CHAPTER 6
Supply Chain Coordination
After reading this chapter you will be able to
- Understand a common supply chain dynamic that is a major contributor to the “boom to bust” business cycle
- Appreciate the factors that contribute to this supply chain dynamic
- Gain an overview of the Global Data Synchronization Network (GDSN) and see how it can help improve supply chain coordination
- Evaluate methods such as collaborative planning, forecasting, and replenishment (CPFR) to improve coordination and combat the bullwhip effect
The spread of high-speed data communication networks and computer technology has made it possible to manage supply chains with a level of precision just not feasible even as recently as the 1990s. Those organizations that learn to use the techniques and technologies that are now available can build supply chains that have a competitive advantage in their markets.
Because the capability exists to react much more quickly to changes in market demand, this capability is now a point of competition. Business competition based on supply chain efficiency is becoming a central fact in many markets. To develop this capability, individual companies and entire supply chains need to learn new behaviors and they need to enable these new behaviors with the use of appropriate technology.
The Bullwhip Effect
One of the most common dynamics in supply chains is a phenomena that has been dubbed “the bullwhip effect.” What happens is that small changes in product demand by the consumer at the front of the supply chain translate into wider and wider swings in demand experienced by companies further back in the supply chain. Companies at different stages in the supply chain come to have very different pictures of market demand and the result is a breakdown in supply chain coordination. Companies behave in ways that at first create product shortages and then lead to an excess supply of products.
This dynamic plays out on a larger scale in certain industries in what is called a “boom to bust” business cycle. In particular this affects industries that serve developing and growth markets where demand can suddenly grow. Good examples of this can be found in the industries that serve the telecommunications equipment or computer components markets. The cycle starts when strong market demand creates a shortage of product. Distributors and manufacturers steadily increase their inventories and production rates in response to the demand. At some point either demand changes or the supply of product exceeds the demand level. Distributors and manufacturers do not at first realize that supply exceeds demand and they continue building the supply. Finally the glut of product is so large that everyone realizes there is too much. Manufacturers shut down plants and lay off workers. Distributors are stuck with inventories that decrease in value and can take years to work down.
This dynamic can be modeled in a simple supply chain that contains a retailer, a distributor, and a manufacturer. In the 1960s a simulation game was developed by the Massachusetts Institute of Technology's Sloan School of Management that illustrates how the bullwhip effect develops. The simulation game they developed is called the “beer game.” It shows what happens in a hypothetical supply chain that supports a group of retail stores that sell beer, snacks, and other convenience items. The results of the beer game simulation teach a lot about how to coordinate the actions of different companies in a supply chain.
Peter Senge in his book, The Fifth Discipline (Senge, Peter M., 1990, The Fifth Discipline: The Art and Practice of the Learning Organization, New York: Doubleday/Currency, Chapter 3), devotes a chapter to exploring how the bullwhip effect gathers momentum and what can be done to avoid it. The beer game starts with retailers experiencing a sudden but small increase in customer demand for a certain brand of beer called Lover's Beer. Orders are batched up by retailers and passed on to the distributors who deliver the beer. Initially, these orders exceed the inventory that distributors have on hand so they ration out their supplies of Lover's Beer to the retailers and place even larger orders for the beer with the brewery that makes Lover's Beer. The brewery cannot instantly increase production of the beer so it rations out the beer it can produce to the distributors and begins building additional production capacity.
At first the scarcity of the beer prompts panic buying and hoarding behavior. Then as the brewery ramps up its production rate and begins shipping the product in large quantities, the orders that had been steadily increasing due to panic buying suddenly decline. The glut of product fills up the distributors' warehouses, fills all the retailers' unfilled back orders, and exceeds the actual consumer demand. The brewery is left with excess production capacity, the distributors are stuck with excess inventory, and the retailers either cancel their beer orders or run discount promotions to move the product. Everybody loses money. Exhibit 6.1 illustrates how each company sees product demand and the distortion that causes such havoc.
The costs of the bullwhip effect are felt by all members of the supply chain. Manufacturers add extra production capacity to satisfy an order stream that is much more volatile than actual demand. Distributors carry extra inventory to cover the variability in order levels. Transportation costs increase because excess transportation capacity has to be added to cover the periods of high demand. Along with transportation costs, labor costs also go up in order to respond to the high demand periods. Retailers experience problems with product availability and extended replenishment lead times. During periods of high demand, there are times when the available capacity and inventory in the supply chain cannot cover the orders being placed. This results in product rationing, longer order replenishment cycles, and lost sales due to lack of inventory.
