Towards smart and sustainable cities: integrating ARIMA-ARNN and IoT for public lighting management - a Romanian town case study

The efficient management of public lighting systems is essential for reducing energy consumption and costs in small towns. This study proposes a hybrid forecasting model that combines the strengths of the Autoregressive Integrated Moving Average and Artificial Neural Network techniques to predict the energy consumption of public lighting in a small town in Romania for 12 months. Leveraging data on the energy usage of the town's public lighting system, the study uses the ARIMA-ARNN hybrid model to provide accurate and reliable forecasts that can assist town managers in budgeting for their energy needs and planning for IoT-enabled smart lighting systems implementation. The findings demonstrate that the hybrid model outperforms traditional individual models in terms of predictive accuracy, making it a valuable tool for energy consumption forecasting and management in the public lighting sector.