Spatial Distribution Prediction of Passenger Flow in Tourist Attractions Based on Improved Gravity Model Method: Using Qujiang New District as an Example

Passenger flow forecasting research is mainly based on past data in accordance with certain algorithms of reasoning to predict the future traffic volume. The use of historical passenger flow data as the basis for passenger flow forecasting has a large error for scenic passenger flow. The gravity model method avoids errors arising from historical passenger flow data. It is a simple model but has limitations, low accuracy, unscientific parameter calibration, and large errors. Therefore, this study focuses on optimizing the gravity model method. The main work is to explore the laws of gravity model parameters over time and describe them in the form of a sine function. The model parameters are no longer a single constant, improving the model fit. Xi’an Qujiang New District was used for validation. The results show that the prediction accuracy is good and can provide a reference basis for passenger flow prediction in scenic spots.