Predicting parking space occupancy poses a formidable challenge due to the multitude of influencing factors, including the day of the week, time of day, and prevailing weather conditions. The forecasting approach for parking space occupancy utilizes a Convolutional Neural Network architecture derived from AlexNet. Renowned for its prowess in image classification tasks, AlexNet has been adapted for predicting parking space utilization. The training process employed a dataset comprised of images captured by parking lot cameras, with each image denoting the status of individual parking spaces. The training dataset encompasses a diverse range of parking lot scenarios, accommodating varying vehicle counts. The outcomes of this investigation highlight the capacity of AlexNet to precisely anticipate parking lot occupancy based on image data. This method harbors the potential to enhance traffic management efforts and mitigate congestion-related challenges.