The University of Edinburgh -
Division of Informatics
Forrest Hill & 80 South Bridge

Research Paper #534

Title:The Use of Domain Information for Comprehension and Construction of Simulation Models
Date:Jun 1991
Presented:Submitted to IJCAI-91
Abstract:We are concerned with two important issues in simulation modelling: model comprehension and model construction. Model comprehension is limited because models are poorly documented. This makes it difficult for models to be modified or used by others. Two key factors hindering model construction are the fact that tools do not cater for how users think about their problems [large conceptual distance], and the vast modelling search space. For some classes of problems, special-purpose paradigms and/or tools may be used to reduce these difficulties to a manageable level. We describe an approach for overcoming the above difficulties in domains for which no such tools exist. The key idea is to build a domain model first, and then define the simulation model in terms of it. This reduces conceptual distance, structures the modelling search space, and automatically documents the simulation model. We describe Elklogic, a wide-spectrum language based on the typed lambda calculus for representing both domain and simulation modelling information. We describe ELK, a working implementation of this approach in the domain of ecological modelling.

[Search These Pages] [DAI Home Page] [Comment]