Khalid, Mohd Nor Akmal
(2015)
Optimizing Crowd Evacuation In The
Emergency Route Planning Problem.
Masters thesis, Universiti Sains Malaysia.
Abstract
Situasi bencana, yang berlaku secara semula jadi (kebakaran, banjir, taufan) atau buatan
manusia (contohnya pengeboman pengganas, tumpahan bahan kimia, dan lain-lain), telah
meragut ribuan nyawa, mencetuskan keperluan untuk pemindahan kecemasan. Biasanya,
mengoptimumkan pelan pemindahan kecemasan melibatkan berkesanan pemodelan orang
ramai dan pemilihan laluan, dimana pelan yang optimum penting dalam masalah perancangan
laluan kecemasan (ERP). Pelbagai pendekatan ERP telah dibangunkan dimana
diklasifikasikan kepada pendekatan matematik, keputusan sokongan, heuristik, dan
meta-heuristik. Ulasan kesusasteraan menyeluruh telah menunjukkan kepentingan untuk
merapatkan jurang antara pemodelan dan pemilihan laluan, di mana di mana pendekatan
bersepadu dan berdaya maju diperlukan. Dalam kajian ini, satu perancangan pemindahan
rangka kerja bersepadu menggunakan model pemindahan orang ramai dan sistem imun (AIS)
algoritma tiruan, yang dipanggil iEvaP, telah dicadangkan. iEvaP telah disahkan terhadap Lu
et al. (2003) dan parameternya telah ditentukan untuk prestasi yang optimum.
Disastrous situations, either natural (e.g. fires, floods, hurricane) or man-made (e.g.
terrorist bombings, chemical spills, etc.), have claimed the lives of thousands, triggering the
needs for emergency evacuation. Typically, optimizing an emergency evacuation plan
involves both the effectiveness in crowd modelling and route selection, where an optimum
evacuation plan is vital in the emergency route planning (ERP) problem. Various ERP
approaches have been developed which are classified into mathematical, decision-support,
heuristic, and meta-heuristic approaches. Exhaustive literature reviews have shown the
significance of bridging the gap between modeling and routing, where an integrated and
viable approach is needed. In this study, an integrated evacuation planning framework
utilizing crowd evacuation model and an artificial immune system (AIS) algorithm, called
iEvaP, was proposed. iEvaP was validated against Lu et al. (2003) and its parameters were
calibrated for optimum performance.
Actions (login required)
 |
View Item |
![[img]](http://eprints.usm.my/style/images/fileicons/application_pdf.png)
