Massey Documents by Type

Permanent URI for this communityhttps://mro.massey.ac.nz/handle/10179/294

Browse

Search Results

Now showing 1 - 10 of 12

Enhancing multi-hazard resilience to tsunami through evacuation simulation : a case study of Napier City : a thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy in Construction Management, School of Built Environment, Massey University, New Zealand. EMBARGOED to 11 July 2026.
(Massey University, 2023-09-12) Fathianpour, Azin
Climate change has intensified the risk of natural hazards, especially tsunamis. This fact has led emergency management decision-makers to question their emergency planning and make coastal cities resilient. Evacuation has been identified as the best course of action in response to tsunamis. Regarding tsunamis, a resilient city must have a resilient evacuation plan. An evacuation plan can be called resilient once all the related infrastructures are resilient and people know how to use them. Based on the literature, simulation has been identified as a reliable source of assessing the resiliency level of infrastructure. Therefore, this doctoral thesis aims to create an evacuation simulation tool that evaluates the resilience of evacuation infrastructure. Recognising the importance of understanding the resilience of the evacuation process in terms of disaster management, current research and practice continually highlight the significance of simulation outputs concerning tsunami responses. Previous evacuation simulation tools primarily focused on monitoring pedestrian movements, neglecting interactions between pedestrians and vehicles. Furthermore, many studies did not consider factors related to human behaviour and decision-making during crises. This study aims to provide fresh insights into evacuation management by incorporating multiple behavioural and spatial factors into the simulation. The outcomes of this research generated realistic evacuation simulation results and translated them into policy and planning protocols.
A systematic approach for developing and manufacturing fruit simulators : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Food Engineering, Massey University, Palmerston North, New Zealand
(Massey University, 2023) Huang, Huijian
Due to the high cost, variable nature and seasonal availability of fruit, conducting large scale experiments for research purposes is not easy. A fruit simulator is a physical tool that mimics the mechanistic features and properties of the targeted fruit; hence, it can be used as a replacement for the fruit in research experiments. This study focuses on developing simulators for heat transfer experiments, especially in horticultural produce precooling. A framework for developing the simulator was established based on the importance of each mechanistic feature. Depending on the application's needs, the simulator can mimic different length scale levels of the targeted fruit, such as the individual fruit, the bulk stacking of the fruit or sub-units of the fruit (e.g., a punnet/bag of table grapes). The scale level determines whether certain mechanistic features are important and affects the values of the thermal properties that must be matched. For example, a simulator that mimics a punnet of fruit with enclosed air pockets has an effective thermal conductivity and volumetric heat capacity that includes contributions from the thermal properties of the fruit and air, which provides more room for material selection. Based on this framework, a systematic approach for the simulator manufacture and material selection was developed. Three different simulators were developed based on the framework: kiwifruit, apple and table grape simulators. The comparison of a simulator and real fruit precooling trials showed good agreement, validating the approach and demonstrating the feasibility of using simulators in postharvest research. The kiwifruit simulator was validated at different experimental scale levels, from individual kiwifruit to multiple kiwifruit boxes containing numerous individual kiwifruit simulators (which reflected pallet scale precooling). During the simulator development, the concept of a time-scaled approach was identified and was explored. In theory, if the volumetric heat capacity of a simulator becomes smaller while the Bi of the simulator remains the same, the heating/cooling time of the simulator in an experiment will decrease proportionally according to the Fourier number (Fo). This approach was validated via the three simulators developed in this study. The validation of the simulators confirms the feasibility of this time-scaled concept. This approach has a significant advantage in reducing the experimental time and easing the material selection process for the simulator manufacture. In the table grape simulator development, a process of using CT scans of the bulk packaged system to study the bulk shape and effective properties of the fruit subunits (bags) were developed, where the bulk shape and effective thermal properties of a bag of table grape were determined based on the process. A set of bag shaped fruit simulators was then manufactured with equivalent bulk thermal conductivity and used to validate the bulk simulator approach by comparison of cooling rates with real fruit. Overall, this study has successfully developed a generalised heat transfer simulator development framework. In addition, this study validated the feasibility and applicability of the time-scaling approach, which could be helpful for any future experiments. Furthermore, this study has developed a process to use CT scanning to determine a bulk object's bulk shape and effective property. The outcomes of the work pave the way for carrying out postharvest and packaging optimisation experimental trials with reduced variability, greater ease and without seasonal constraints. The simulator development framework provides a basis for further expansion of these concepts into other applications beyond the heat transfer focus that they were developed for in this work.
