Smart sensing and predictive learning for optimization and simulation of airport ground equipment and staff scheduling: a cloud-based mobile integrated decision support platform
Author(s)
Fei, Yiyun
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Airports as major transportation hubs need to transport a large number of travelers
every day. Effectively managing such a large transportation system without compromising
safety has always been a huge challenge. One of the keys to enhancing overall airport
management is improving operational efficiency which includes optimizing the allocation
and scheduling of limited resources, such as workforce and equipment.
To address the joint job scheduling problem for both the GSS and GSE in the airport
apron area, this thesis proposed an intelligent decision support system, which integrates the
use of smarting sensing, predictive learning for maintenance scheduling, and algorithm and
simulation for job schedule optimization. First, the demands, information flow, interaction
between the stakeholders and the system, and the functionalities of the proposed system
have been analyzed using various design tools. Second, a real-time GSE status monitoring
solution using non-invasive sensor hardware and smart sensing has been proposed; this
method enables status displaying both locally on each GSE and remotely on a cloud-based
platform. Third, by processing and analyzing raw status data collected by sensor hardware,
the time range and duration of GSE maintenance could be predicted using recurrent neural
networks. Next, optimal job schedules will be generated for GSE and GSS using a 2-
dimensional genetic algorithm and validated in a discrete event simulation model. Finally,
the system architecture and process flow of the integrated decision support system are
analyzed. Moreover, a sample mobile app has been created and the functionalities of the
app are introduced in this thesis.
Sponsor
Date
2022-05-04
Extent
Resource Type
Text
Resource Subtype
Thesis