Product Demand Distortion “The Bullwhip Effect”
Inventory levels in supply chain over time illustrating the wild swings that develop as product demand distortion moves from customer to retailer to distributor to manufacturer. Swings in product demand appear more pronounced to companies further up the supply chain. This distortion makes effective supply chain management very difficult.
Coordination in the Supply Chain
Research into the bullwhip effect has identified five major factors that cause the effect. These factors interact with each other in different combinations in different supply chains, but the net effect is that they generate the wild demand swings that make it so hard to run an efficient supply chain. These factors must be understood and addressed in order to coordinate the actions of any supply chain. They are:
- Demand Forecasting
- Order Batching
- Product Rationing
- Product Pricing
- Performance Incentives
Demand Forecasting
Demand forecasting, based on orders received instead of end user demand data will inherently become more and more inaccurate as it moves up the supply chain. Companies that are removed from contact with the end user can lose touch with actual market demand if they view their role as simply filling the orders placed with them by their immediate customers. Each company in a supply chain sees fluctuations in the orders that come to them that are caused by the bullwhip effect. When they use this order data to do their demand forecasts, they just add further distortion to the demand picture and pass this distortion along in the form of orders that they place with their suppliers.
Clearly, one way to counteract this distortion in demand forecasts is for all companies in a supply chain to share a common set of demand data from which to do their forecasting. The most accurate source of this demand data is the supply chain member closest to the end-use customer (if not the end-use customers themselves). Sharing point-of-sales (POS) data among all the companies in a supply chain goes a long way toward taming the bullwhip effect because it lets everyone respond to actual market demand instead of supply chain distortions.
Order Batching
Order batching occurs because companies place orders periodically for amounts of product that will minimize their order-processing and transportation costs. As discussed in the section on inventory control in Chapter 2, companies tend to order in lot sizes determined by the economic order quantity (EOQ). Because of order batching, these orders vary from the level of actual demand and this variance is magnified as it moves up the supply chain.
The way to address demand distortion caused by order batching is to find ways to reduce the cost of order-processing and transportation. This will cause EOQ lot sizes to get smaller and orders to be placed more frequently. The result will be a smoother flow of orders that distributors and manufacturers will be able to handle more efficiently. Ordering costs can be reduced by using electronic ordering technology. Transportation costs can be reduced by using third-party logistics suppliers (3PLs) to cost effectively pick up many small shipments from suppliers and deliver small orders to many customers.
Product Rationing
This is the response that manufacturers take when they are faced with more demand than they can meet. One common rationing approach is for a manufacturer to allocate the available supply of product based on the number of orders received. Thus if the available supply equals 70 percent of the orders received, the manufacturer will fill 70 percent of the amount of each order and backorder the rest. This leads distributors and retailers in the supply chain to raise their order quantities artificially in order to increase the amount of product that gets rationed to them. This behavior greatly overstates product demand and it is called “shortage gaming.”
There are several ways to respond to this. Manufacturers can base their rationing decisions on the historical ordering patterns of a given distributor or retailer and not on their present order sizes. This eliminates much of the motivation for the shortage gaming that otherwise occurs. Manufacturers and distributors can also alert their customers in advance if they see demand outstripping supply. This way product shortages will not take buyers by surprise and there will be less panic buying.
Product Pricing
Product pricing causes product prices to fluctuate, resulting in distortions of product demand. If special sales are offered and product prices are lowered, it will induce customers to buy more product or to buy product sooner than they otherwise would (forward buying). Then prices return to normal levels and demand falls off. Instead of a smooth flow of products through the supply chain, price fluctuations can create waves of demand and surges of product flow that are hard to handle efficiently.
Answers to this problem generally revolve around the concept of “everyday low prices.” If the end customers for a product believe that they will get a good price whenever they purchase the product, they will make purchases based on real need and not other considerations. This in turn makes demand easier to forecast and companies in the supply chain can respond more efficiently.
Performance Incentives
These are often different for different companies and individuals in a supply chain. Each company can see its job as managing its position in isolation from the rest of the supply chain. Within companies, individuals can also see their jobs in isolation from the rest of the company. It is common for companies to structure incentives that reward a company's sales force on sales made each month or each quarter. Therefore as the end of a month or a quarter approaches, the sales force offers discounts and takes other measures to move product in order to meet quotas. This results in product for which there is no real demand being pushed into the supply chain. It is also common for managers within a company to be motivated by incentives that conflict with other company objectives. For instance, a transportation manager may take actions that minimize transportation costs at the expense of customer service or inventory carrying costs.
