مجلة الاربعين المحكمة
Religious events are held globally, where the Arbaeen pilgrimage is considers one of the largest events. During this event, millions of visitors arrive in Iraq from different countries, creating an urgent need to ensure a safe and organized mass movement for people and vehicles. Therefore, in this research an intelligent framework for crowd movement analysis and anomaly detection was designed specifically for this event. Synthetic datasets were created to simulate the dynamics of the visit and test the proposed framework. Six machine learning algorithms were applied (Isolation Forest, Local Outlier Factor (LOF), Rolling Z-Score, ARIMA Residuals, DBSCAN Clustering, and Kernel Density Estimation (KDE)) to analyze Data identify anomalies (such as sudden overcrowding or irregular movement patterns) which may indicate possible threats to safety, unusual stops, or irrational movements. In addition to calculating the accuracy of each algorithm by comparing its results with the majority vote results and measuring the execution time of each algorithm. Moreover, to maintain data privacy and integrity, with the results signed using a digital signature (SHA3-256) to ensure security and integrity. The system aims to support the shrine authorities by monitoring crowd behavior and enabling proactive measures to enhance the safety of visitors and the smooth execution of procedures.