Why does a business need to plan demand? The answer to this question is found in the economic uncertainty, risk, and mitigation graphic (see Figure 5.1):
Figure 5.1 Economic uncertainty, risk and mitigation
The demand plan is a realistic view of future sales based on all known activities and trends. It’s used by sales and marketing to focus on or change the commercial direction of the business. It is also a formal request for the supply chain to make the relevant materials and schedule capacity for anticipated customer requirements (see Figure 5.2). A demand plan is also a financial commitment made by the business for top-line revenue and bottom-line margin.
Figure 5.2 SIOP: a consistent cycle of events
At its most basic, a demand plan is a realistic view of future sales based on all known activities and trends. It is also the main business process that anticipates and acts on a changing business outlook. It consists of:
•A review of the product portfolio, understanding lifecycle and implications
•Planning for future demand, considering risks and opportunities
•Developing a supply response and the ability to react to changes
•Reconciling all plans with financial assessment of risk and trade-offs
•Team decisions presented to executives for review and agreement.
The following definitions are used in planning demand:
Demand is what customers would buy if they could. This is unconstrained demand.
Demand forecast is a projection into the future of expected demand, given a stated set of environmental assumptions. This is an unconstrained forecast.
Outputs are the managerial actions that result from the balancing of demand with a supply and an operational, demand, and revenue plan.
Then, does the business have the resources and training to execute the plan? The process for one business might not work in another.
The only function of economic forecasting is to make astrology look respectable.
—John Kenneth Galbraith, U.S. (Canadian-born) Economist and Author (1908–2006)
I once attended a supply chain forum at the University of Tennessee, where the president of a very well-known consumer electronics company said, “My sales force turned out to be the world’s worst forecasters, but the world’s best adjusters.”
I always thought this was a great statement on the art of forecasting. The goal of the forecasting process is to provide an ongoing, sustainable 12- to 18-month forward-looking forecast for the business using the collaboration of sales and product management knowledge based on historical actuals. It also uses the future expectations of existing and new customers for both existing and new product listings.
The objective is to provide an accurate forecast of unconstrained product demand by listing product quantity with 12- to 18-month visibility and to overlay the forecast with market intelligence, trends, and exceptions.
As indicated in the preceding, a forecast is the basis for all planning decisions in a supply chain and is used for both push and pull processes*.
Forecasts are also used for:
•Production scheduling, inventory, and aggregate planning
•Sales force allocation, promotions, and new production introduction
•Plant/equipment investment and budgetary planning
•Workforce planning, hiring, and layoffs.
All decisions of this process are interrelated. Accuracy, honesty, and integrity are essential in creating a forecast demand plan. Forecasts are always inaccurate and should thus include both the expected value of the forecast and a measure of forecast error. Long-term forecasts are usually less accurate than short-term forecasts, and aggregated forecasts are usually more accurate than disaggregated forecasts. In general, the farther up the supply chain a company is, the greater the distortion of information it receives.
•Lead time of product replenishment
•Planned advertising or marketing efforts
•Planned price discounts
•State of the economy
•Actions that competitors have taken.
The components and methods of forecasting include these factors:
•Qualitative—are primarily subjective and rely on judgment
•Time series—use historical demand only, and are best used with stable demand
•Causal—involve the relationship between demand and some other factor
•Simulation—imitate consumer choices that give rise to demand. This can include forecast error, the difference between the forecast and actual demand.
Time series components look backward at historical demand to try to identify future patterns by examining these three components:
•Continuing pattern of demand increase or decrease
•Pattern can be a straight line or a curve
•Repeating pattern of demand increases or decreases
•Normally think of seasonality as occurring within a single year, and cycles as occurring over longer than one-year periods
•That part of demand history which the other time series components cannot explain.
•Forecasting level: At what level of granularity is the forecast expressed (stock-keeping unit (SKU), product, family, etc.)?
•Forecasting horizon: How far out into the future is demand forecast?
•Forecasting interval: How frequently is the forecast updated?
•Forecasting form: How is the forecast expressed? In units? Weight? Dollars?
The Figure below shows how the sales forecasting system fits into the entire process.
More important points to remember with respect forecasting are as follows:
•You need a data warehouse (see above figure) to guarantee data integrity
•Your forecasting system must integrate seamlessly with other corporate systems
•Forecasting systems are not the answer.
Demand planners input information at the level they know and take it out at the level they need. Most basic forecasting software will have many algorithms embedded in it. Even so, the analyst should understand the idea behind the time series techniques (Figure 5.3).
Figure 5.3 The forecasting hierarchy has three separate “faces”
One tool, regression analysis, is useful when:
•You think there are measurable factors that affect demand
•Demand is your dependent variable
•These measurable factors are your independent variables (which can either be internal or external).
Regression analysis can calculate the lift that occurs when these promotional activities are implemented. Understanding this lift is important for both strategic and operational forecasting.
Examples of internal factors include promotional and advertising spending, pricing changes, number of salespeople, and number of distribution outlets.
Examples of external factors can be leading indicators or simultaneous indicators and include the following:
•Average work week
•Manufacturers’ new orders
•Plant and equipment purchases
•Corporate profits after taxes
•Index of stock prices.
•Index of help wanted
•Index of industrial production
•Gross national product
•Index of wholesale prices.
Regression analysis using those types of variables is most appropriate for more long-term forecasting. There is enormous benefit from looking in the re