Alignment of performance incentives with supply chain efficiencies is a real challenge. It begins with the use of accurate activity-based costing (ABC) data that can highlight the associated costs. Companies need to quantify the expenses incurred by forward buying due to month-end or quarter-end sales incentives. Companies also need to identify the effect of conflicting internal performance incentives. The next step is to experiment with new incentive plans that support efficient supply chain operation. This is a process that each company needs to work through in its own way.
Eliyahu Goldratt wrote a book titled, The Goal, about a factory manager's quest to save his factory from being closed down for lack of profitability. It chronicles the process that the manager and his staff go through as they learn how to save their factory. What they learn is how to apply the principles of what Mr. Goldratt calls the “Theory of Constraints.”
Mr. Goldratt and others have realized that the theory of constraints applies equally well to the operation of a whole supply chain as to the operation of a single factory within a supply chain. Lawrence Fredendall and Ed Hill in their book, Basics of Supply Chain Management (Fredendall, Lawrence D., and Ed Hill, 2001, Basics of Supply Chain Management, Boca Raton, FL: St. Lucie Press), have put forth a clear explanation for how to apply the theory of constraints to synchronize the operations of a supply chain.
The theory of constraints provides a useful model to conceptualize and manage the supply chain within a single company or across a collection of companies. The theory of constraints is based upon the idea that all systems have at least one constraint and that it is better to manage constraints than to try to eliminate them. This is because when one part of a system ceases to be a constraint, a different constraint will occur in another part of the system. This is inevitable because the capacities of each part of a system are not all the same. So instead of forever reacting to new constraints or bottlenecks as they appear, why not choose a small group of constraints and manage them deliberately and efficiently?
To apply this model, the first step is to define the goal and decide what measurements will be used to measure progress toward the goal. Mr. Goldratt's definition of the goal for a manufacturing company also works for a supply chain. The goal is defined as “Increase throughput while simultaneously reducing both inventory and operating expense.” Throughput is the rate at which sales to end customers occur.
Once a goal has been defined and there is agreement on how to measure progress toward the goal, it is possible to apply the five focusing steps. These steps help clarify the situation being investigated and lead to the decisions necessary to reach the goal. The five steps are:
Identify the System's Bottlenecks or Constraints—Trace out the workflows and the paths that materials travel in a factory or a supply chain. Find out where slowdowns and backups occur.
Decide How to Exploit These Bottlenecks—Figure out how to maximize the operation of those activities that are bottlenecks. The rate of throughput for the entire system is set by the rate of throughput achieved by the bottlenecks. Ensure the bottlenecks operate at maximum capacity by providing them with enough inventory so that they can continue to operate even if there are occasional slowdowns elsewhere in the system.
Subordinate Everything Else to the Above Decision—Do not try to maximize the operation of a non-bottleneck operation. Additional productivity achieved by non-bottleneck operations that exceeds the capacity of the bottlenecks to process will be neutralized anyway by the slowdowns and backups caused at the bottlenecks. Synchronize all system operations to the rates that can be efficiently processed by the bottleneck operations.
Elevate the System's Bottlenecks—Add additional processing capacity to the bottleneck activities. Since the rate of throughput of the entire system is set by the throughput of the bottlenecks, improvements in the bottlenecks will increase the efficiency of the entire system and provide the best return on investment.
If, in a Previous Step, a Bottleneck Has Been Broken, Go Back to Step 1—As the capacity of one system bottleneck is elevated, it may cease to be a bottleneck. The bottleneck may transfer to another operation that could keep up before but now cannot keep up with the new increase in capacity. Watch the entire system to see where slowdowns and backups occur; they may shift from one area to another. If this occurs, start again at Step 1.
The theory of constraints says that the throughput of the whole system is set by the capacity of the bottlenecks. Exhibit 6.2 shows a sample diagram of workflows and bottlenecks in a factory. This model of workflows in a factory can be applied to the workflows in a supply chain. One constraint or bottleneck in every supply chain is the demand that is generated by the market that the supply chain serves. In many cases, market demand is the only constraint because supply of products equals or exceeds demand. In cases where demand exceeds supply there will be some other constraints elsewhere in the supply chain. If we apply this model to a supply chain, we get a powerful method to organize and manage supply chain operations.
Flow of Work and Inventory through a Factory
FLOW OF WORK AND INVENTORY THROUGH A FACTORY
The bottlenecks or constraints in the flow of work through this factory are operations C and E in Exhibit 6.2. The productivity set by these two operations sets the pace for the entire factory. Productivity improvements in the other operations will not result in any improvement in the productivity of the factory as a whole. Apply the five focusing steps to manage this system and move it toward the goal defined for it.
