How OR can aid Management Decision Making Modern businesses have more need to predict future operations than those of the past do. Managers in large corporations have to summarise and analyse the various data available to them when making decisions. The U.K. OR Society defines the operational research decision-making techniques as a scientific model of the system, incorporating measurements of factors such as chance and risk, with which to predict and compare the outcomes of alternative decisions, strategies/ controls. The purpose of these techniques is to help management determine its policy and action scientifically. The models of OR are symbolic or abstract representations of real life problems.Examples of techniques that can be used by managers for use in decision making are for example, forecasting. Statistical forecasting is to an extent, an extension of the prediction of a dependent variable. A reasonably accurate forecast can be extremely valuable for a marketing or production strategy.Time series forecasting attempts to capture the past behaviour of the time series and uses this information to predict future values. No external predictors are considered. (Kvanli et al, chapter 17)The types of factors that determine the strengths of forecasting are the time horizon of the forecast; the stationarity of the data and the presence of trend, seasonality or cyclical activity. The accuracy of forecasting can be measured by calculating the MAD, MSE and the MAPE. These are useful for comparing the accuracy of a particular forecasting technique on two different time series.The advantage of forecasting is that there is no need to search for external predictors to explain the behaviour of dependent variables. The main disadvantage is that the observed values can be extremely complex and difficult to work out. Such methods are often hard to sell to managers who may not be able to understand the technique. However as mentioned earlier, if ...
