Major modern theme parks, consisting of tens or even close to one hundred of tourist attractions, are growing continually more complex. The operation of such parks is becoming increasingly more difficult which makes developing their management systems a very challenging endeavor. More so when many different smart objects like cameras, mobile terminals and various sensors (e.g. RFIDs) are deployed throughout the environment which must be integrated into the system. Moreover, in order for the system to be conformable with the concept of Internet of Things and with the idea of Smart City, it should make an intelligent use of a distributed sensor network and provide smart capabilities, i.e. improving the process of tourists management within a smart territory, e.g. automatic congestion avoidance. Implementing such algorithms which involve Big Data analytics can be a very demanding task, especially when the system is required to be scalable to support a huge number of smart devices. Therefore the ability to accurately simulate theme park and test many different scenarios (e.g. distinct hardware configurations, varying positions of attractions, contrasting tourists’ actions) becomes imperative. In this paper we describe a method for simulating IoT-based theme parks. Presented methodology integrates several models: tourists behavior model, tourist attractions model, theme park model and simulation model. Our goal is to create a computer simulation which is able to efficiently model smart theme park.