|dc.description.abstract||The four major stages of disaster management are mitigation, preparation, response and recovery. Waste management is one of the core activities in the recovery stage and focuses on collecting, reducing or recycling, and final disposal of the remaining waste. The volume of waste generated from a single event can reach 5 to 15 times the annual waste normally produced by affected communities. The clearance, removal and disposal of such large amounts of debris are costly and time-consuming operations. However, there has been little literature dedicated to the improvement of disaster waste management (DWM) procedures compared to other operations in disaster management. The main objective of this thesis is to develop an integrated framework to improve DWM. Two sets of models that focus on two topics have been developed, namely reliability analysis of a DWM system and the two-echelon DWM system optimisation. The framework is tested for its validity and capacity for an improved understanding of the challenges in disaster waste clean-up.
In the first part of the thesis, a mathematical model is built to implement the First Order Reliability Method (FORM) to investigate reliability based on variables have an impact on the system, which were identified and summarised in the literature review. An optimisation model is developed that consider the total cost and clean-up period constraints to improve reliability. To solve the optimisation model, which is non-linear, a genetic algorithm is developed. The methodology is validated using a case study in Victoria, Australia. Sensitivity analysis was conducted to identify the impact of total cost and total clean-up time on the reliability of the system.
In addition, a methodology has been developed for estimating waste accumulation caused by disasters and the reliability of DWM system. These consider the uncertainty of return period and scale of disasters. To estimate the reliability of the system, FORM is used to evaluate the system's reliability. Two case studies are presented to illustrate how the methods can be applied in the real world. The reliability index curve of the system developed from sensitivity analysis can provide information for decision-makers regarding disaster waste clean-up arrangements. The approach developed can be used to analyse the effects of different parameters involved in the system after disasters.
In the second part of the thesis, initially, a methodology is presented to select candidate Temporary Disaster Waste Management Sites (TDWMS) that can be regarded as a land suitability analysis problem. ArcGIS was used to conduct the analysis, which includes four main steps: identifying and determining criteria, weighting criteria, mapping standardised layers, and overlapping standardised layers. The Modelbuilder function was applied to build the analysis model. Boolean logic was used to standardise the criteria map layers. A total of 45 candidate sites were selected within the case study area in Murrindindi, Victoria, Australia. According to the analysis, the distance from groundwater, drinking water resources, and public water supplies are the most sensitive criteria.
Using the location of TDWMS candidates, an optimisation model was developed for small-scale and large-scale disasters. In small-scale disasters demand from each customer node is smaller than the capacity of collection vehicles. Therefore, the problem can be seen as a Multi-Period Two-echelon Location Routing Problem (MP-2ELRP) in which the main decisions are the location of the TDWMS and the routing of vehicles in both echelons. In this thesis, both a Mixed Integer Programming (MIP) and a genetic algorithm were developed to model and solve the problem, respectively. A methodology for generating data for a case study and some alternative testing instances, which are used to evaluate the efficiency of both the model and the heuristic were developed. Computational tests indicate the robust performance of the genetic algorithm and allow for a thorough analysis of the effect of using TDWMS in terms of both the cost and the duration of the clean-up process.
In large-scale disasters, the arrangement to demolish damaged buildings as well as selecting the location of TDWMS, in which the demand from each customer node is larger than the capacity of collection vehicles is considered. A multi-objective MIP model is developed that consider the limitations on the working time of vehicles, vehicle capacities and the capacity of TDWMS. The goal of the model is to minimise the total cost and total time spent in the clean-up process. Three different approaches are developed to solve the model, which are tested with artificial instances containing different numbers of customer nodes. A case study in Kinglake, Victoria, Australia, which was badly affected by the 2009 Black Saturday bush-fires, is conducted to validate the model and analyse the significance of building demolition sequences.||en_US