Proceedings of the 2011 Winter Simulation Conference S. Jain, R.R. Creasey, J. Himmelspach, K.P. White, and M. Fu, eds. TOWARDS AN ONTOLOGICAL FOUNDATION OF AGENT-BASED SIMULATION Giancarlo Guizzardi Gerd Wagner Federal University of Espírito Santo (UFES) Brandenburg University of Technology Computer Science Department Institute of Informatics Av. Fernando Ferrari P. O. Box 101344 s/n29060-970 Vitória, Espírito Santo, BRAZIL 03013 Cottbus, GERMANY ABSTRACT A simulation model is intended to capture a real-world system. Consequently, the modeling language used for making the simulation model should have a “real-world semantics” guaranteeing some kind of ontological faithfulness for the models made with it. In this paper, we propose to use ABDESO, a founda- tional ontology for agent-based discrete event simulation, for evaluating agent-based simulation lan- guages. 1 INTRODUCTION In recent years, there has been a growing interest in the application of foundational ontologies (also known as upper level, or top-level ontologies) for providing real-world semantics for conceptual modeling languages, and methodological guidelines for evaluating and improving the models made using these lan- guages. While the value of an ontologically well-founded conceptual modeling language is widely acknowledged in the areas of information system and software system engineering, as indicated by the great number of recent publications in this area, the issue of investigating the ontological foundations of simulation languages did not yet receive much attention in the scientific literature. This paper on the onto- logical foundations of agent-based simulation (ABS) is a follow-up of (Guizzardi and Wagner 2010a), which was concerned with the ontological foundations of basic discrete event simulation (DES). While there are several research papers on the ontological foundations of organization and business process modeling (see Section 2), there has been no attempt yet to demonstrate the value of an ontologi- cally well-founded modeling language for ABS. The main benefit obtained from establishing the ontolog- ical foundations of the core concepts of agent-based modeling languages is a clarification of their real world semantics. An ontological semantics of a simulation modeling language leads to a higher overall quality of the simulation models expressed in that language with respect to comprehensibility, maintaina- bility, interoperability and evolvability. We argue in section 3 that it is natural to consider the ABS paradigm as an extension of the DES par- adigm. We, therefore, also speak of agent-based DES or, in short, ABDES. In a series of publications (Guizzardi and Wagner 2004, 2005, 2010b; Guizzardi, Falbo and Guizzardi 2008) we have reported about our project for developing a foundational ontology called “UFO” (for Uni- fied Foundational Ontology) by employing theories from Formal Ontology, Cognitive Psychology, Lin- guistics, Philosophy of Language, and Philosophical Logics. The core of UFO has been established through the development of an ontology of endurants by the first author in Guizzardi (2005). This founda- tional ontology has been successfully applied in the analysis of several important conceptual modeling constructs such as Roles, Types, Part-Whole Relations, Attributes, and Data types, among others. 284 978-1-4577-2109-0/11/$26.00 ©2011 IEEE