A few years ago, our research team worked with a company that manufactured and marketed women's lingerie. Like so many companies in this age of global supply chains, much of the manufacturing process had been moved offshore to Asian sewing operations, where sewing costs were so much lower than those costs would be in the United States.
Of course, this made this company's supply chain very long, and lead times of 100 days became commonplace. However, women's lingerie is, to a large degree, a fashion business with very unpredictable demand. This company would place orders with its Asian manufacturing partner for a particular style or color of merchandise, wait 100 days for that merchandise to arrive, only to find that it had "guessed wrong" about the market's enthusiasm for this product. Once they discovered that they had "guessed wrong," they had 100 days worth of merchandise in various places in the supply chain, and ultimately a lot of inventory that had to be heavily discounted because of lack of demand. If they "guessed wrong" in the other direction, then it would take at least 100 days to adjust colors or styles, and "catch up" to their competitors, who were riding the wave of fashion acceptance.
What this example illustrates is that in order for a company to effectively manage its supply chains, it must effectively understand -- and manage -- demand. There were two critical processes that the lingerie company had failed to develop: demand forecasting and demand/supply integration. In the nearly 30 years that the Sales Forecasting Research Team at the University of Tennessee has been working with companies, we often are astonished at how little attention is paid to the art and science of forecasting future demand. Often, when we dig into the details of a company's forecasting processes, we find that the entire process consists of asking the salespeople what they think they will sell next year -- a number that is usually influenced by their natural inclination to "manage expectations" (sometimes referred to as "sandbagging").
Another common forecasting disaster is when the forecast is simply determined by the firm's financial targets. In this scenario, the forecast is created to assure Wall Street that financial targets will be met, without regard to the existence (or lack of existence) of true demand in the marketplace. Excellence in forecasting future demand requires a disciplined process that combines analysis of historical demand (statistical, or quantitative forecasting) with careful analysis of expected changes in future demand patterns (judgmental, or qualitative forecasting). This excellence also requires a spirit of collaboration among those functional units that are responsible for generating and managing demand, as well as the discipline that comes from strong metrics and performance rewards. The lingerie company did not commit adequate resources - financial, human, or system resources -- to this critical process of demand forecasting, and the result was that dreaded combination of too much slow-moving inventory, and back-orders on popular merchandise.
But excellence in demand forecasting is not enough. Companies must also have a formal, disciplined process in place for Demand/Supply Integration, or DSI. This process, which sometimes is referred to as Sales and Operations Planning, or S&OP, is the mechanism by which information about future demand (the demand forecast), is matched up with information about future supply capability (the capacity forecast), and where both strategic and tactical decisions are made about how to bring supply and demand into harmony.
At a tactical level, short-term excess capacity might be addressed by increasing demand through promotional activity or expanded distribution. Short-term excess demand might be addressed through price increases or by curtailing planned advertising. At a more strategic level, this DSI process is critical for evaluating need for additional manufacturing capacity or for placement of new distribution centers. The key here is that an accurate demand forecast is of no value to the firm unless it is part of a formal, disciplined process for decision-making, and we refer to this process as DSI.
Now, let's return to the lingerie company and see how they addressed their supply chain problems. First, they dedicated resources to become more adept at demand forecasting. They implemented a sophisticated forecasting system that helped them analyze historical demand patterns. They hired and trained forecast analysts, who became skilled at analyzing that historical demand. They engaged both salespeople and their retail customers in providing valuable subjective information about changing demand patterns. They measured forecasting performance and rewarded improved accuracy. They stopped basing their forecasts on financial goals, but rather paid more attention to actual demand in the marketplace. They got better at forecasting.
But they did more. They used these better forecasts to make better supply chain decisions. They implemented an S&OP process which included strong engagement from senior marketing and supply chain executives. This S&OP process brought together demand and supply information, and allowed them to make good, data-driven decisions. By carefully analyzing historical demand, they discovered that many of their products had quite predictable demand patterns which lent themselves very well to the longer supply chains involved in their Asian sewing operations. But their "fashion" products, with much less predictable demand, were moved to sewing operations in Central America or Mexico, where the supply chains could be much shorter and more responsive. This company used the insights gained from careful analysis of demand to make good tactical and strategic decisions. The result? Inventories are down and customer fill rates are up.
Sound easy? It isn't. There is no silver bullet. Effecting this type of change requires considerable corporate soul-searching, and a whole lot of change management efforts to change processes, systems, and most importantly, corporate culture. But the payoffs can be huge. Excellence in supply chain management brings reduced costs, greater responsiveness to customers, and ultimately profitability, but such excellence is not possible with understanding, and managing, demand.
Mark A. Moon, Ph.D., is associate professor of marketing at the University of Tennessee. His research interests include forecasting, demand planning, sales and operations planning, and buyer/seller relationships. He is director of the University of Tennessee's Sales Forecasting Management Forum and has published extensively in both academic and practitioner journals. Moon teaches at the undergraduate, MBA, and executive MBA level and is actively involved in UT's Center for Executive Education.
For over 50 years, University of Tennessee (UT) faculty have played a major role in the supply chain/logistics arena -- conducting innovative research, publishing leading-edge findings, writing industry-standard textbooks, and creating benchmarks for successful corporate supply chain management. Programming is top-ranked in Supply Chain Management Review, U.S. News & World Report, and Journal of Business Logistics. Certification is available. http://SupplyChain.utk.edu www.bus.utk.edu/ivc