Forecasting Coal Consumption to Compare Biomass Ratios

This study examines sub-bituminous coal usage in the meat-processing industry to compare statistical forecasting models. To expand a coal-to-biomass action plan, it is necessary to forecast future coal demand. This analysis used monthly coal consumption data from January 2017 to September 2022, split 80:20 into two datasets. The first data set is used to develop forecast models such as seasonal naïve, exponential smoothing, and Box-Jenkins, while the second is used to compare forecast accuracy using the mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE). The findings revealed that time series data lack a trend component but indicate an additive seasonal pattern, and the appropriate model for the series includes the seasonal naïve model, the additive seasonal exponential smoothing model, and the Box-Jenkins seasonal model. The SARIMA(0,0,0)(1,0,0)₁₂ model was found to be the most accurate in forecasting coal consumption with an accuracy of 83.18 per cent. Furthermore, a comparative study of the initial properties of coal according to ASTM D2234/D2234M-10 standards, consisting of net calorific value, total moisture, and ash, revealed that biofuels alternative masses such as palm kernel shells, wood chips, and sawdust briquettes are examples of alternative masses have standard-compliant property values, and if used to replace coal as a fuel with a calorific value of 6,000 kilocalories, such biomass must be replenished by 1.36, 1.28, and 1.19 times more than coal, respectively. In addition, the ash generated by the combustion of palm kernel shells, wood chips, and sawdust briquettes with 32.8, 66, and 72 times the amount of coal, respectively. As a result, biomass is another viable fuel option for industrial applications. Each type of biomass' suitability can be weighed against other factors such as available resources, prices, and so on.