This paper describes a model that simulates crowd movement incorporating an efficient follow-the-leader technique based on cellular automata (CA). The scope of the method is to derive principal characteristics of collective motion of biological organisms, such as flocks, swarms or herds and to apply them to the simulation of crowd movement. Thus, the study focuses on the massive form of the movement of individuals, which is lastingly detected macroscopically, during urgent circumstances with the help of some form of guidance. Nevertheless, on a lower level, this formation derives from the application of simple local rules that are applied individually to every single member of the group. Hence, the adoption of CA-based formation has allowed the development of a micro-operating model with macro-features. Furthermore, the model takes advantage of the inherent ability of CA to represent sufficiently phenomena of arbitrary complexity. The response of the model has been evaluated through different simulation scenes that have been developed both in two and three dimensions.