an order at their convenience. In some cases, this option is considered more convenient because it does

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 83 Table 4-5 Performance Characteristics of Network with Consumer Pickup Sites Cost Factor Performance Inventory Can match any other option, depending on the location of inventory. Transportation Facilities and handling Lower than the use of package carriers, especially if using an existing delivery network. Information Service Factor Facility costs can be high if new facilities have to be built. Costs are lower if existing facilities are used. The increase in handling cost at the pickup site can be significant. Significant investment in infrastructure required Performance Response time Similar to package carrier delivery with manufacturer or distributor storage. Same-day delivery possible for items stored locally at pickup site. Product variety Similar to other manufacturer or distributor storage options. Product availability Similar to other manufacturer or distributor storage options. Customer experience Lower than other options because of the lack of home delivery. Experience is sensitive to capability of pickup location. Time to market Similar to manufacturer storage options. Order visibility Difficult but essential. Returnability Somewhat easier given that pickup location can handle returns. not require the customer to be at home at the time of delivery. Time to market for new products can be as short as with manufacturer storage. Order visibility is extremely important for customer pickups. The customer must be informed when the order has arrived, and the order should be easily identified once the customer arrives to pick it up. Such a system is hard to implement because it requires integration of several stages in the supply chain. Returns can potentially be handled at the pickup site, making it easier for customers. From a transportation perspective, return flows can be handled using the delivery trucks. The performance characteristics of manufacturer or distributor storage with consumer pickup sites are summarized in Table 4-5. The main advantages of a network with consumer pickup sites are that it can lower the delivery cost and expand the set of products sold and customers served online. The major hurdle is the increased handling cost and complexity at the pickup site. Such a network is likely to be most effective if existing retail locations are used as pickup sites, because this type of network improves the economies from existing infrastructure. In particular, such a network can be effective for firms like Seven-Eleven Japan, Wal-Mart, and W.W. Grainger, which have both a network of stores and an online business. Unfortunately, such retail sites are typically designed to allow the customer to do the picking and need to develop the capability of picking a customer-specific order. Retail Storage with Customer Pickup In this option, often viewed as the most traditional type of supply chain, inventory is stored locally at retail stores. Customers walk into the retail store or place an order online or by phone and pick it up at the retail store. Examples of companies that offer multiple options of order placement include Albertsons, which uses part of the facility as a grocery store and part as an online fulfillment center. Customers can walk into the store or order online. A B2B example is

84 Chapter 4 • Designing Distribution Networks and Applications to Online Sales W.W. Grainger: Customers can order online, by phone, or in person and pick up their order at one of W.W. Grainger’s retail outlets. Albertsons keeps its inventory at the pickup location itself; W.W. Grainger stores some items at the pickup locations, whereas others may come from a central location. Local storage increases inventory costs because of the lack of aggregation. For fast- to very fast-moving items, however, there is marginal increase in inventory even with local storage. Albertsons uses local storage because most of its products are relatively fast moving and are stocked at the supermarket in any case. Similarly, W.W. Grainger keeps its inventory of fast-moving items at pickup locations, whereas slow-moving items are stocked at a central warehouse. Transportation cost is much lower than with other solutions because inexpensive modes of transport can be used to replenish product at the retail store. Facility costs are high because many local facilities are required. A minimal information infrastructure is needed if customers walk into the store and place orders. For online orders, however, a significant information infrastructure is needed to provide visibility of the order until the customer picks it up. Good response times can be achieved with this system because of local storage. For example, both Albertsons and W.W. Grainger offer same-day pickup from their retail locations. Product variety stored locally is lower than under other options. It is more expensive than with all other options to provide a high level of product availability. Customer experience depends on whether or not the customer likes to shop. Time to market is the highest with this option because the new product has to penetrate through the entire supply chain before it is available to customers. Order visibility is extremely important for customer pickups when orders are placed online or by phone. Returns can be handled at the pickup site. Overall, returnability is fairly good using this option. The performance characteristics of a network with customer pickup sites and local retail storage are summarized in Table 4-6. The main advantage of a network with retail storage is that it can lower delivery costs and provide a faster response than other networks. The major disadvantage is the increased inventory and facility costs. Such a network is best suited for fast-moving items or items for which customers value rapid response. Table 4-6 Performance Characteristics of Retail Storage at Consumer Pickup Sites Cost Factor Performance Inventory Higher than all other options. Transportation Lower than all other options. Facilities and Higher than other options. The increase in handling cost at the pickup handling site can be significant for online and phone orders. Information Some investment in infrastructure required for online and phone orders. Service Factor Performance Response time Same-day (immediate) pickup possible for items stored locally at pickup site. Product variety Lower than all other options. Product availability More expensive to provide than all other options. Customer Related to whether shopping is viewed as a positive or negative experience experience by customer. Time to market Highest among distribution options. Order visibility Trivial for in-store orders. Difficult, but essential, for online and phone orders. Returnability Easier than other options because retail store can provide a substitute.

