Study on human oriented system simulation: comparison of different methods to represent human behaviour

Wicaksono, Kukuh Nasrul (2015) Study on human oriented system simulation: comparison of different methods to represent human behaviour. [Dissertation (University of Nottingham only)]

[thumbnail of MIT-dissert_74_Wicaksono Kukuh_2015.pdf] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (11MB)

Abstract

Recently, researchers from any subject fields started to adopt agent based modelling to get deeper insight into complex system which are not well-captured by using traditional modelling approach. By using this, modeller can build the simulation from the bottom up which means, even the modeller may not have any idea about how the system as a general works and behaves, as long as modeller knows a little bit about how the objects works or behaves individually within the system, then the simulation can be built. This advantage may be useful to model the complicated system which the individual objects within the system are easier to be observed such as human oriented system. Therefore, human behaviour modelling will be great concern since it will affect the internal decision making system. Different methods have been used to represent human behaviour in the previous researches such as probabilistic method, fuzzy logic method, BDI, etc.

Through this study, comparison of different methods to represent human behaviour has been conducted to see whether it can influence the outcomes of the simulation with the same scenario. Three methods that have been compared were probabilistic method, fuzzy logic method, and combination of fuzzy logic and probabilistic method. In this study, three simulation models that modelling daily work of software development working team have been built. Each simulation model implemented different method separately.

At the end of the study, it can be concluded that different methods of human behaviour representation in the simulation model can affect the outcome simulation even if it simulated the same scenario.

Item Type: Dissertation (University of Nottingham only)
Depositing User: Gonzalez-Orbegoso, Mrs Carolina
Date Deposited: 09 Dec 2015 15:33
Last Modified: 19 Oct 2017 15:10
URI: https://eprints.nottingham.ac.uk/id/eprint/30804

Actions (Archive Staff Only)

Edit View Edit View