Demand Forecasting
Forecasting is an important element of demand management. It provides forecasts of future demand
and a basis for sound business planning and decisions. Because all organizations face an uncertain
future, some errors exist between forecasts and actual demand. Thus, the goal of a good forecasting
technique is to minimize the deviation between actual and forecast demand. Since forecasts are
predictions of the future, the factors affecting demand must be considered in developing an accurate
forecast. In addition, buyers and sellers must share all relevant information to come up with a single
consensus forecast so that informed decisions about supply and demand can be made. Having
accurate demand forecasts allows the purchasing department to order the right amount of product,
the operations department to produce the right amount of product, and the logistics department to
deliver the right amount of product. Inaccurate estimates will cause an imbalance in supply and
demand. In today's competitive business environment, collaboration (or cooperation and sharing of
information) between buyers and sellers is indispensable. The benefits of better forecasting are lower
inventories, reduced stockouts, production planning, reduced costs, and improved customer service.
Many would argue that demand forecasting is both an art and a science. The impact of poor
communication and inaccurate forecasting reflects across the supply chain and leads to stockouts, lost
sales, inventory costs, high obsolescence, material shortages, unresponsiveness to market dynamics,
and poor profitability. Many examples exist that demonstrate the problems companies face when
sales forecasts do not match customer demands during new product introductions.
Forecasting Techniques
An inaccurate forecast doesn't mean that nothing can be done to improve the forecast. Both
quantitative and qualitative forecasts can be improved by seeking input from trading partners.
Qualitative forecasting methods are based on opinion and intuition, whereas quantitative forecasting
methods use mathematical models and relevant historical data to generate forecasts. Quantitative
methods can be divided into two groups: time series and associative models.
1. Qualitative Method
Qualitative forecasting methods are based on intuitive evaluation or judgment and are
generally used when data are limited and data are not available. This approach can be very
low-cost, but its effectiveness depends heavily on the skill and experience of the forecaster
and the amount of relevant information available. Qualitative techniques are often used to
develop remote projections when rental data is no longer very useful and for new product
introductions when data is lacking. The following are four qualitative forecast models:
a) Executive Opinion Jury: A group of senior management executives knowledgeable
about the market, competitors, and business environment collectively develop the
forecast. This technique has the advantage of involving several individuals with fairly
joint work experience, but if one member's view dominates the discussion, the value
and reliability of the results can be reduced. This technique is applicable to long-range
planning and new product introductions. High-fashion forecasting, for example, is a
risky business because there is often no theoretical basis for generating forecasts.
b) Delphi method: A group of internal and external experts is surveyed over a period of
time against long-term demand forecasts. Group members did not physically meet
and avoided scenarios where one or a few experts could dominate the discussion.
Answers from experts are accumulated after each survey round and summarized. A
summary of the responses is then sent to all experts, where they can modify their
responses based on the group. Summary of responses This process continues until a
, consensus is reached. The process can be time-consuming and very expensive. This
approach is applicable to high-risk technology forecasting, large and expensive
projects, or new product introductions. The quality of forecasts is highly dependent
on the knowledge of experts.
c) Sales Force Composite: The sales force is a good source of market information. This
type of forecast is generated based on the salesperson's knowledge of the market and
forecasts of customer needs. Because of the salespeople's proximity to the customer,
their forecasts tend to be reliable, but individual bias can negatively affect the
effectiveness of this approach. For example, if bonuses are paid when actual sales
exceed estimates, there is a tendency for salespeople to under-estimate.
d) Consumer Surveys: Questionnaires were developed that solicit customer input on
important issues such as future buying habits, new product ideas, and opinions on
existing products. Surveys are conducted by telephone, mail, internet, or personal
interviews. The data collected from the surveys was analyzed using statistical and
scoring tools to derive a set of meaningful results. The challenge is to identify a sample
of respondents that is representative of the larger population and to obtain an
acceptable response rate.
2. Quantitative Method
Quantitative forecasting models use mathematical techniques that are based on historical
data and can include causal variables to forecast demand. Because these forecasts rely solely
on past demand data, all quantitative methods become less accurate as the time horizon
estimates increase. Thus, for the long term, it is generally recommended to utilize a
combination of quantitative and qualitative techniques.
• 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 results in one or more factors (independent variables)
related to demand and therefore can be used to predict future demand.
