Layout based model generation

  • Authors:
  • Peter Lorenz;Thomas Schulze

  • Affiliations:
  • Department of Simulation and Graphics, Otto-von-Guericke University Magdeburg, D-39016 Magdeburg, PSF 4120, Germany;Department of Simulation and Graphics, Otto-von-Guericke University Magdeburg, D-39016 Magdeburg, PSF 4120, Germany

  • Venue:
  • WSC '95 Proceedings of the 27th conference on Winter simulation
  • Year:
  • 1995

Quantified Score

Hi-index 0.00

Visualization

Abstract

Layouts of systems (e.g. barber shops, job shop, road traffic and copper smelter) contain information important for modelling. Layout based model generation (LBMG) entails: extracting all model relevant data from layouts, complementing the initial layout by simulation specific components, extracting additional model relevant data from other data sources, generating a simulation model from these extraction's, asking the user for complementary input data, running the simulation model, and presenting the simulation results and animation on the layout, generated in the second step. The paper describes an efficient approach to simulation modelling. This approach intends to avoid any doubled or repeated data acquisition and any avoidable mouse click or drop and drug. The modeler can make precise, realistic models with little cost in terms of time or money. The models that are created are now free of bugs that are normally introduced by the manual model construction and data acquisition process. The preconditions for LBMG are the the existence of a defined set of object classes in the domain of application and the existence of layouts or other pictures which contain model relevant information and which are of interest for the presentation of results by animation. The first ideas for LBMG were described by Lorenz, Schriber and Schulze (1994) and by Lorenz and Schulze (1995). The LBMG approach is described and verified by the implementation of a prototype. The prototype uses AutoCAD, DXF and Proof Layout file types as the layout basis for model generation. The approach can be used for any simulator with a language oriented model description interface, e.g. GPSS/H, MODSIM, SIMAN and other simulation systems.