Development of a decision support system through modelling of critical infrastructure interdependencies : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Emergency Management at Massey University, Wellington, New Zealand
(Massey University, 2021) Syed, Yasir Imtiaz
Critical Infrastructure (CI) networks provide functional services to support the wellbeing of a community. Although it is possible to obtain detailed information about individual CI and their components, the interdependencies between different CI networks are often implicit, hidden or not well understood by experts. In the event of a hazard, failures of one or more CI networks and their components can disrupt the functionality and consequently affect the supply of services. Understanding the extent of disruption and quantification of the resulting consequences is important to assist various stakeholders' decision-making processes to complete their tasks successfully. A comprehensive review of the literature shows that a Decision Support System (DSS) integrated with appropriate modelling and simulation techniques is a useful tool for CI network providers and relevant emergency management personnel to understand the network recovery process of a region following a hazard event. However, the majority of existing DSSs focus on risk assessment or stakeholders' involvement without addressing the overall CI interdependency modelling process. Furthermore, these DSSs are primarily developed for data visualization or CI representation but not specifically to help decision-makers by providing them with a variety of customizable decision options that are practically viable. To address these limitations, a Knowledge-centred Decision Support System (KCDSS) has been developed in this study with the following aims: 1) To develop a computer-based DSS using efficient CI network recovery modelling algorithms, 2) To create a knowledge-base of various recovery options relevant to specific CI damage scenarios so that the decision-makers can test and verify several ‘what-if’ scenarios using a variety of control variables, and 3) To bridge the gap between hazard and socio-economic modelling tools through a multidisciplinary and integrated natural hazard impact assessment. Driven by the design science research strategy, this study proposes an integrated impact assessment framework using an iterative design process as its first research outcome. This framework has been developed as a conceptual artefact using a topology network-based approach by adopting the shortest path tree method. The second research outcome, a computer-based KCDSS, provides a convenient and efficient platform for enhanced decision making through a knowledge-base consisting of real-life recovery strategies. These strategies have been identified from the respective decision-makers of the CI network providers through the Critical Decision Method (CDM), a Cognitive Task Analysis (CTA) method for requirement elicitation. The capabilities of the KCDSS are demonstrated through electricity, potable water, and road networks in the Wellington region of Aotearoa New Zealand. The network performance has been analysed independently and with interdependencies to generate outage of services spatially and temporally. The outcomes of this study provide a range of theoretical and practical contributions. Firstly, the topology network-based analysis of CI interdependencies will allow a group of users to build different models, make and test assumptions, and try out different damage scenarios for CI network components. Secondly, the step-by-step process of knowledge elicitation, knowledge representation and knowledge modelling of CI network recovery tasks will provide a guideline for improved interactions between researchers and decision-makers in this field. Thirdly, the KCDSS can be used to test the variations in outage and restoration time estimates of CI networks due to the potential uncertainty related to the damage modelling of CI network components. The outcomes of this study also have significant practical implications by utilizing the KCDSS as an interface to integrate and add additional capabilities to the hazard and socio-economic modelling tools. Finally, the variety of ‘what-if’ scenarios embedded in the KCDSS would allow the CI network providers to identify vulnerabilities in their networks and to examine various post-disaster recovery options for CI reinstatement projects.
