Boid-based simulation and modeling of crowd flow during the Sri Pada pilgrimage

The Sri Pada pilgrimage is a culturally and religiously significant event in Sri Lanka that draws thousands of devotees annually. During peak seasons, managing the dense and dynamic flow of pilgrims along the narrow and steep paths becomes a critical challenge. This study presents a Boid-based crowd simulation model to analyze and predict crowd behavior in the Sri Pada pilgrimage. The Boid algorithm, inspired by the flocking behavior of birds, enables the modeling of individual pilgrims as agents governed by simple local rules—separation, alignment, and cohesion—leading to realistic collective movement patterns. The simulation was developed using Python and related libraries (Pygame, scikit-learn, pandas, Matplotlib, NumPy), and it captures the natural interactions of the crowd along the pilgrimage path. Synthetic datasets were generated based on simulation results and used to train a linear regression model for estimating crowd density. The model achieved a high R 2 score of 0.85, with an MAE of 30 and an MSE of 250, indicating strong prediction capabilities. Visual outputs such as heatmaps were used to identify areas of high congestion and potential risk. The findings demonstrate that Boid-based modeling can effectively simulate crowd flow, providing valuable insights for authorities to improve crowd management, enhance safety, and design better infrastructure for future pilgrimages [ 3 ].