Page 92 - DMGT523_LOGISTICS_AND_SUPPLY_CHAIN_MANAGEMENT

Unit 4: Demand Planning and Forecasting

4.4.1 Time Series Notes

Time series is a characterization of change that takes place over a period of time. It is a quantitative
model that reflects the change in demand for goods and services and the pattern in the order of
occurrence, using historical data. These patterns or characteristics of the change process are
known as a ‘times series’.

One major theme in the continuing development of inventory theory is to develop inventory
models that are realistic and reflect product demand. In real life, demand is uncertain and hard
to forecast. Furthermore, as product life cycles get shorter, the randomness and unpredictability
of these demand processes have become even greater.
In practice, inventory managers often rely on forecasts based on a time series of prior demand,
such as a weighted moving average. Typically, these forecasts are predicted on a belief that the
most recent demand observations are the best predictors for future demand. Time series are,
therefore, commonly used for inventory decisions in order to:
 Generate and maintain forecasts at different levels of product,

 Provide appropriate forecasts for planning and replenishment for product and location,
and
 Optimize demand history through demand cleansing and seasonal profiling,
The method chosen will depend on the accuracy demanded of the forecast and the pattern of
historical demand. Some of the methods used to solve such problems are discussed below.

Moving Average Method

The simplest form of time series is the ‘moving average method’. In this type of model, the raw
data is converted into a moving average that reflects the trend in change of demand. The moving
average is an arithmetic average of data over a period of time. By averaging historical data, the
attempt is to remove the random fluctuations.

In this method, the data is updated regularly by replacing the item in the average by the new
item. This type of model is especially useful when demand has no pronounced trend or seasonal
influence. It is generally used to study this type of data, which is superior to the raw data because
it eliminates or smoothens out the irregularity in the time series.
The general formula for moving average is:
F = (A + A + A + A + ……+ A )/n
t + 1 t t – 1 t – 2 t – 3 t – n + 1
Where: F is the moving average for the period t + 1,
t + 1
A , A , A , A etc. are actual values for the corresponding period, and ‘n’ is the total number
t t – 1 t – 2 t – 3
of periods in the average.
For example, suppose the prices for a product are given for 12 months and a five monthly
average is to be computed. Each month sequentially designated as A , A , A , A , A .…………etc.
1 2 3 4 5
Then, the first 5-month moving average would be;

F = [(A + A + A + A + A )/5]
5 1 2 3 4 5
The second moving average of the next five months would be;
F = [(A + A + A + A + A )/5]
6 2 3 4 5 6
And so on.

LOVELY PROFESSIONAL UNIVERSITY 87