 This paper presents a novel approach to predict the expansion of pandemic plastic in mega cities of Iran. A deep learning-based model was developed and evaluated against other machine learning algorithms such as decision trees, random forests, support vector machines, logistic regression, and multi-layer perceptrons. The results showed that the DNN-based model outperformed all other models and had the lowest error rate, MSE equals 0.024, IMSE equals 0.027, MAP equals 0.025, indicating its high accuracy. Furthermore, the RLC curve analysis indicated that the DNN model had the highest overall accuracy, AUC equals 0.99. This study provides a reliable tool for predicting the expansion of pandemic plastic in mega cities of Iran and can be used to develop effective strategies for managing the pandemic plastic. This article was authored by Yasser Aynanakirin, Chulitsai, Mohamed Azouafza, and others.