Parallel simulation methods for large-scale agent-based predator-prey systems : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Computer Science at Massey University, Albany, New Zealand
(Massey University, 2019) Quach, Dara (Minh) Quang
The Animat is an agent-based artiﬁcial-life model that is suitable for gaining insight into the interactions of autonomous individuals in complex predator-prey systems and the emergent phenomena they may exhibit. Certain dynamics of the model may only be present in large systems, and a large number of agents may be required to compare with macroscopic models. Large systems can be infeasible to simulate on single-core machines due to processing time required. The model can be parallelised to improve the performance; however, reproducing the original model behaviour and retaining the performance gain is not straightforward. Parallel update strategies and data structures for multi-core CPU and graphical processing units (GPUs) are developed to simulate a typical predator-prey Animat model with improved perfor- mance while reproducing the behaviour of the original model. An analysis is presented of the model to identify dependencies and conditions the parallel update strategy must satisfy to retain original model behaviour. The parallel update strategy for multi-core CPUs is constructed using a spatial domain decomposition approach and supporting data structure. The GPU implementation is developed with a new update strategy that consists of an iterative conﬂict resolution method and priority number system to simultaneously update many agents with thousands of GPU cores. This update method is supported by a compressed sparse data structure developed to allow for efﬁcient memory transactions. The performance of the Animat simulation is improved with parallelism and without a change in model behaviour. The simulation usability is considered, and an internal agent deﬁnition system using a CUDA device Lambda feature is developed to improve the ease of conﬁguring agents without signiﬁcant changes to the program and loss of performance.
Molecular dynamics simulation of inter-molecular interactions : a thesis submitted to Massey University in Albany, Auckland in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Computational Biochemistry
(Massey University, 2019) Shadfar, Shamim Zahra
Many aspects of the operation of chemical and biological systems are based on intermolecular interactions. In this work, binding modes and interactions between molecules at a range of scales have been studied, using molecular dynamics (MD) simulations. The first chapter provides an introduction of each of the different chemical and biological systems that are studied in this work. It also introduces MD and its role in the context of this research. The second chapter corresponds to the study of host-guest interactions for cyclodextrin- bullvalene complexes. Bullvalene is a shapeshifter molecule, which interconverts between different isomers at room temperature. The goal of this chapter is to capture one favourable isomer of bullvalene (guest molecule) by binding it to cyclodextrin as a host molecule. This chapter consists of two smaller chapters (2i and 2ii). The former details the development and validation of a “host-guest binding potential energy profiling” (HGBPEP) method, which is a rotational interaction energy screening method designed for prediction of the most favourable orientation and position of bullvalene isomers with respect to cyclodextrin. The latter investigates the interaction of bullvalene isomers and cyclodextrin molecules, and finally binding free energy values of the complexes are calculated. The third chapter describes KstR, a transcriptional repressor in Mycobacteria. KstR is required for Mycobacterium tuberculosis (Mtb) pathogenesis as well as regulating the initial steps in cholesterol degradation by controlling the expression of the enzymes that carry out the early stages of cholesterol catabolism. Therefore, this protein is of great interest for development of new tuberculosis treatments. In this chapter, the stability and conformational changes of KstR in its different states – apo, DNA-bound and ligand-bound –have been studied. The main goal is to investigate the binding mechanism of KstR to DNA, as well as the effect of DNA and ligand binding on the structure and dynamics of KstR more generally, using MD simulations. In the fourth chapter, KstR2, another Mtb transcriptional repressor, is studied. KstR2 represses a 14-gene regulon involved in the later steps of cholesterol degradation. It is structurally similar to KstR, but has been proposed to act through a novel scissor-like mechanism. This chapter investigates two key questions regarding the mechanism of action of KstR2: first, the effect of mutating the key switch residue ARG170 to ALA, and second, the effect of ligand binding on its structure and motion. The focus of the fifth and final chapter is phosphatidylinositide 3-kinases (PI3Ks), which are proteins that take part in signalling pathways regulating factors like cell growth, survival and proliferation, which in turn are involved in cancer. The interaction between PI3Kα and another protein, RAS, is very important in the formation, growth and maintenance of RAS- driven tumours. A model of PI3Kα (class IA PI3K) has therefore been built, as well as of RAS associated with a model cell membrane, and MD simulations used to investigate the process by which the two proteins interact with one another and with the lipid bilayer. Altogether, this thesis uses MD simulations to provide insight into intermolecular interactions at a range of scales, with a particular focus on proteins involved in tuberculosis and in cancer.
