Repository hosted by TU Delft Library

Home · Contact · About · Disclaimer ·
 

Simulation independent model configuration

Publication files not online:

Author: Wermeskerken, F.P.J. van · Ferdinandus, G.R. · Berg, T.W. van den · Bosch, K. van den · Smelik, R.M. · Henderson, H.C.
Type:article
Date:2018
Source:Proceedings of the 2018 Winter Simulation Innovation Workshop (SIW), Orlando, FL
Identifier: 783908
Keywords: Simulation · Simulation scenario · Model configuration · Ontology · OWL

Abstract

The models of all entities involved in a simulation need to be configured to behave in line with the purpose of the scenario. Simulation model configuration is quite complex for several reasons. For one, a simulation model such as a human behavior model or a sensor model has many detailed parameters and specialist knowledge is needed to configure and use these models appropriately. The people that develop scenarios for simulation systems are seldom specialists in the underlying simulation models. Moreover, different simulation systems that model the same or similar concept like a human or a specific type of vehicle, in practice all use different models and model configuration parameters. Configuration parameters for similar concepts have different names or meanings in different systems and simulation model configurations cannot be exchanged between simulation systems. This paper investigates the concept and envisioned use of a so called "Simulator Independent Model Configuration" (SIMC) language based on the ontology language OWL to describe simulation model configurations in a simulation system independent way. Using the SIMC language it should, for example, be possible to model system-neutral simulation model configurations that can be translated into system-specific simulation model configurations in a systematic fashion. Furthermore, the semantic properties of OWL should open possibilities for reasoning about concepts, properties and parameters, and should provide the capability to infer further information. This paper presents several use-cases and an analysis of a case study to evaluate the suitability of the OWL language for defining simulation models configurations.