Quantitative Method:

These methods utilize mathematical models based on historical data considered relevant to the future. Many quantitative models, forecast accuracy, long range forecasts and short range forecasts are discussed in following paragraphs.

Forecast Accuracy

This is the closeness of forecasts to the actual data. Because of time gap between forecast and determination of actual data, the forecast accuracy gets determined after a long passage of time. Based upon the degree of closeness one may have a high accuracy or low accuracy.

Long Range Forecasts

This involves estimating future conditions over one year or more. These kinds of forecasts are essential to support strategic decisions about planning products, processes, technologies and facilities.

Cycles, Trends and Seasonality

From the data patterns in a long range forecasting, historical sales data appear to have several components which include trends, cycles, seasonality, and random fluctuation. Trends are upward or downward sloping line, cycles are the data patterns covering various years before its repetition and seasonality is a data pattern that repeats itself over a period of time. Random fluctuation is a pattern resulting from random variation or unexplained causes.

Linear Regression and Correlation

In this forecasting model establishment of a relationship among a dependent variable and independent variable is done. Suitable regression equation is utilized for the forecasting.

Short Range Forecasts

These forecasts are generally estimates of anticipated conditions over a short span of time. For these forecast cycles, seasonality and trends have little impact and enable the production managers to take following decisions:

1. Amount of inventory to be carried next month.

2. Amount of product to be scheduled for next week.

3. Amount of each of raw material to be ordered next week.

4. Number of employees to be scheduled to work on a straight time and overtime basis next week.