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 85 Selecting a Distribution Network Design A network designer needs to consider product characteristics as well as network requirements when deciding on the appropriate delivery network. The various networks considered earlier have different strengths and weaknesses. In Table 4-7, the various delivery networks are ranked relative to one another along different performance dimensions. A ranking of 1 indicates the best performance along a given dimension; as the relative performance worsens, the ranking number increases. Only niche companies end up using a single distribution network. Most companies are best served by a combination of delivery networks. The combination used depends on product characteristics and the strategic position that the firm is targeting. The suitability of different delivery designs (from a supply chain perspective) in various situations is shown in Table 4-8. An excellent example of a hybrid network is that of W.W. Grainger, which combines all the aforementioned options in its distribution network. The network, however, is tailored to match the characteristics of the product and the needs of the customer. Fast-moving and emergency items are stocked locally, and customers can either pick them up or have them shipped, depending on the urgency. Slower moving items are stocked at a national DC and shipped to the customer within a day or two. Very slow-moving items are typically drop-shipped from the manufacturer and carry a longer lead time. Another hybrid network is used by Amazon, which stocks fast-moving items at most of its warehouses, slower moving items at fewer warehouses, and very slow-moving items may be drop-shipped from distributors or publishers. We can now revisit the examples from the computer industry discussed at the beginning of the chapter. Gateway’s decision to create a network of retail stores without exploiting any of the supply chain advantages such a network offers was flawed. To fully exploit the benefits of the retail network, Gateway should have stocked its standard configurations (likely to have high demand) at the retail stores, with all other configurations drop-shipped from the factory (perhaps with local pickup at the retail stores if that was economical). Instead, it drop-shipped all configurations from the factory. Apple has opened several retail stores and actually carries products for sale at these stores. This makes sense given the low variety and high demand for Apple products. In fact, Apple has seen consistent growth in sales and profits through its retail outlets. Table 4-7 Comparative Performance of Delivery Network Designs Retail Manufacturer Manufacturer Distributor Distributor Manufacturer Storage Storage with Storage with Storage with Storage with Storage with with Direct In-Transit Package Last-Mile Pickup Customer Shipping Merge Carrier Delivery Delivery Pickup Response time 1 4 4 3 2 4 Product variety 4 1 1 2 3 1 Product availability 4 1 1 2 3 1 Customer Varies from 4 3 2 1 5 experience 1 to 5 Time to market 4 1 1 2 3 1 Order visibility 1 5 4 3 2 6 Returnability 15 5 4 3 2 Inventory 41 1 2 3 1 Transportation 1 4 3 2 5 1 Facility and handling 6 1 2 3 4 5 Information 1 4 4 3 2 5 Key: 1 corresponds to the strongest performance and 6 the weakest performance.