Forecasting is an important element of demand management. It provides forecasts of future demand
and a basis for sound business planning and decisions. Because all organizations face an uncertain
future, some errors exist between forecasts and actual demand. Thus, the goal of a good forecasting
technique is to minimize the deviation between actual and forecast demand. Since forecasts are
predictions of the future, the factors affecting demand must be considered in developing an accurate
forecast. In addition, buyers and sellers must share all relevant information to come up with a single
consensus forecast so that informed decisions about supply and demand can be made. Having
accurate demand forecasts allows the purchasing department to order the right amount of product,
the operations department to produce the right amount of product, and the logistics department to
deliver the right amount of product. Inaccurate estimates will cause an imbalance in supply and
demand. In today's competitive business environment, collaboration (or cooperation and sharing of
information) between buyers and sellers is indispensable. The benefits of better forecasting are lower
inventories, reduced stockouts, production planning, reduced costs, and improved customer service.
Many would argue that demand forecasting is both an art and a science. The impact of poor
communication and inaccurate forecasting reflects across the supply chain and leads to stockouts, lost
sales, inventory costs, high obsolescence, material shortages, unresponsiveness to market dynamics,
and poor profitability. Many examples exist that demonstrate the problems companies face when
sales forecasts do not match customer demands during new product introductions.
Forecasting Techniques
An inaccurate forecast doesn't mean that nothing can be done to improve the forecast. Both
quantitative and qualitative forecasts can be improved by seeking input from trading partners.
Qualitative forecasting methods are based on opinion and intuition, whereas quantitative forecasting
methods use mathematical models and relevant historical data to generate forecasts. Quantitative
methods can be divided into two groups: time series and associative models.
1. Qualitative Method
Qualitative forecasting methods are based on intuitive evaluation or judgment and are
generally used when data are limited and data are not available. This approach can be very
low-cost, but its effectiveness depends heavily on the skill and experience of the forecaster
and the amount of relevant information available. Qualitative techniques are often used to
develop remote projections when rental data is no longer very useful and for new product
introductions when data is lacking. The following are four qualitative forecast models:
a) Executive Opinion Jury: A group of senior management executives knowledgeable
about the market, competitors, and business environment collectively develop the
forecast. This technique has the advantage of involving several individuals with fairly
joint work experience, but if one member's view dominates the discussion, the value
and reliability of the results can be reduced. This technique is applicable to long-range
planning and new product introductions. High-fashion forecasting, for example, is a
risky business because there is often no theoretical basis for generating forecasts.
b) Delphi method: A group of internal and external experts is surveyed over a period of
time against long-term demand forecasts. Group members did not physically meet
and avoided scenarios where one or a few experts could dominate the discussion.
Answers from experts are accumulated after each survey round and summarized. A
summary of the responses is then sent to all experts, where they can modify their
responses based on the group. Summary of responses This process continues until a
, consensus is reached. The process can be time-consuming and very expensive. This
approach is applicable to high-risk technology forecasting, large and expensive
projects, or new product introductions. The quality of forecasts is highly dependent
on the knowledge of experts.
c) Sales Force Composite: The sales force is a good source of market information. This
type of forecast is generated based on the salesperson's knowledge of the market and
forecasts of customer needs. Because of the salespeople's proximity to the customer,
their forecasts tend to be reliable, but individual bias can negatively affect the
effectiveness of this approach. For example, if bonuses are paid when actual sales
exceed estimates, there is a tendency for salespeople to under-estimate.
d) Consumer Surveys: Questionnaires were developed that solicit customer input on
important issues such as future buying habits, new product ideas, and opinions on
existing products. Surveys are conducted by telephone, mail, internet, or personal
interviews. The data collected from the surveys was analyzed using statistical and
scoring tools to derive a set of meaningful results. The challenge is to identify a sample
of respondents that is representative of the larger population and to obtain an
acceptable response rate.
2. Quantitative Method
Quantitative forecasting models use mathematical techniques that are based on historical
data and can include causal variables to forecast demand. Because these forecasts rely solely
on past demand data, all quantitative methods become less accurate as the time horizon
estimates increase. Thus, for the long term, it is generally recommended to utilize a
combination of quantitative and qualitative techniques.
• 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 results in one or more factors (independent variables)
related to demand and therefore can be used to predict future demand.
