Linked Open Simulations: An Ontology-Based Approach for System Dynamics Models on Insight Maker

This thesis introduces a System Dynamics ontology, SDMOnto, for the System Dynamics models (SDMs) available on the modeling platform Insight Maker (IM). The aim was to capture the metadata associated with these models and automate their translation to RDF using a programming approach. This involves integration with SDMOnto to enhance model interoperability and promote the reusability of cross-domain SDMs. SDMOnto consists of 24 classes, 14 object properties, 11 data properties, and 129 pre-defined individuals. Three test cases (models) were used to develop and evaluate the ontology's capability to store SDMs from IM as intended, and to identify any inconsistencies within the ontology. We found that each test case generated corresponding triples and individuals without resulting in any inconsistencies, indicating a successful conversion. In addition to the main research, we proposed a preliminary implementation for the integration of external, tabular data and its units of measure based on CSVW (CSV on the Web) standards. This implementation also generated consistent conversions. However, further inspection revealed inaccuracies, which require refinements in the implementation and more testing to prove feasibility. Since this paper is a first attempt at a standardized representation for SDMs, we recommend that future research focuses on evaluating the scalability and reusability of the ontology to address any issues that may arise on a larger scale.