A very effective response to the bullwhip effect is to manage the entire supply chain as a single entity and to synchronize it to the timing of actual market demand. Exhibit 6.3 illustrates this idea. This can happen if the supply chain participants closest to the end-use customers share their sales numbers and their sales forecasts with the other companies in the supply chain. Each company can then manage their actions based on the most accurate data about market demand.
Buffers in the supply chain are determined by the degree of uncertainty about future market demand and the service levels required by the market. The lower the uncertainty about demand, the smaller the buffers can be and still maintain high service levels. Companies can manage their buffers by using either productive capacity or inventory, whichever is most cost effective for them.
Synchronized supply chains avoid the volatile waves of demand that are generated by the bullwhip effect. And increased predictability makes the productivity of each company easier to manage and the supply chain as a whole becomes more efficient and profitable.
FLOW OF INVENTORY THROUGH A SYNCHRONIZED SUPPLY CHAIN
This model is called “drum-buffer-rope.” Market demand is the constraint on the system and it sets the drumbeat or pace of the supply chain. Individual companies manage uncertainty in their stage of the supply chain by using a buffer of either inventory or productive capacity. Buffers are kept low because uncertainty is minimized by sharing market demand data. This data is the rope that ties the participants together and allows them to synchronize their actions.
Flow of Inventory through a Synchronized Supply Chain
Supply Chain Product Data Standards
Historically, companies have assigned their own part numbers to the items that they buy and sell. This worked well enough in a slower time when supply chains were less complex and when products themselves were less complex. Those were times we now refer to as the “good old days.” Increasing competition and demands from customers to deliver products faster and cheaper shapes the world we live in today. At the same time, the array and complexity of products in our economy have increased dramatically and that trend will clearly continue and even accelerate.
In order to be competitive and also profitable, companies need to find ways to reduce or eliminate costs associated with routine and repetitive business transactions. Those transactions often fall in the areas of purchasing, billing, accounts receivable, and accounts payable. It is in these areas that the confusion caused by translating part numbers is most noticeable. Time spent translating one part number to another part number for the same item adds very little, if any, value to the transaction. The errors that result from errors in translation are the cause of many problems in invoicing and making payments. These problems consume people's time and slow down cash flow. All these expenses simply eat away at profit margins that are already thin enough.
In addition to the operating problems caused by using different part numbers for the same item, another consequence is a lack of accuracy and clarity in sales history data. Part number translation errors result in sales of some items being undercounted and sales of other items being overcounted. And sales of many items are simply not counted at all or they are lumped under a miscellaneous part number such as the famous “9999” part number. Sales history data is usually the basis for forecasting future demand and this fuzziness in the data hampers efforts to improve demand forecasts, production scheduling, and inventory management.
In order for companies to coordinate effectively, they need to have a single part number that stays with a part as it makes its way through the supply chain. That number is the electronic product code (EPC) number. Companies that do business together need to be able to tag every item that they buy and sell with an EPC number. They can still use their internal part numbers for internal operations if they wish. But when they communicate with each other they need to use EPC numbers so as to eliminate the need to do part-number translations. There are more valuable and profitable things that can be done with the time and money that now goes into translating part numbers and dealing with translation-related problems.
Global Data Synchronization Network
The Global Data Synchronization Network (GDSN) is a network of independently owned and operated databases that can exchange data with each other and the GS1 Global Registry. The GS1 Global Registry acts as a central coordinator between all the other databases to provide for timely and traceable distribution of verified product descriptive information between all the databases. It is the locator and routing mechanism for finding source data and sending requested data between databases.
GS1 is a global, not-for-profit organization of member organizations, including GS1 US, representing more than 100 countries around the world. GS1 is based in Brussels, Belgium. GS1 US is the former Uniform Code Council and consists of the EAN UCC System, UCCnet, EPCglobal US, RosettaNet, and UNSPSC. GS1 US is based in New Jersey, USA (www.gs1us.org).
Companies join the GS1 Global Registry to keep connected to their trading partners that also join the Global Registry. Parties in a supply chain—manufacturers, logistics providers, distributors, retailers—then subscribe to the database of their choice and through these databases they can both publish data about their products to other parties as well as request and receive data from other companies about their products. This is illustrated in Exhibit 6.4.
