Customer wait times pose significant challenges in service environments such as banks, airports, hospitals, and petrol stations, making efficient queue management essential. While traditional queuing theory (e.g., M/M/1, M/M/s) has been widely used for queue optimization, it struggles to account for real-world uncertainties in customer arrivals and service durations. This study enhances Fuzzy Queuing Theory by incorporating fuzzy sets to model such variability, comparing traditional queuing models with fuzzy queuing models that integrate triangular and trapezoidal fuzzy numbers. The Dong Shah Wong technique is employed to develop membership functions for key performance metrics, including queue length system length, waiting time in queue, and total time spent in the system. Unlike traditional models that assume fixed arrival and service rates, fuzzy queuing models introduce interval-based estimates, enabling dynamic adaptation to fluctuations in customer flow. This results in greater flexibility and improved accuracy in queue predictions, leading to reduced waiting times and enhanced service efficiency. Performance evaluations using real-world banking data from three bank branches confirm that fuzzy models outperform traditional models, particularly under high-variability conditions. These findings demonstrate that fuzzy queuing models provide a more adaptive and reliable framework for queue optimization, making them highly suitable for practical banking applications and other intensive service industries.