Melting temperatures of the noble gases from ab-initio Monte Carlo simulations : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Physics at Massey University, Albany, New Zealand
(Massey University, 2019) Smits, Odile R.
This thesis describes simulations to determine the melting temperatures of the noble gases based on first-principles ab-initio methods. The melting temperatures of bulk krypton, xenon, radon and oganesson are determined using parallel-tempering Monte Carlo with the interaction potential approximated by two- and three-body contributions. The employed interaction potentials are obtained from relativistic coupled cluster theory including spin-orbit coupling and are the most accurate ab-initio potentials to this date. These potentials are fitted to computationally efficient functions utilized to calculate the interaction energy during the Monte Carlo melting simulation. Two different techniques of obtaining the melting temperature are presented. First, the melting temperature is studied by simulating finite clusters in a canonical ensemble. The melting temperature is then deducted from extrapolation of the finite cluster results to the bulk. Second, the melting temperature is determined by direct sampling of the bulk using cells with periodic boundary conditions in the isobaric-isothermal ensemble. Upon correction for superheating, an excellent agreement to the melting temperatures obtained from cluster simulations is obtained. The numerically determined melting temperatures of krypton and xenon are in close agreement with available experimental data. That is, for krypton a melting temperature of 109.5 K and 111.7 K is obtained for cluster and periodic simulations respectively, which is approximately 5 Kelvin lower than the corresponding experimental value of 115.78 K. The melting point of xenon is determined to be 156.1 K and 161.6 K respectively, which compares to the experimental value of 161.40 K. The long debated value of the radon melting temperature of 202 K is confirmed by our simulations (200 K for both techniques). And finally, the melting point of oganesson is determined to be 330 K and therefore surprisingly high compared to the other rare gases. This implies that oganesson is a solid at room temperature. Furthermore, an analytical formula to compute the temperature of the solid-liquid phase transition based on the analytically expressed bulk modulus and interaction potential is presented, and the superheating correction factor is evaluated.
Network models of mindfulness : a thesis presented in partial fulfilment of the requirements for the degree of Doctor in Clinical Psychology at Massey University, Albany campus, New Zealand
(Massey University, 2019) Smith, Joseph Hendry
Contemporary mindfulness research at the level of self-report has often represented mindfulness as a latent (trait) variable. Recently, a novel psychometric methodology has been developed which allows mindfulness to be modelled as a complex system or network at the level of self-report. This network perspective is argued to provide a more plausible conceptualisation of mindfulness. A network perspective implies that a more densely connected network of practices may be indicative of a greater level of development of mindfulness. It also implies that certain practices may be more strongly interconnected or central than others. These highly central practices may be potentially useful targets for interventions. Mindfulness networks were estimated for practitioners and non-practitioners using the Friedberg Mindfulness Inventory (Study 1) and an adapted version of the Applied Mindfulness Process Scale (Study 2). A total of 371 regular mindfulness practitioners, 224 non-practitioners and 59 irregular practitioners were recruited online from the Amazon Mechanical Turk database. Across both measures, comparisons between practitioners and non-practitioners’ networks indicated that network density did not significantly differ, whereas evidence was found in support of a significant difference in network structure. Exploratory analyses revealed practitioners’ networks to be characterised by greater differentiation in their community structures relative to non-practitioners across both measures. In Study 1, Acceptance was revealed to be much more central to the practitioners’ network relative to non-practitioners; and Returning to the Present much more peripheral. The practice of Attending to Actions and/or the negative path it shared with Self-kindness were identified as possible targets to facilitate mindfulness in non-practitioners. In Study 2, highly eudemonic practices were revealed to be more central to the practitioners’ network relative to non-practitioners, whilst more foundational de-centering practices were more peripheral. These studies provide support for the plausibility of investigating mindfulness as a complex network at the level of self-report. However, the lack of difference in network density indicates that future research is needed to examine network dynamics in the context of regular mindfulness practice. Future research is also required to establish whether the networks estimated are behavioural or semantic.