86 Chapter 4 • Designing Distribution Networks and Applications to Online Sales Table 4-8 Performance of Delivery Networks for Different Product/Customer Characteristics Retail Manufacturer Manufacturer Distributor Distributor Manufacturer Storage Storage with Storage with Storage with Storage Storage with with with Direct In-Transit Package Last-Mile Pickup Customer Shipping Merge Carrier Delivery Delivery Pickup High-demand product +2 −2 −1 0 +1 −1 Medium-demand +1 −1 0 +1 0 0 product −1 +1 0 +1 −1 +1 Low-demand product −2 +2 +1 0 −2 +1 Very low-demand +1 −1 −1 +2 +1 0 product −1 +2 +1 Many product sources +2 −2 −2 +1 0 +2 High product value −1 +2 Quick desired response −2 +1 0 −1 +1 −2 High product variety +2 Low customer effort +1 0 +2 +2 +2 −1 Key: +2 = very suitable; +1 = somewhat suitable; 0 = neutral; −1 = somewhat unsuitable; −2 = very unsuitable. 4.4 ONLINE SALES AND THE DISTRIBUTION NETWORK In this section, we use ideas discussed earlier in the chapter to see how the Internet has affected the structure and performance of various distribution networks. The goal is to understand what drove the successful introduction of online sales in some networks and not others, and how these networks are likely to evolve. Similar to our consideration with distribution networks, we build a scorecard based on how online sales affect a supply chain’s ability to meet customer needs and the cost of meeting those needs. We now detail the contents of each scorecard category. Impact of Online Sales on Customer Service As with distribution networks considered earlier, we start by studying how online sales affect customer service elements such as response time, product variety, availability, customer experience, time to market, visibility, and returnability. We also look at factors such as direct sales and the ability to offer flexible pricing that help companies selling online. RESPONSE TIME TO CUSTOMERS When selling physical products that cannot be downloaded, online sales take longer to fulfill a customer request than does a retail store because of the shipping time involved. Thus, customers who require a short response time may not use the Internet to order a product. There is no such delay, however, for information goods. The Internet has facilitated almost instantaneous access to movies, music, and books in digital form. PRODUCT VARIETY A company selling online finds it easier to offer a larger selection of products than a bricks-and-mortar store. For example, Netflix offers a much larger selection of movies than any video rental store. Offering the same selection at a store would require a huge location with a correspondingly large amount of inventory. PRODUCT AVAILABILITY By aggregating its inventory, a company selling online improves product availability. Better information on customer preferences also allows firms selling online to improve availability.

Chapter 4 • Designing Distribution Networks and Applications to Online Sales 87 CUSTOMER EXPERIENCE Online sales affect customer experience in terms of access, customization, and convenience. Unlike most retail stores that are open only during business hours, the Internet allows a customer to place an order at any convenient time. In fact, W.W. Grainger has observed a surge in online orders after their bricks-and-mortar stores close. Online sales also allow a firm to access customers who are geographically distant. Using the Internet, a small specialty retail store located near Chicago can reach customers all over the United States, or even the world. Access to online sales is limited only by the customers’ access to the Internet. The Internet offers an opportunity to create a personalized buying experience for each customer. For example, Amazon displays products that are related to what customers have recently purchased or browsed. Firms that focus on mass customization can use the Internet to help customers select a product that suits their needs. For example, Dell allows customers to customize their computers by using the options available on the Dell Web site. For both consumers and companies, online sales can increase the ease with which one does business. Customers do not have to leave home or work to make a purchase. For many companies selling online, such as Peapod, information from past purchases is used to significantly speed up order placement for the customer. FASTER TIME TO MARKET A firm can introduce a new product much more quickly online as compared to physical channels. A firm that sells PCs through physical channels must produce enough units to stock the shelves at its distributors and retailers before it starts to see revenue from the new product. A firm selling online, in contrast, makes a new product available online as soon as the first unit is ready to be produced. This is evident at Walmart.com, where larger new TVs go on sale well before they are sold at Wal-Mart stores. ORDER VISIBILITY The Internet makes it possible to provide visibility of order status. From a customer’s perspective, it is crucial to provide this visibility because an online order has no physical equivalent to a customer shopping for an item at a retail store. RETURNABILITY Returnability is harder with online orders, which typically arrive from a centralized location. It is much easier to return a product purchased at a retail store. The proportion of returns is also likely to be much higher for online orders because customers are unable to touch and feel the product before their purchase. Going online thus increases the cost of reverse flows. DIRECT SALES TO CUSTOMERS The Internet allows manufacturers and other members of the supply chain that do not have direct contact with customers in traditional channels to get customer feedback and build a relationship with the customer. Social networking channels such as Facebook and Twitter allow firms to directly pitch products and promotions to customers. FLEXIBLE PRICING, PRODUCT PORTFOLIO, AND PROMOTIONS Given the ease of changing prices and assortments online, the Internet allows a company selling online to manage revenues from its available product portfolio much more effectively than do traditional channels. Promotion information can be conveyed to customers quickly and inexpensively using the Internet as long as the business has access to its customer network. Groupon is a company that has used social networking online to push promotions to customers. EFFICIENT FUNDS TRANSFER The Internet and cell phones can enhance the convenience and lower the cost of revenue collection, especially in small amounts. For example, after the earthquake in Haiti in 2010, Mercy Corps transferred $40 automatically into each Haitian person’s account allowing him or her to buy food at local merchants. This was much more efficient than handing out cash or vouchers.