ABSTRACT
In this chapter, the authors discuss three time series methods of forecasting; namely smoothing, trend projection, and trend projection adjusted for seasonal influence. The main problem are concerned with when using time series analysis is how to identify the trend, seasonal variations, cyclical variations and random one-off causes in a set of data over a period. The moving average is an average taken at the end of each successive time period of the results of a fixed number of previous periods. The difference between the actual data and the predicted data is the residual. The predicted data is the sum of trend data and the seasonal variation. The residual shows how much actual data were affected by external factors other than the trend and seasonal variations. The trend line can be extrapolated to predict total lost time for the 6th week. This can be done either by summing the expected daily losses or using a weekly total plot.
