Title :
Fuzzy set methods for qualitative and natural language oriented simulation
Author :
Fishwick, Paul A.
Author_Institution :
Dept. of Comput. & Inf. Sci., Florida Univ., Gainesville, FL, USA
Abstract :
The author discusses the approach of using fuzzy set theory to create a formal way of viewing the qualitative simulation of models whose states, inputs, outputs, and parameters are uncertain. Simulation was performed using detailed and accurate models, and it was shown how input and output trajectories could reflect linguistic (or qualitative) changes in a system. Uncertain variables are encoded using triangular fuzzy numbers, and three distinct fuzzy simulation approaches (Monte Carlo, correlated and uncorrelated) are defined. The methods discussed are also valid for discrete event simulation; experiments have been performed on the fuzzy simulation of a single server queuing model. In addition, an existing C-based simulation toolkit, SimPack, was augmented to include the capabilities for modeling using fuzzy arithmetic and linguistic association, and a C++ class definition was coded for fuzzy number types
Keywords :
Monte Carlo methods; discrete event simulation; fuzzy set theory; queueing theory; C++ class definition; C-based simulation toolkit; Monte Carlo; SimPack; correlated; discrete event simulation; encoding; fuzzy arithmetic; fuzzy set theory; linguistic association; natural language oriented simulation; qualitative; qualitative simulation; single server queuing model; trajectories; triangular fuzzy numbers; uncorrelated; Artificial intelligence; Automation; Computational modeling; Computer simulation; Fuzzy sets; Fuzzy systems; Humans; Natural languages; Robots; Statistics;
Conference_Titel :
Simulation Conference, 1990. Proceedings., Winter
Conference_Location :
New Orleans, LA
Print_ISBN :
0-911801-72-3
DOI :
10.1109/WSC.1990.129569