Demand Assessment

 Problem

 Identification of forecasting techniques which can predict the demand for existing and new products.

 Issues

 Ease of using forecasting process

 Reliability of the sales forecast

 Impact of occasional price promotions

 Use of economic information to facilitate forecasts

 Methods to forecast demand of new product

Rationale:

 The demand data for PVB valve displays trend and seasonality so we use Time Series Analysis & Projection- Moving Average to forecast future demand.

 Summary of results

o Forecast for 1st Quarter'2005 = 45,182 units, current forecasting methods yield forecast value of 53,560 units.

o Actual sales = 48,159 units

 The minimum sample size data required for conducting regression analysis is 25, since the regression was conducted on sample size of 13, the relationship between the dependent variable and predictor variables needs to be re-checked

 Relationship with other predictor variables such as advertising expense, price promotions and other economic indicators should be established to check if they account for fluctuation in the dependent variable, Quarterly sales.

 The sales data for Fire Valves shows very random trend and hence moving average method would be most appropriate for forecasting here also.

 Summary of results

 Forecast for 1st Quarter'2005 = 310 units, current forecasting methods yield forecast value of 559 units. Actual sales = 580 units

Results of regression Analysis

o Regression between Quarterly sales and Unemployment Rate

 R2 = 0.35

 Adjusted R2 = 0.29

o Regression between Quarterly sales and Bank Prime Loan Rate

 R2 = 0.09

 Adjusted R2 = 0.00

o Regression between Quarterly sales and Housing starts

 R2 = 0.10

 Adjusted R2 = 0.02

o The values of R2 tells us that the percentage of fluctuation in the dependent variable, Quarterly sales explained by the predictor variables - Unemployment Rate, Bank Prime Loan Rate and Housing starts is low hence we discard causal forecasting for predicting demand