Page 84 - DMGT523_LOGISTICS_AND_SUPPLY_CHAIN_MANAGEMENT

Unit 4: Demand Planning and Forecasting

Supply and demand reflect the time dimension. It is important to recognize that both supply and Notes
demand can be influenced by management actions. In business and economics, forecasting has
various meanings. There are two distinct quantities involved in forecasting, a forecast and a
prediction. A prediction is a broader concept. It is an estimate of a future event achieved through
subjective considerations other than just past data; this subjective consideration need not occur
in any predetermined way.
In supply chain management, we adopt a rather specific definition of a forecast, which is given
below:
A forecast is an estimate of a future event achieved by systematically combining and casting
forward in a predetermined way data about the past.

The supply chain has both space and time dimensions. That is, the supply chain professional
must know where demand volume will take place as well as when it will take place. Spatial
location of demand is needed to plan warehouse locations, balance inventory levels across the
supply chain network, and geographically allocate transportation resources.
The nature of demand can differ greatly, depending on the operations of the firm and the
activity for which the forecast is required. There are two types of demand.

Notes The first is when demand is generated from many customers, each of whom
purchases only a small fraction of the total volume. This type of demand is said to be
independent. The second type of demand comes into play when the demand is derived
from a production schedule. This type of demand is said to be dependent.

Independent demand uses statistical forecasting techniques. These models are based on
independence and randomness of demand. In contrast, the demand is known in the case of
dependent demand.

4.1.1 Forecasting Methods

Different forecasting methods can be used to develop the forecast. The appropriate method will
depend on the nature of the item being forecast and the availability of historical data. These are
two factors that often determine the method you choose to form the forecast.

There are four common approaches to forecasting which are given below:
 Qualitative: These forecasts are used where there is little or no historical performance
data to determine demand. They are typically based on an expert’s familiarity of products,
the industry and customer preferences. An expert’s opinion is usually useful when new
products are being introduced into the market.

 Time Series: Time series forecasts rely on historical demand in order to predict the future
demand. There are a variety of computational methods that can be used. Usually, this
method is ideal for items that have a generally defined historical pattern that does not
change radically from one year to the next e.g. “staple stock” items in a retail store.
 Causal: Causal forecasting is used when there is a visible correlation between one or
more variables to the demand for the product. For example, disposable income, lifestyle
indicators, etc. may be used to determine the demand for many consumer durable items.
The method, however, requires a high level of sophistication in modelling.
 Simulation: This method is highly sophisticated and is mainly used where the organization
needs to generate multiple ‘what-if’ scenarios. For example, such a model would be able to

LOVELY PROFESSIONAL UNIVERSITY 79