Gym crowd is the total number of people present at a gym at any particular time. This can comprise gym members, staff, trainers, and casual visitors. The density of the gym crowd or traffic might vary depending on several circumstances, including time of day, day of the week, season, and special events. Thus, there is a need to understand the dynamics of the gym crowd for physical activity patterns, space utilization, social interactions, and their impact on equipment usage and facility management. This research leverages linear and seasonal patterns for gym crowd prediction. It compares the performance of various machine learning models, such as the autoregressive integrated average (ARIMA), seasonal autoregressive integrated moving average (SARIMA), exponential smoothing state space model (ETS), SDGR_V1 (linear regressor), and SDGR_V2 (standard scaler) models. The SDGR_V1 and SDGR_V2 models had the best performance, with a mean absolute error (MAE) of 13.065 and 12.052, respectively, the mean squared error (MSE) of 304.909 and 253.234, and the root mean squared error (RMSE) of 17.462 and 15.913. Both models had the highest R2 scores. The R2 score of the SDGR_V1 was 0.408, but after training and evaluation, the R2 score of SDGR_V2 was 0.509, which showed a slight improvement by 0.101.