Multi-objective Optimization and GIS to Improve Climate Change Induced Disaster Risk Management in Africa
(2024)- Abstract
- We live in an era that suffers from climate change issues, not least disaster risks induced by climate change. While measures, such as proper evacuation plans, are required to reduce the negative impacts of disasters when they hit, actions should also be taken to reduce climate change effects by e.g. increasing the use of renewable energies or proper urban land-use allocation, where climate change factors are considered among other criteria. Planning and decision-making on these issues are usually complex and complicated, since several criteria, usually conflicting with each other, should be taken into consideration.
Multi-objective optimization (MOO) has proven to be a proper technique for solving multi-criteria decision... (More) - We live in an era that suffers from climate change issues, not least disaster risks induced by climate change. While measures, such as proper evacuation plans, are required to reduce the negative impacts of disasters when they hit, actions should also be taken to reduce climate change effects by e.g. increasing the use of renewable energies or proper urban land-use allocation, where climate change factors are considered among other criteria. Planning and decision-making on these issues are usually complex and complicated, since several criteria, usually conflicting with each other, should be taken into consideration.
Multi-objective optimization (MOO) has proven to be a proper technique for solving multi-criteria decision analysis problems, where criteria are conflicting. Metaheuristic algorithms, inspired from nature, has been developing and showing a proper performance for solving complex MOO problems. Meanwhile, these algorithms yet need to be modified and adjusted to perform well, for each specific case study project. This research aims to improve metaheuristic algorithms to make them suitable for solving some spatial problems related to disaster risk management.
The research started by making a comparative study between well-known multi-objective optimization algorithms in order to not only learn about MOO, but also get an insight about the performance of algorithms and how they could be enhanced (Paper 1). Then the study continued by proposing a modified multi-objective cuckoo search algorithm for evacuation planning (Paper 2). The third study was in the context of the impact of urban land-uses on climate change, and hence modified and applied the Non-dominant Sorting Genetic Algorithm – III (NSGA-III) for urban land-use planning in Mozambique (Paper 3). The last study was about modifying NGSA-II for solar farm site selection in Mozambique (Paper 4).
The results of the above studies demonstrated the high potential of metaheuristic algorithms and multi-objective optimization to solve complex spatial problems that in turn can facilitate planning and decision-making to prevent or respond climate change induced disaster risks. The performances of the modified and original algorithms were compared. The evaluation showed improved performance for each of the selected case studies.
(Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/3d143ca3-e5b8-4821-8874-451e88df0c42
- author
- Sicuaio, Tomé Eduardo LU
- supervisor
- opponent
-
- Professor Zhou, Qiming, Department of Geography, Hong Kong Baptist University, China.
- organization
- publishing date
- 2024-06-11
- type
- Thesis
- publication status
- published
- subject
- keywords
- Urban planning, multi-objective optimization, Evacuation planning, Land use allocation, Geographical Information Science, Metaheuristic algorithms
- pages
- 108 pages
- publisher
- Lund University
- defense location
- Världen, Geocentrum I, Sölvegatan 10, Lund.
- defense date
- 2024-06-11 10:00:00
- ISBN
- 978-91-89187-39-9
- 978-91-89187-40-5
- language
- English
- LU publication?
- yes
- id
- 3d143ca3-e5b8-4821-8874-451e88df0c42
- date added to LUP
- 2024-05-14 10:28:16
- date last changed
- 2024-06-13 15:02:24
@phdthesis{3d143ca3-e5b8-4821-8874-451e88df0c42,
abstract = {{We live in an era that suffers from climate change issues, not least disaster risks induced by climate change. While measures, such as proper evacuation plans, are required to reduce the negative impacts of disasters when they hit, actions should also be taken to reduce climate change effects by e.g. increasing the use of renewable energies or proper urban land-use allocation, where climate change factors are considered among other criteria. Planning and decision-making on these issues are usually complex and complicated, since several criteria, usually conflicting with each other, should be taken into consideration.<br/> <br/>Multi-objective optimization (MOO) has proven to be a proper technique for solving multi-criteria decision analysis problems, where criteria are conflicting. Metaheuristic algorithms, inspired from nature, has been developing and showing a proper performance for solving complex MOO problems. Meanwhile, these algorithms yet need to be modified and adjusted to perform well, for each specific case study project. This research aims to improve metaheuristic algorithms to make them suitable for solving some spatial problems related to disaster risk management.<br/> <br/>The research started by making a comparative study between well-known multi-objective optimization algorithms in order to not only learn about MOO, but also get an insight about the performance of algorithms and how they could be enhanced (Paper 1). Then the study continued by proposing a modified multi-objective cuckoo search algorithm for evacuation planning (Paper 2). The third study was in the context of the impact of urban land-uses on climate change, and hence modified and applied the Non-dominant Sorting Genetic Algorithm – III (NSGA-III) for urban land-use planning in Mozambique (Paper 3). The last study was about modifying NGSA-II for solar farm site selection in Mozambique (Paper 4).<br/> <br/>The results of the above studies demonstrated the high potential of metaheuristic algorithms and multi-objective optimization to solve complex spatial problems that in turn can facilitate planning and decision-making to prevent or respond climate change induced disaster risks. The performances of the modified and original algorithms were compared. The evaluation showed improved performance for each of the selected case studies.<br/>}},
author = {{Sicuaio, Tomé Eduardo}},
isbn = {{978-91-89187-39-9}},
keywords = {{Urban planning; multi-objective optimization; Evacuation planning; Land use allocation; Geographical Information Science; Metaheuristic algorithms}},
language = {{eng}},
month = {{06}},
publisher = {{Lund University}},
school = {{Lund University}},
title = {{Multi-objective Optimization and GIS to Improve Climate Change Induced Disaster Risk Management in Africa}},
url = {{https://lup.lub.lu.se/search/files/183455200/Tom_Eduardo_Sicuaio_-_WEBB.pdf}},
year = {{2024}},
}