The GDSN is being administered by GS1 and increasingly is being used by companies in consumer goods retail and related areas. It allows data about products to be continuously updated as new products are released, existing products evolve, and obsolete products are discontinued. The benefits are significant beginning with the fact that each company needs only to make a single connection to their selected database or “data pool” as GDSN calls them. Once they do this they can send and receive data to and from any other company that is connected to any other data pool that is part of the GDSN.
Other benefits include things such as elimination of the need for companies to maintain massive cross-reference tables to translate between the different part numbers for the same product that are used by different supply chain partners. This reduces many ordering and billing errors that consume people's time and result in delays in product deliveries and cash flows between companies. It also simplifies order tracking and tracing individual items as they move through a supply chain.
Global Data Synchronization Network (GDSN)
- Load Item and Location Data – the seller or manufacturer registers with a GS1 certified data pool and uploads item and location data to their data pool.
- Register Data – a small subset of item and location data is sent by the data pool to GS1 Global Registry.
- Request for Subscription Data – the buyer or retailer subscribes to a data pool and to categories of products or to particular suppliers to receive the related item and location data. Buyer requests data from their data pool. The data pool requests this data from the GS1 Global Registry and the Global Registry sends the request to the data pool containing this data.
- Publish Requested Data – the seller's data pool provides requested item and location data to the data pool of the buyer and the buyer's data pool sends the data to the buyer. Buyer updates its systems with this data. Buyer and seller now have identical item and location data—data synchronization is complete.
Source: GS1 US (formerly known as Uniform Code Council Inc)
Product Classification
Products that move through a supply chain need to be identified and traced so that people know how many products are moving through their supply chains. Products also need to be classified so that people know what types of products they are handling. All supply chains handle a mix of different product types and that mix changes over time. As the product mix changes, the supply chain itself must change.
There are two major standards presently in use for product classification. The first one is the United Nations Standard Products and Services Code (UNSPSC). The United Nations Development Program (UNDP) and Dun & Bradstreet Corporation (D&B) jointly developed the UNSPSC in 1998. The UNSPSC is a hierarchical classification with five levels. These levels allow analysis by drilling down or rolling up to analyze expenditures and product usage at each level. Each level in the hierarchy has its own unique number. Starting with the highest level, the five levels are segment, family, class, commodity, and business function.
The second major product classification standard presently in use is the GS1 Global Product Code or GPC. The GPC was developed by GS1 and is used in the GDSN to identify different types of products. The GPC is also a hierarchical classification scheme and it has four levels: Segment; Family; Class; and Brick. These two product-classification schemes are not mutually exclusive and they can be used together. It does require all parties to agree on the rules they will use to translate product codes between UNSPSC numbers and GPC numbers.
Collaborative Planning, Forecasting, and Replenishment
To facilitate the coordination that is needed in supply chains, an industry group known as the Voluntary Interindustry Commerce Standards (VICS) (www.vics.org) has set up a committee to investigate collaborative planning, forecasting, and replenishment issues (CPFR). This committee documents best practices for CPFR and creates guidelines to follow for CPFR.
Supply chains where people use technology to support a CPFR process are the most efficient supply chains, because they can best manage the factors that give rise to the bullwhip effect.
The CPFR process is divided into the three activities of planning, forecasting, and replenishment. Within each activity there are several steps:
Collaborative Planning
- Negotiate a front-end agreement that defines the responsibilities of the companies that will collaborate with each other
- Build a joint business plan that shows how the companies will work together to meet demand
Collaborative Forecasting
- Create sales forecasts for all the collaborating companies
- Identify any exceptions or differences between companies
- Resolve the exceptions to provide a common sales forecast
Collaborative Replenishment
- Create order forecasts for all the collaborating companies
- Identify exceptions between companies
- Resolve the exceptions to provide an efficient production and delivery schedule
- Generate actual orders to meet customer demand
CPFR in Action
For an example of how CPFR can work let's return to the example of Nimble Co. In the section on product design in Chapter 3, we saw how Nimble Co. developed a home entertainment system that was much simpler to manufacture than a competitor's system. This simpler design is in turn supported by a less-complex supply chain that reduces production costs and increases responsiveness to market demands. All of this is central to the competitive success that Nimble Co. is enjoying.
Nimble Co. has collaboration agreements in place with its supply chain partners and has an ongoing planning, forecasting, and replenishment process in place with these partners. Nimble Co. receives point of sale (POS) data that shows the actual sales of its systems in retail stores. From these same retailers, Nimble Co. receives regular updates of their sales forecasts and their inventory levels of Nimble Co. home entertainment systems. Nimble Co. uses this data to plan its production schedule and it also shares this data with the component manufacturers who provide parts for its home entertainment system. This way the component manufacturers can plan their own production schedules.
