Enhancing street light fault detection in Smart Cities using Machine Learning and Deep Neural Network Approaches

In today’s smart city era, efficient management of infrastructure, including street lighting, is vital for ensuring safety, reducing energy consumption, and fostering sustainability. To improve security and safety, investing in sufficient lighting for public areas and highways encourages energy conservation and long-term environmental viability. Recent developments in the discipline are examined, along with significant obstacles and opportunities, by conducting a comprehensive analysis of the most recent scientific literature. In this paper, novel techniques have been explored to improve street light defect identification through the application of ML and DNN algorithms. In this paper, the historical data and sensor measurements are analysed using various machine learning algorithms to detect faults in street lights, including support vector machines, decision trees, and ensemble approaches. The paper highlights the merits, drawbacks, and performance metrics of various proposed models through comparative analyses. To promote sustainable urban development and enhance the general quality of life, our research aims to acquire knowledge regarding effective methodologies for enhancing the detection of street light malfunctions in smart cities.