Accurate forecasting of electricity consumption is essential for energy planning and policy-making in rapidly developing economies. Therefore, this research aims to propose a new hybrid ARIMA-ANN electricity demand forecasting model for Bahrain that can combine the advantages found in methods for capturing the linear patterns of autoregressive integrated moving average (ARIMA) models with the ability to model nonlinear relationships typically identified in artificial neural networks (ANN). Using historical consumption data from 1980 to 2023, the results from the hybrid model indicate a very low error in predictions (R2 = 1.0), which indicates that the model can explain the underlying patterns of electricity consumption exceptionally well (RMSE = 1.57, MAPE = 11.4%). In terms of model convergence, the hybrid model achieved convergence at epoch 4, demonstrating computational efficiency and strong generalization capacity. The hybrid approach was successful in outperforming both the ARIMA and ANN models utilized and provided the best fit for the first step in forecasting the complex, nonlinear, growing electricity consumption patterns demonstrated in the region of study. Additionally, these findings showcase the model's strengths in overcoming the boundaries of energy forecasting that previous studies have suggested are crucial (non-stationarity, volatility). This research has important implications for energy planners and policymakers, as a finalized hybrid ARIMA-ANN model provides a reliable option for future planning of energy infrastructure and network management and deals with persistently rapid growth in sustainable energy plans for the public and private sectors. The findings of the hybrid ARIMA-ANN model contribute to the limited literature on hybrid modeling methods for energy forecasting in the Gulf Cooperation Council (GCC) and provide an adaptable methodological framework for similar countries. The model allows for optimization of generation planning, improved demand-side management, and enhancements to renewable energy integration into the electricity supply chain. While the findings from the hybrid ARIMA-ANN model yielded strong results, additional research is warranted to explore additional variables to provide additional validation, as well as test the goodness of fit applied in real-time procedures to fully understand operational robustness. Thus, this research sets the foundation for a novel hybrid ARIMA-ANN electricity demand forecasting model that can be utilized in rapidly urbanizing, economically expanding nations. Studying and developing a hybrid model in Bahrain exemplifies a process that can be replicable for consideration in other energy-reliant economies that desire to follow a development directive while simultaneously managing sustainable energy needs. Future possibilities include additional research that examines exogenous variables and develops adaptive learning for energy consumption patterns that change over time.