RESEARCH PAPER Modelling “Occident/Orient” duality and migration process with mobile agents Ahmed M'hamdi 1 | Fatima‐Ezahra Sfa 1 | Mohamed Nemiche 1 | Saida Hachimi El Idrissi 1 | Rafael Pla‐López 2 1 IMI Laboratory, Ibn Zohr University, Agadir, Morocco 2 University of Valencia, Valencia, Spain Correspondence Mohamed Nemiche, IMI Laboratory, Ibn Zohr University, Agadir, Morocco. Email: nemiche@uv.es Abstract The history of the evolution of human societies up to the present day has been marked by large‐scale collective migrations as well as by a clear duality between a West dominated by individualistic social behaviours and an East dominated by collective social behaviours. Only, at the end of the 20th century, the two lines of evolution converge towards a capitalist globalization. The main objective of this work is to build an agent‐based model from the mathematical model of Rafael Pla‐López in order to simulate the social evolu- tion of an artificial society. The simulations of our model make it possible to understand the collective migration process and different lines of social evolu- tion such as Occident/Orient duality. This model can serve as a virtual labora- tory to understand the processes involved in human social evolution and explore different trajectories of human social evolution. KEYWORDS agent‐based modelling, migration, North/south diversity, Occident/Orient duality 1 | INTRODUCTION Social simulation is a research field emerging from the intersection of computer science, statistics, and social sci- ences in which new computer tools and mathematical methods are used to answer societal questions. The field is intrinsically collaborative: Social scientists provide a vital context and insight into relevant research questions and data sources, whereas statisticians and computer sci- entists provide expertise in developing mathematical models and computer tools (Sallach & Macal, 2001). The methods most used for social simulation are based on two approaches depending on their level of abstraction (Cioffi‐Revilla, 2014). The first approach “top‐down” is a holistic approach. The second one is “bottom‐up” approach, which proposes to explicitly model the behav- iours of the entities and considers that the global dynamics at the macroscopic level results directly from the interaction of behaviours at the microscopic level. The social simulation models based on the “bottom‐ up” approach generally use cellular automata or agent‐ based models (Cioffi‐Revilla, 2014; Gilbert & Troitzsch, 2005; M'hamdi & Nemiche, 2018). Cellular automata are adequate when the interactions are planned just between a cell and its neighbours. But, if the elements of the sys- tem are more complex in their behaviour, we use agent‐ based simulation (M'hamdi & Nemiche, 2018). The agent‐based simulation allows to model autonomous, het- erogeneous, and mobile agents as well as its environment (Bonabeau, 2002). Understanding the history of the evolution of human societies with the classical methods used in the social sciences is a difficult task due to its complex and inde- composable character (Gilbert, 2005). Rafael Pla‐López DOI: 10.1002/sres.2642 Syst Res Behav Sci. 2019;1–15. © 2019 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/sres 1