Demand Forecasting

Why Is forecasting necessary?

 Material/
Knowing
Inventory
Demand
Manpower Needed
Requirement

Facilities/
Warehouse
Machine and (Space) needed
other resources
 Characteristics of good information

• Information must be accurate

• Information must be accessible in a timely fashion
• Information must be of the right kind
 Value of Information

Information
• Helps reduce variability
• Helps improve forecasts
• Enables coordination of systems and strategies
• Improves customer service
• Facilitates lead time reductions
• Enables firms to react more quickly to changing market conditions
 Demand Forecasting
• Forecasting is an important element of demand management
• It provides an estimate of future demand and the basis for planning and sound business decisions
• The goal of a good forecasting technique is to minimize the deviation between actual demand and the
forecast
• Buyers and sellers should share all relevant information to generate a single consensus forecast so
that the correct decisions on supply and demand can be made
• Improved forecasts benefit not only the focal company but also the trading partners in the supply
chain
• The benefits of better forecasts are lower inventories, reduced stockout, smoother production plans,
reduced costs and improved customer service
 Forecasting Techniques - Qualitative
Methods

• Qualitative forecasting methods are based on intuition or judgmental evaluation and are
generally used when data are limited, unavailable, or not currently relevant

• Often when current data is no longer very useful, and for new product introductions when
current data does not exist

• This approach can be very low cost, its effectiveness depends to a large extent on the skill
and experience of the forecaster
 Forecasting Techniques - Qualitative
Methods

Jury of Executive Opinion

• A group of senior management executives who are knowledgeable about the
market, their competitors and the business environment collectively develop the
forecast
• The challenge is if one member’s views dominate the discussion
• This technique is applicable for long-range planning and new product
introductions
 Forecasting Techniques - Qualitative
Methods
Delphi Method
• A group of internal and external experts are surveyed during several rounds in terms of future events and
long-term forecasts of demand
• The answers from the experts are accumulated after each round of the survey and summarized
• The summary of responses is then sent out to all the experts in the next round, wherein individual experts
can modify their responses based on the group’s response summary
• This iterative process continues until a consensus is reached.
• Time-consuming and very expensive.
• This approach is applicable for high-risk technology forecasting; large, expensive projects; or major, new
product introductions.
• The quality of the forecast depends largely on the knowledge of the experts.
 Forecasting Techniques - Qualitative
Methods
Sales Force Composite
• The sales force represents a good source of market information
• This type of forecast is generated based on the sales force’s knowledge of the market and estimates of customer needs
• Due to the proximity of the sales personnel to the consumers, the forecast tends to be reliable, but individual biases
could negatively impact the effectiveness of this approach

Consumer Survey
• A questionnaire is developed that seeks input from customers on important issues such as future buying habits, new
product ideas and opinions about existing products
• The survey is administered through telephone, mail, Internet, or personal interviews
 Forecasting Techniques - Quantitative
Methods
• Quantitative forecasting models use mathematical techniques that are based on
historical data and can include causal variables to forecast demand
• Time series forecasting is based on the assumption that the future is an extension of
the past; thus, historical data can be used to predict future demand
• Cause-and-effect forecasting assumes that one or more factors (independent
variables) are related to demand and, therefore, can be used to predict future demand
• Since these forecasts rely solely on past demand data, all quantitative methods
become less accurate as the forecast’s time horizon increases
 Quantitative Methods - Components of
Time Series
• Trend Variations - Trends represent either increasing or decreasing
movements over many years, and are due to factors such as population
growth, population shifts, cultural changes and income shifts

• Cyclical Variations - These variations are wavelike movements that are

longer than a year and influenced by macroeconomic and political factors.
 Quantitative Methods- Components of
Time Series
• Seasonal Variation shows peak and valley that repeat over a consistent
interval such as hourly, daily, weekly, monthly or season. Due to
seasonality, many companies do well in certain months and not so well in
other months. For example, snow blower sales , fast-food restaurant
• Random Variations are due to unexpected or unpredictable events such
as natural disasters(hurricanes,flood,strikes,war).Volcano eruption in
Iceland causing shut down of flights in Europe due to Ash clouds
 Time Series Forecasting Models
• Time series forecasts are dependent on the availability of historical data.
• Forecasts are estimated by extrapolating the past data into the future.
• A survey of forecasting models used shows that time series models are the
most widely used
• The study also finds that within the time series models, the ones most
commonly used are the simple models (averages and simple trend) and
exponential smoothing.
 Time Series Forecasting Models
Naïve Forecast
• Using the naïve forecast, the estimate for the next period is equal to the actual
demand for the immediate past period
• Ft +1 =At
• where Ft+1 = forecast for period t+1;
• At = actual demand for period
• Inexpensive to understand, develop, store data and operate
• No consideration of causal relationships and the method may not generate
accurate results
 Time Series Forecasting Models
The Simple Moving Average forecast

• The Simple Moving Average forecast uses historical data to generate a

forecast and works well when the demand is fairly stable over time
• The advantage of this technique is that it is simple to use and easy to
understand.
• A weakness of the simple moving average method is its inability to
respond to trend changes quickly.
 Time Series Forecasting Models
Weighted Moving Average Forecast

• The simple moving average forecast places equal weights (1/n) on each of
the n-period observations. Under some circumstances, a forecaster may
decide that equal weighing is undesirable
• Weights used tend to be based on the experience of the forecaster and this
is one of the advantages of this forecasting method.
 Time Series Forecasting Models
Exponential Smoothing Forecast
• The exponential smoothing forecast is a sophisticated weighted moving
average forecasting technique in which the forecast for next period’s
demand is the current period’s forecast adjusted by a fraction of the
difference between the current period’s actual demand and forecast
• Most widely used forecasting technique
 Linear Trend Forecast
• A linear trend forecast can be estimated using simple linear regression to fit a line to a series of
data occurring over time
• This model is also referred to as the simple trend model
• The trend line is determined using the least squares method, which minimizes the sum of the
squared deviations to determine the characteristics of the linear equation.
• The trend line equation is expressed as
• Ŷ = b0 + b1x Where Ŷ = forecast or dependent variable; x = time variable
• b0 = intercept of the vertical axis
• b1 = slope of the trend line
 Cause-and-Effect Models
• The cause-and-effect models have a cause (independent variable or
variables) and an effect (dependent variable)
• One of the more common models used is regression
• Simple Linear Regression Forecast - When there is only one
explanatory variable, we have a simple regression forecast equivalent to
the linear trend forecast described earlier. The difference is that the x
variable is no longer time but instead an explanatory variable of demand
 Forecast Accuracy
• The ultimate goal of any forecasting endeavor is to have an accurate and unbiased
forecast
• The costs associated with prediction error can be substantial and include the costs of lost
sales, safety stock, unsatisfied customers and loss of goodwill
• Forecast error is the difference between the actual quantity and the forecast. Forecast
error can be expressed as:
• et = At − Ft
• where et = forecast error for period t; At = actual demand for period t; Ft = forecast for
period t.