Demand forecasting is an assumption of demand in future. By using demand forecasting, a company makes suitable plans for upcoming challenges or demands and takes suitable action to tackle that them.
Demand forecasting can be divided into the following two major types −
Short run forecasting − is made to fulfill short-term targets, like preparation of suitable sales policies to increase the sales or proper planning for inventory as per the required demand.
Long run forecasting − is assumption made for long-term targets like planning of capital or assets.
Short run and long run demand forecasting is used as per the requirement of the enterprise. These forecasting types are explained in further section.
Following factors should be considered for assumption and fulfillment of short and long term demand forecasting.
Identifying the most relevant method for forecasting.
Predicting factors involved, which affect the demand of the product.
Acquiring the data about the factors that affect demand.
Finding the most suitable relation among independent variables and dependent variables.
Preparing the demand forecast and analyzing the results.
Demand forecasting can be accomplished by following the above steps.
The tools or methods used to forecast demand are of the following two types −
Quantitative techniques
Qualitative techniques
These techniques are used for both short run and long run forecasting; however, for short and long run forecasting, this method can further be sub divided as per forecasting type. The following are the tools for short-run forecasting −
This method is used to plot a trend in the demand. In this, average demand of different time frame is taken (for example, 2 years, 3years, etc.) for getting an assumption of future demand.
Example − Find the 3 yearly moving averages of the following −
Year | Production |
---|---|
1999 | 42 |
2000 | 46 |
2001 | 47 |
2002 | 39 |
2003 | 54 |
2004 | 65 |
2005 | 66 |
2006 | 60 |
Year | Production | 3 Yearly MT | 3 Yearly MA |
---|---|---|---|
1999 | 42 | − | − |
2000 | 46 | 136 | 45.33 |
2001 | 48 | 133 | 44.33 |
2002 | 39 | 141 | 47 |
2003 | 54 | 158 | 52.67 |
2004 | 65 | 185 | 61.67 |
2005 | 60 | − | − |
This method is mostly used for short-term forecasting. It is derived from moving average and modified. It is based on weighted averaged of observed value. It smoothens the trend where weighted value remains between 0 and 1.
St = W.Yt + (1-W). St-I [St= Current smoothened value (predicted)]
Yt = Current observed value.
W = weighted value or rate of trend.
Time series analysis is commonly used for long term demand forecasting. The following are some of its components −
Seasonal variation
Cyclical variation
Random variation
Irregular variation
To measure the components of time series, the following three methods are used −
Semi Average Method
Moving Average Method
Method of Least Square
These methods can be used for time series analysis as per demand forecasting requirement of an enterprise.
This method for demand forecasting is an analytical method. In this method, different methods of economics and mathematics are used to forecast the demand.
This method provides the liberty to assume multiple variables so it is more accurate in real business situations.
This method is based on the following criteria −
Demand for a product is based on several factors.
The determinants are independent variables but the demand is the dependent variable.
There is a constant interaction between demand and its determinants.
There is a constant interaction between the independent variables. The independent variables are divided into two types − Exogenous (non-economics) and Endogenous (economics).
This type of interaction can be estimated by statistical method. The forecast is divided into the set of linear or non-linear equations. These principles should be taken into consideration while using the econometrics method for demand forecasting.
Let us now discuss some of the qualitative techniques of Demand Forecasting −
In buying intention survey method, the survey is conducted on the product; several questions regarding the product are formulated. The participants are asked for reviewing/rating the product based on different criteria like taste, preference, cost, expectation, etc. These reviews are summarized and a report is prepared for consumer demand of the product.
In sales force opinion method, different territorial sales demands are collected to forecast the demand of a product. Then individual territory demand is combined to produce a final report of the market demand. This method is difficult to execute due to improper skill of salesmen. However, with appropriate skills, accurate predictions can be forecasted.