In looking at the sales data and forecasts, Nimble Co. sees that demand for their product is growing faster than anticipated in their yearly plan, and they need to increase production. Nimble Co. revises its production schedule for the year and takes the new plan to its key component suppliers to negotiate additional purchases of their components. It turns out that one component supplier cannot quickly ramp up their production but a second supplier has a component that could fill the need with just a slight modification to the design of one part of Nimble Co.'s home entertainment system. Because all affected parties know what is going on and have enough lead time, the design changes are made and production schedules are increased to meet the rise in product demand without any retailers running out of inventory.
The benefits illustrated in this scenario are numerous. To begin with, the bullwhip effect is diminished because all companies in the supply chain can see real-time sales data and share sales forecasts. This allows everyone to optimize their production schedules, inventory levels, and delivery schedules. Next there are the benefits associated with Nimble Co. being able to quickly see a real rise in customer demand and coordinate with its suppliers to increase production schedules over previously planned levels. Even though one component supplier was not able to accommodate Nimble Co.'s increased production schedule, another supplier had a workable substitute. Changes were made to the product design, production was increased, and no retailer lost sales revenue due to running out of inventory.
Those companies that can create collaborative supply chains will have a significant competitive advantage. Collaboration is not easy to implement and it will take time to become more common in business. However, prominent companies are already beginning to lead the way. Companies such as Wal-Mart, Dell, and Procter & Gamble share POS data with all the other companies in their respective supply chains. The companies in these supply chains are also starting to share inventory data with each other. Sharing this kind of information provides a basis for each company to make decisions about its own activities that will yield better efficiencies and profits for itself and for the supply chain as a whole.
How to Start Supply Chain Collaboration
The best place to start in any effort to promote collaboration is to measure the bullwhip effect within your company. Over a period of time such as a quarter or a year, compare the volume and frequency of orders you receive from your customers with the volume and frequency of orders you place with your suppliers. Plot them out on a graph so everyone can see the divergence between incoming customer orders and your outgoing supplier orders. What is the extent of this divergence? Where is your company located in the supply chain—is it toward the front of the chain close to the end customer or is it further toward the back of the chain? Remember, the distortion caused by divergence of incoming orders with outgoing orders increases as it moves back through the supply chain.
Many companies are not aware of the cost of the bullwhip effect on their supply chains. Traditionally, demand variability caused by the bullwhip effect was taken as a given and companies worked on their own to develop better capabilities to respond to fluctuations in demand. It may instead be far more efficient for companies to work together to actually reduce the fluctuations in demand. A company can either try to optimize its individual response to fluctuating demand or it can collaborate with other companies to reduce the fluctuations themselves.
Once you have established the magnitude of the bullwhip effect in your company, then get some estimates of the cost consequences in different areas of the company. What is the effect of this demand variability on production costs and scheduling? What is the effect on transportation costs and shipping and receiving costs? What inventory levels are needed to maintain service levels in such a volatile situation and what is the associated carrying cost? What is the effect on product availability and order lead times—are sales lost because of lack of inventory? These estimates show the cost to the company of dealing with demand fluctuations. This is the basis upon which to discuss what it might be worth to fix the bullwhip effect.
EXECUTIVE INSIGHT
The Tao of Supply Chains: Effective supply chain collaboration requires that the people involved be able to see accurate and timely data showing inventory levels at different stages in the supply chain. What follows is a story about creating a simple supply chain visibility system that enabled very effective collaboration.
Sun Tzu was a Taoist philosopher who lived in China about 2,500 years ago. He wrote a book called The Art of War. It isn't so much a book about war as it is a book about the art of competition and collaboration—whether in business, politics, the military, or even sports. I've puzzled through this book several times, and the concepts that I've taken away have helped me develop and preserve a reputation for IT agility.
For six years I was the chief information officer (CIO) of Network Services Company. Network Services is a nationwide distribution cooperative that provides foodservice items, janitorial supplies, and printing paper. It is wholly owned by its 80+ member companies. They each have their own facilities and internal IT systems and they have their own local customers. They also work together to serve national account customers. The members' collective revenue was then more than $7 billion, and Network's total national account revenue was more than $500 million, growing by double-digit percentages every year. We provided customers with a tailored package of products and supply chain services to lower their overall operating costs.
One of our biggest national account customers was a chain of stores that each holiday season used specially printed paper items to promote its holiday theme. Those items were used in the customer's 4,500 stores during November and December, and when January arrived any remaining inventory had to be written off. The same holiday print designs were never used two years in a row. In years past, there had been excess inventory of around 4 percent, amounting to almost $600,000 in costs that had to be written off by the customer.
