ABSTRACT
Forecasts of future demand provide the basis for supply chain decision making. For a retailer, customer demand information is needed to make replenishment decisions. To forecast demand, one must first identify the factors that influence future demand and then understand the relationships between these factors and future demand. Causal forecasting methods assume that the demand is highly correlated with certain factors. Forecasting methods can be classified into four categories: qualitative, causal, time series, and simulation. Given the forecasted rainfall in the future month, one can use the relationship to forecast the demand for umbrella in the same month. Time series forecasting methods assume that past demand is a good indicator of future demand. Sales and marketing personnel are good examples of “experts” for forecasting the demand of a new product. Simulation forecasting methods mimic consumer behaviors that give rise to demand to arrive at a forecast.
