Simple Moving Average
In a business, when the demand of a given product does not rapidly grow, decline nor have seasonal variations, the moving average is the most helpful strategy in removing the random variations in its forecasting. The simple moving average is determined through the calculation of the average annual sales of a given product over a certain period of time (Nadler & Kros, 2007). Often, it is based on using the closing sales. The longer the duration taken to be used in the forecasting, the smoother the final effect. A longer period should also be chosen in case the data in hand is considered to be fairly constant with some substantial randomness ( Makridakis et al., 2008) . The period used in calculating the moving average is consistent. Some of the limitations to using this strategy is that it does not take into account any data beyond the period of average. When an unadjusted moving average is used, it can result to an underlying seasonal variation.
From the calculations of Dafra Motors Manufacturers in excel, it can be noted that the average of a given past period is used as the forecast for the next period. It is called the moving average since it is the averages which move to the next period. In comparison with other techniques such as that of exponential moving average, this method is more simplified. In relation to the business, the forecast is done on a monthly basis.
Delegate your assignment to our experts and they will do the rest.
Exponential Smoothing Technique
A limitation to techniques such as the simple moving average, is that a large amount of historic data needs to be carried on to the next period. Whenever a new data is added, the old observation is dropped and not used in the computation. The exponential moving average helps in minimizing the lags created by applying more weight on the recent sales. The most recent occurrence can be considered to be a better indicator of the future than data in the distant past. This technique suggests that most data that diminishes in the past becomes more distant. In relation to this, exponential smoothing becomes the most logical strategy to use in forecasting data. It is referred to as the exponential smoothing because each increment of data is decreased by a (1 - α).
From the excel computations of the company, this technique involves an automatic weighing of the past data which then decreases exponentially with time. This means that the most recent forecast has a decreasing weight. The new forecast is the sum of the old forecast plus a certain proportion of the error forecasted. From the computations, it can also be noted that only three pieces of data is used in the forecasting process. The value of the forecast which is most recent, the actual demand or sales of the given product in the specific period and the smoothing alpha constant. The smoothing constant is used to determine the level of smoothing and the reaction speed towards the difference in the actual and forecasted occurrence. The value of this constant is determined by the product’s nature and the what is viewed to be an effective response rate ( Makridakis et al., 2008) .
Naive Method
This is the simplest technique method which uses the most recent observation as the forecast value for the next period. This implies that the other historic data is not used as part of the forecast (Nadler & Kros, 2007). This method is more helpful when using the time series data. The latest data value which is represented in the operations dataset is thus the new expected value for the nest period under consideration. This is well elaborated from the excel computations. From the data and computations on excel, a trend line is generated from the data. This line is elongated and used to predict the value of the next period. From the line drawn, the value of the 1991 sales can be estimated to be 119 after the elongation. This method is however not accurate when the historical data does not seem to be consistent.
Recommendations and Impact on Firm
From the above analysis, the exponential potential would be the best quantitative forecasting technique that can be used by Dafra Motors. This is because it is able to successfully eliminate various limitations that are posed by the other two forecasting techniques. The exponential method is the most accurate amongst the three methods. Despite the method being complex in the computation process, it is able to present a well calculated value of the forecasted data, since it considers various factors such as accuracy of the procedure and its feasibility when producing the forecast ( Mentzer & Bienstock, 1998). Additionally, this method is also more advantageous since it is well placed to minimize the lags which are present in the simple moving averages. The shorter durations used in the prediction process also helps in boosting the accuracy of the predictions. The most recent data value used in this tactic is a better indicator of the future as compared to previously distant data.
The forecasts will have various positive impacts on the firm. This for instance is during the decision making process. It will be in a position to know what it anticipates to generate as revenue for a given future period. This information will be important as it will help in the cost planning process ( Sanders & Manrodt, 2003). The management team will formulate strategies and policies that will guide the firm to achieving their future goals and deviate from what is forecasted. The firm will also be best placed to determine the costs that they can eliminate to achieve its financial goals.
References
Makridakis, S., Wheelwright, S. C., & Hyndman, R. J. (2008). Forecasting methods and applications . John wiley & sons.
Mentzer, J. T., & Bienstock, C. C. (1998). Sales forecasting management: understanding the techniques, systems and management of the sales forecasting process . SAGE Publications, Incorporated.
S. Nadler & Kros, John F. (2007). Forecasting with Excel: Suggestions for Managers, Spreadsheets in Education (eJSiE): Vol. 2. Retrieve from: http://epublications.bond.edu.au/ejsie/vol2/iss2/5
Sanders, N. R., & Manrodt, K. B. (2003). The efficacy of using judgmental versus quantitative forecasting methods in practice. Omega , 31 (6), 511-522.
