Return to search

Integrated intelligent management platform for the built environment incorporating agent-based simulation and geographic informationsystem

In the area of indoor built environment modelling, the main focus was usually put on the

accuracy of the modelled equipment within the given environment. There was not enough

attention placed on the modelling of occupants inside the building, who represent the

ultimate end?users and should be the most important element of an indoor built

environment model. This deficiency was mainly due to the difficulties of modelling human

behaviour with the traditional mathematical modelling methods usually adopted in the field.



Therefore, in order to model the occupants more effectively, a more appropriate alternative

approach is needed. This thesis examined the suitability of agent?based modelling and

simulation (ABMS) for the indoor built environment. Philosophies of ABMS for the built

environment were discussed, with the natural flow of progression stages in indoor built

environment models outlined and explored, showing the need for switching a problem of

simplicity to one in which a complex organised system was defined.



Scenarios of single active agent case in ABMS for indoor built environment were used to

demonstrate the effectiveness of this approach, with the finding that ABMS was versatile in

modelling different types of occupants. It provided an illustration of the difference between

people with different sensation time, and estimated energy savings between occupants with

varied understanding of the environment.



A classification scheme for different types of indoor built environment models with different

levels of complexities was proposed and discussed. Based on the classification scheme,

models with different levels of complexities were applied to the same building to explore the

feasibility and practicality of the scheme. It was found that the simulation results provided

insightful figures on probable energy consumption for different types of occupants and their

crowd behaviour when interactions between occupants were considered. Findings included

the amount of energy saved (around 10%) by a knowledgeable occupant over a less

informed one, and the tendency for the majority (?66%) of occupants trying to save energy

by assessing the environment themselves under different conditions.



To further assist building management and the implementation of ABMS approach in built

environment for simulations, the concepts of geographic information system (GIS) were

modified for applications in the indoor environment, based on its similarity with ABMS for a

grid?based environment. A prototype program was developed to test the applicability of the

proposed concepts, with OPC connectivity for connections to real?world equipment. It was

found that the idea of indoor GIS helps to solve problems such as the facility layout problem

or the difficulties of information storage/retrieval in facility management, and this concept

also integrated well with ABMS.

With the successful modelling of different types of occupants and simulation results

consistent with real data, it can be concluded from the work that ABMS is a promising

approach for future generations of indoor built environment models. Also, with the inspiring

framework of indoor GIS, the combination of ABMS and GIS in an integrated platform will

certainly bring a powerful tool to the field. / published_or_final_version / Mechanical Engineering / Doctoral / Doctor of Philosophy

Identiferoai:union.ndltd.org:HKU/oai:hub.hku.hk:10722/180812
Date January 2012
CreatorsLee, Ho-yin, 李浩賢
PublisherThe University of Hong Kong (Pokfulam, Hong Kong)
Source SetsHong Kong University Theses
LanguageEnglish
Detected LanguageEnglish
TypePG_Thesis
Sourcehttp://hub.hku.hk/bib/B47244410
RightsThe author retains all proprietary rights, (such as patent rights) and the right to use in future works., Creative Commons: Attribution 3.0 Hong Kong License
RelationHKU Theses Online (HKUTO)

Page generated in 0.0034 seconds