Demand Forecasting: Need, methods and technique of demand forecasting

Demand Forecasting: Need, methods and techniques of demand forecasting

1. Demand Forecasting By Prof Roma Gupta

2. Introduction A forecast is an estimate of future demand & provides the basis for a number of managerial decisions like: I n ve n t o ry m anag e m e n t Pricing F inan c i a l r e qu i r e m e n t Supp l y c ha i n m anag e m e n t Capacity planning/operations Good forecasting provides reduced inventories, costs, & stockouts, & improved production plans & customer service

3. D e m a n d F o r eca s t i n g : P u r p o se Production Scheduling : in order to have a proper schedule of production so as to check over and under production. Cost Control : if future demand is estimated with accuracy cost can be minimized by purchasing raw materials at lower price in advance. Sales Target: Sales targets can be fixed based on the demand scenario. Advertising Policy Decision: cold drinks are advertised more in summer than winter . Inventory Management: inventory has a cost. Hence, proper inventory can be planned depending on the demand forecast. Long Term Planning : like capacity augmentation, building new plant, hiring etc.

4. Demand Forecasting: Important points Purpose of Forecast : The purpose of forecast determines the choice of technique and accuracy required. Level of Forecast : There are three levels usually- firm, industry, and macro level T ime Ho r izon o f F o r e c a s t : T h e f o r ec a s t i n g c a n be done for short, intermediate and long period. Unit of Forecast : individual product or aggregate.

5. Common Characteristics of Forecasting Forecasts are rarely perfect Forecasts are more accurate for aggregated data than for individual items Forecast are more accurate for shorter than longer time periods. Forecast are more accurate for industry rather than firm.

6. Techniques of Demand Forecasting

7. Demand Forecasting: Survey Methods Survey Methods : These methods are highly subjective and usually used for short term forecasting. They can be further classified as: Consumer survey Opinion Poll

8. Demand Forecasting: Survey Methods Consumer Survey : Under this method a list of potential buyers would be drawn and each buyer will be approached and asked about their buying plans. This method is ideal and it gives firsthand information, but it is costly and difficult to conduct. This may be undertaken in three ways : Complete Enumeration – In this method, all the consumers of the product are interviewed . Sample survey - In this method, a sample of consumers is selected for interview. Sample may be random sampling or non random . End-use method – The demand for the product from different sectors such as industries, consumers, export and import are found out.

9. Demand Forecasting: Survey Methods Opinion Poll Methods : Under this methods the experts opinion is used to forecast the demand. The main tec hn i qu e s ca n be put und e r t w o h ea d s : E xp er t op i n i on M a r ke t S t ud i e s The important expert opinion methods are: 1. Opinion Survey method : This method is also known as Sales- Force –Composite method. Under this method, the company asks its salesmen to submit estimate for future sales in their respective territories. This method is more useful and appropriate because the salesmen are more knowledgeable about their territory.

10. Demand Forecasting: Survey Methods 2 . Experts’ Opinion Poll : In this method, the experts on the particular product whose demand is under study are requested to give their ‘opinion’ or ‘feel’ about the product. These experts, dealing in the same or similar product, are able to predict the likely sales of a given product in future periods under different conditions based on their experience. If the number of such experts is large and their experience-based reactions are different, then an average- simple or weighted –is found to lead to unique forecasts.

11. Demand Forecasting: Survey Methods 3. Delphi Method : It is a sophisticated statistical method to arrive at a consensus (a general agreement) . This method employs a panel of experts in different locations who independently fill out a series of questionnaires. However, the results from each questionnaire are provided with the next one, so each expert then can evaluate this group information in adjusting his or her responses next time. The process is repeated until the responses converge to along a single line.

12. Demand Forecasting: Survey Methods Market Studies : Under this method a representative market is elected and the product is sold by varying price, advertisement etc. It can be of several types. It is mostly used for new products and is called test marketing.

13. Demand Forecasting: Statistical Methods It is used for long term forecasting. In this method, statistical and mathematical techniques are used to forecast demand. This method is relies on past data. This includes three methods: Trend Projection Barometric Methods Econometric Methods

14. Demand Forecasting: Statistical Methods Trent projection method : Under this method, demand is estimated on the basis of analysis of past data. This method makes use of time series (data over a period of time). Here we try to ascertain the trend in the time series. Three techniques can be used to project the trend: Graphical method Fitting Trend Equations Box Jenkins Method

15. Demand Forecasting: Statistical Methods Barometric method : This consists in discovering a set of series of some variables (indicators) which exhibit a close association in their movement over a period or time. The indicators can be of 3 types: Leading indicators Coincidental indicators Lagging indicators For example, the sale of tractors can be forecasted by the movement of agricultural income (AY series). The movement of AY is similar to that of ST, but the movement in ST takes place after a year’s time lag compared to the movement in AY.

16. Demand Forecasting: Statistical Methods 3. Econometric methods : These methods combine economic theories with mathematics and statistics to estimate the demand. It is mainly of two types: i. Regression methods: it can be bivariate or multivariate ii. S i m ulta n e o us eq u a ti o n m o del s : T hi s m e a ns t h e development of a complete economic model which will explain the behaviour of all variables which the company can control.