ABSTRACT

A life span forecasting model for KHS DMG-VF84 bottling line system was developed in this study to predict life span for a preventive maintenance planning. A Linear multiple regression analysis model and as well as the polynomial (second order) model were used. Eight independent variables were used for the model, but six key predictors were significant in developing the regression model and they are Availability (A), Reliability (R), Mean time to Repair (T), Failure rate (F), Operational time (O) and Mean time before failure (B).The eliminated variables were lost hour (H) and expected run time (E), implying that lost hour and expected run time did not significantly impact to Life span. The first and second model utilized linear multipleregression and second order polynomial respectively of Life span of the KHS DMG-VF84bottling line system as its dependent variables. The test data showed that the mean absolute percentage error for the first model is 8.3% and has the ability to predict Life span for bottling line with a good degree of accuracy of 75.68% with ± 0.24% error and the coefficients of determination R2for the developed first model is 0.76. This indicates that the relationship between the dependent variable and the independent variables of the developed model is good and the predicted values from a forecast model fit with the real-life data. The test data also showed that the mean absolute percentage error for the second model is 7.5% and has the ability to predict Life span for bottling line with a good degree of accuracy of 79.65% with ± 0.20% error and the coefficients of determination R2for the developed second model is 0.7965. This indicates that the relationship between the dependent variable and the independent variables of the developed second model is good and the predicted values from a forecast model fit with the real-life data as well. The average accuracy percentage for the first model is 91.7% while that of the second model is 92.5%, which shows that the second model is more accurate than the first one.This study established contributions to knowledge in the area of enhancing budget and control for production in a manufacturing company and support prompt and effective decision making.