This retail chain hired a new purchasing manager who decided we could all do better than in prior holiday seasons. He called us out to the company's headquarters that summer for a meeting. There he announced his intention to reduce excess inventory of the specially printed holiday items by 50 percent or more. We still had to maintain 100 percent product availability for all of its stores and minimize expensive movements of inventory from one region to another to meet unexpected demand. He asked us how we were going to work with him to make that happen. I told him we understood what he wanted and that we'd be back in touch with the specifics in a few weeks.
As we flew home, our sales director on the account told me this was a high-visibility project with the customer, and we had to figure out how to do it. He reminded me that it was already halfway through the summer, so we had to be ready to go in 90 days because we would begin stocking inventory in our distribution centers by October. And, of course, we couldn't spend lots of money on this because margins were tight. In addition, all the parties in this supply chain used different enterprise resource planning (ERP) systems. And even within Network Services, the 26 member companies that served the account used different ERP systems. Several times on that flight, I experienced a sudden falling sensation in my stomach, and it wasn't due to air turbulence.
At times like these my identity as IT Agility Man hangs in the balance. Can I rise to the challenge, or will I flee in panic? Agility means doing three things: First, taking a deep breath; second, taking another deep breath; then, remembering The Art of War and asking, “What would Master Sun Tzu do?”
The concepts that I've been able to absorb from Master Sun tell me that apparent complexity is really composed of simple underlying patterns. If I can discern those underlying patterns, then I can devise simple and effective responses. So what was the pattern here? As I saw it, the need was to track daily product usage, constantly update demand forecasts, move inventory so as to cover demand, and use it all up by the end of the season.
That meant effective collaboration among all parties in the supply chain to respond as actual demand unfolded. If our initial assumptions about demand were not entirely accurate (and they never are), we needed to be able to reposition inventory among distribution centers earlier and more efficiently. No sudden air freighting of paper goods to stores across the country. So, I asked myself, “What can IT provide that will enable this collaboration?” Obviously, what was needed was a continually updated, end-to-end view of product in the supply chain that was visible at all times to people at my company, the manufacturers, and the customer. That would be the basis for our collaboration and decision making.
I knew of several fine software vendors' products that could do that, but they cost more money than I had to spend and took more time to install than I had available. So much for the orthodox ideas. What else could I do? Master Sun says, “Therefore, those skilled at the unorthodox are infinite as heaven and earth, inexhaustible as the great rivers.” Wow. What unorthodox ideas could I come up with?
Master Sun says, “There are only five notes in the musical scale, but their variations are so many that they cannot all be heard. There are only five basic colors, but their variations are so many that they cannot all be seen.” Does this mean that there is a combination of basic IT components that I could use to quickly create my end-to-end supply chain picture and keep it constantly updated?
What basic IT components do all parties in this supply chain have easy access to, and how could I combine them into the system I needed? I'm not going to give you the whole answer because then you wouldn't get to practice your own agility and figure it out for yourself. But I will give you some hints. The components are spreadsheets, text files, e-mail, a few Web pages, a relational database, and some Java programs that took about three weeks to write and test.
We assembled these components into a system that collected data from all members of the supply chain. The data consisted of inventory amounts that were in production, in warehouses, and on order. It also included invoice data that showed our deliveries to the customer's stores, which allowed us to track actual demand at the store levels and regional levels.
The system was up and running by October. It was extremely cost effective to build. We used it to facilitate conference calls that increased in frequency as the season progressed. On those calls, we all reviewed the numbers and projected run-out dates. We made decisions and continued to tweak the system to incorporate new views of the data and new calculations.
We reduced excess inventory from 4 percent in prior years to 1.3 percent that year on increased total sales, and the dollar value of the excess inventory dropped to less than $200,000. As we reviewed the holiday season results in January, the new purchasing manager said he was quite pleased with our performance. We worked with him and the manufacturers to document what we learned, make further improvements, and extend the system to cover the rollout of other new products—not just holiday items. Thank you, Master Sun.
If you want to know more about how I designed and built this supply chain visibility system you can contact me by e-mail at: [email protected] and I'll be glad to share the details with you.
Sales and Operations Planning
Sales and operations planning (S&OP) is, to a large degree, a further elaboration of the planning process described in Chapter 2 (page 42). It also shares elements in common with the CPFR process described in this chapter (page 199). As companies develop effective S&OP processes internally and collaboratively with their supply chain partners, they will see significant improvements in their supply chain management capabilities.
