Home > Research > Publications & Outputs > Graphical representation of agent-based models ...
View graph of relations

Graphical representation of agent-based models in Operational Research and Management Science using UML

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review



Agent-Based Modelling and Simulation (ABM/S) is still struggling to become one of the main stream simulation methods in Operational Research (OR) and Management Science (MS), despite its generally accepted usefulness when it comes to representing human behaviour in human-centric systems. In other fields, as for example Business Studies, Economics, and Social Science, it is flourishing. One of the technical differences between ABM/S and the well-established OR/MS
simulation methods System Dynamics Simulation (SDS) and Discrete Event Simulation (DES) is that ABM/S traditionally uses an equation based modelling approach while SDS and DES use a graphical notation for the model description. We believe that having a graphical notation for ABM/S would help establish it in OR/MS. The Unified Modelling Language (UML) is a graphical notation
commonly used in software engineering for the purpose of software design. Use case and state machine diagrams, which are part of the UML notation seem to lend themselves particularly well to ABM/S. In this paper we introduce UML to the OR/MS community. First we explain step-by-step how to use UML for developing ABM/S models. Then we demonstrate the application of this graphical notation by presenting two conceptual models we built for real world OR/MS case studies.