Strategic Forecasting in Energy Management: Integrating ANN and Exponential Smoothing Techniques for Business Innovation and Advancing UNSDGs

Forecasting electricity consumption with high precision is imperative for optimizing energy distribution, promoting sustainable energy policies, and ensuring economic efficiency. This study undertakes a comparative examination of two forecasting methodologies, namely Exponential Smoothing and Artificial Neural Networks (ANN), to assess their effectiveness in predicting electricity consumption within the United Kingdom for the period spanning from 2000 to 2020. We employed robust statistical measures, namely Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and the Correlation Coefficient (R), to assess the accuracy of the forecasts produced by each method. All performance measures showed that ANNs were faster than Exponential Smoothing. A more accurate and consistent prediction capability was demonstrated by the ANN model, which achieved a significantly lower root-mean-square error (RMSE) of 5.06 vs 6.86 for Exponential Smoothing, a MAPE of 0.92% vs 5.71%, and an R-value of 0.93, which is superior. These findings demonstrate the promise of ANNs in improving forecast accuracy by accounting for the complicated, nonlinear interdependencies inherent in data on electricity consumption. The study recommends that ANNs are better able to handle the complexities of modern energy consumption patterns and give a significant improvement in predicted accuracy compared to Exponential Smoothing, which is still a faster and less computationally demanding method. Although context-specific demands, such as the availability of historical data and computational resources, should drive the final selection of a forecasting approach, the results do support integrating ANN models into energy system strategic planning. In addition to supporting sustainable development’s overarching goals, this study offers a thorough comparison of forecasting approaches. This study’s immediate contribution to SDG 7 Affordable and Clean Energy is an improvement in the reliability of power consumption predictions. If predictions are accurate, we can better distribute energy, cut down on waste, and speed up the switch to renewable power. The study's findings also pertain to SDG 13: Climate Action, since better energy management and efficiency are critical to lowering emissions of greenhouse gases. Hence, ANNs and other advanced forecasting methods are vital for both short- and long-term operational needs, as well as for fostering ecological and economic resilience.