The Association for Operations Management (APICS, formerly American Production and Inventory Control Society), endorses a definition of sales and operations planning provided by Tom Wallace in his book Sales & Operations Planning:
Sales & Operations Planning (S&OP) is a business process that helps companies keep demand and supply in balance. It does that by focusing on aggregate volumes—product families and groups—so that mix issues— individual products and customer orders—can be handled more readily. It occurs on a monthly cycle and displays information in both units and dollars. S&OP is cross functional, involving General Management, Sales, Operations, Finance, and Product Development. It occurs at multiple levels within the company, up to and including the executive in charge of the business unit, that is, division president, business unit general manager, or CEO of a smaller corporation. S&OP links the company's Strategic Plans and Business Plan to its detailed processes—the order entry, master scheduling, and purchasing tools it uses to run the business on a week-to-week, day-to-day, and hour-to-hour basis. Used properly, S&OP enables the company's managers to view the business holistically and gives them a window into the future. (Tom Wallace, Sales & Operations Planning, T.F. Wallace & Co., Cincinnati, OH, 2000.)
The purpose of S&OP is to routinely review customer demand for different products and the sources of supply for those products, and then re-plan or adjust existing plans to best match supply with demand. The process focuses on changes to earlier supply and demand forecasts and helps managers understand how well the company is doing in the process of balancing supply and demand. The primary aim is to continuously adjust company operating procedures so as to accomplish its strategic goals and annual sales targets in light of expected future conditions.
The continuous adjustment of sales and operations plans as forecasts change is the method by which a company takes steps each month to best meet its annual sales targets. The S&OP process can be thought of as having five main steps as shown in Exhibit 6.5 below.
S&OP takes a phased and iterative approach to the planning and operations in a company. Instead of attempting to do one master plan each year and then spend the rest of the year following that one plan, S&OP takes the approach of continuously reevaluating demand and supply conditions and continuously adjusting plans in light of changing conditions. It is a process of constantly responding to change. Fixed annual plans worked better in the slower and more predictable industrial economy of the last century. The real-time unpredictable nature of the economy of this century makes it critical for companies to constantly respond to changing conditions.
Sales & Operations Planning Cycle
This planning cycle is done on a monthly basis, and increasingly, it needs to be done on a weekly or even daily basis in order to respond effectively as conditions change.
Companies are encouraged to take an “outside-in” approach to their S&OP planning. People should begin their forecasting and planning by considering the effect of events that are largely outside of their control such as the actions of customers, partners, and competitors. These actions typically have the greatest impact, either positive or negative, on a company's ability to meet its sales and operating targets. Collaboration with customers and partners (if not with competitors) provides for more accurate information about these possible actions and thus makes the S&OP process more effective. This is where practices from CPFR can be beneficial to S&OP.
In a world now awash in data thanks to computers, the Internet, and the global reach of telecommunication networks, it is important for people in the S&OP process to know the data they need and find ways to get that data quickly and accurately. The collection of massive amounts of unnecessary data or inaccurate data will slow down the planning process and prevent people from making effective decisions in a timely manner. Good S&OP practice defines the minimum amount of data needed to make certain decisions and focuses on getting that data quickly and guaranteeing its accuracy.
Chapter Summary
One of the most common dynamics in supply chains is a phenomenon that has been dubbed “the bullwhip effect.” What happens is that small changes in product demand by the consumer at the front of the supply chain translate into wider and wider swings in demand, as experienced by companies further back in the supply chain. Companies at different stages in the supply chain come to have very different pictures of market demand and the result is a breakdown in supply chain coordination. Companies behave in ways that at first create product shortages and then lead to an excess supply of products.
Many companies are not aware of the cost of the bullwhip effect on their supply chains. Traditionally, difficulties in predicting demand caused by the bullwhip effect were taken as a given and companies worked on their own to develop better capabilities to respond to these hard-to-predict fluctuations in demand. It may instead be far more efficient for companies to work together to actually reduce the fluctuations in demand or more accurately predict those changes in demand. A company can use CPFR to better forecast changes in demand and supply and it can use S&OP to optimize its individual response to fluctuating demand and supply conditions.
The GDSN is a network of interoperable databases that allows all parties in a supply chain to continuously update and request data about different products. The GDSN is administered by the GS1 organization. GS1 is a global not-for-profit organization devoted to setting standards and providing for efficient data transfer between all parties in global and regional supply chains. Efficient transfer of up-to-date and accurate information is a basic requirement for supply chain coordination.