Simulation plays a crucial role in increasing customer satisfaction and strengthening long-term relationships. Focusing on speed and convenience of service, as well as efficient management is essential in business strategy. Understanding customer characteristics and applying technology supports service efficiency. The method in this research is observation, where the results of the observation become the basis for analysis and decision making. This research aims to compare simulation methods between existing models using Arena and alternatives using Flexim. Based on the results of the t-test, with the hypothesis of testing the significance of the differences between the two models. Based on t-alpha analysis and t-test statistical tests on data processing using the data analysis feature in Ms. Excel, it was found that the critical limit for a 2-sided acceptance chart is +-0.9861. The t-test result of 0.9814 is in the acceptance area, supporting the statistical sustainability of the model with the observation results. Thus, it can be concluded that the data can accept Ho and the system is considered valid. The recapitulation shows that service 3 in Alternative 1 has the highest sales, but with a longer stay time. Alternative 1 on service 1 has the fastest stay time. The existing model has an even distribution without distinguishing the number of goods purchased. The result of using the existing model with the arena is replication in the range 60-769 seconds with the number of queues in the range 77-409 seconds. Meanwhile, using flexsim on the existing model gets a queue time on 10 replication models of around 66-70 seconds. Based on the results of observation data, it shows that the queue time range is 112-115 seconds. Therefore, flexsim data is also considered invalid because the time span is quite different. So, it can be concluded that the existing service and queue models are good.