Novel methods to characterise texture changes during food breakdown : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosphy in Food Technology at Massey University, New Zealand
(Massey University, 2018) Ng, Cui Fang, Grace
The purpose of the mastication process is to break down food for bolus formation so that it can be swallowed safely. Although light has been shed on the criterion for a swallow safe bolus, quantifying these in terms of the bolus properties is not fully understood. There is a lack of suitable measurement techniques to quantify these identified bolus properties. Thus, the purpose of this work was to develop novel techniques that would be useful in in-vitro studies of food breakdown for the characterisation of bolus properties. A mastication robot (MR) had been previously developed to enable the reproducible mastication of food so that masticatory efficiency and food breakdown dynamics can be assessed quantitatively. To evaluate if the MR could be a controllable and reproducible alternative to subjects for food break down studies, a series of experiments involving the mastication of peanuts using a range of machine parameters was conducted. The bolus particle size distributions were used to characterise the breakdown of the peanuts. There were significant differences in the average particle size of the particles chewed by the different chewing trajectories during the initial chews. The performance of the mastication robot was validated against human subjects (n=5) by comparing the particle size distribution (PSD) of peanut boluses collected from subjects and the MR. Although the MR was unable to achieve similar breakdown capability as that for the human subjects, the MR proved to have good reproducibility in bolus preparation. Two novel techniques were developed to characterise bolus properties. The slip extrusion test was developed to characterise two determinant properties for safe swallowing, the bolus deformation and slippage properties. The test measures the force needed to extrude a bolus through a test bag imitating the swallowing action of a bolus. The multiple pin penetrometer was previously developed to measure the spatial distribution of texture in foods exhibiting heterogenous structures. The forces experienced by each pin is measured independently as they pushed through the food, providing a pressure distribution for each food. This allowed the characterisation of fibrous (non-fracturable) foods in a similar way to PSD analysis, offering a method to characterise boluses that do not form discrete particles. The variability in the structure of the boluses was also characterised using the grey level co-occurrence matrix through the image textural features: contrast, energy and homogeneity. Finally, these developed novel techniques were applied to five real foods with varying textures to show how the MR and these techniques may be used to characterise the changes in bolus properties across the mastication stages. Subjects (n=5) were asked to masticate the foods to determine their chewing behaviour and the bolus properties (deformation and slip properties) at swallow point. The chewing parameters from the median subject (subject A) was used to establish the parameters for the mastication robot’s set up for the factorial design of experiments. The developed models from the factorial study were used to optimize the conditions needed for the MR to achieve boluses with similar DR and SR properties as subject A. The five foods were then broken down using the MR configured in this way, and bolus properties were evaluated at various stages of the mastication process through the application of the slip extrusion test, textural mapping using the multiple pin penetrometer, and the back-extrusion test. Factor analysis was applied to the various data collected, which showed that the properties related to the hardness, swallowability and homogeneity attributes were best at describing the changes in the boluses as they were masticated to swallow point. In conclusion, the mastication robot could be used to replicate human chewing trajectories to consistently produce boluses in a controlled trajectory with controlled “simulated saliva” rates throughout the various stages of mastication. Thus, it is relevant as a tool to produce boluses for comparative analysis especially for studies investigating the properties of boluses collected from various stages of the mastication process. In addition, the developed characterisation techniques could be used to track the dynamic changes in the bolus properties for most of the mastication stages from initial chews to the swallow point and beyond that.
Policy analysis using microsimulation : a thesis presented in partial fulfilment of the requirements for the degree of Master of Arts, Social Policy, at Massey University
(Massey University, 1993) Bade, Keith
Recent changes in the social policy development arena in New Zealand mean that traitional methods of social policy analysis are not now adequate for all analyses. Microsimulation is a technique that can provide another dimension to social policy analysis. The thesis starts by discussing some of the major social policy developments in New Zealand pointing out some of the weaknesses in the analyses accompaning them. The thesis then goes on to introduce microsimulation as a technique that can help improve the analysis of social policy. However, the main body of the thesis consists of the development of a microsimulation model, a discussion of the database upon which the model is based, and an analysis carried out using the model. The thesis demonstrates the usefulness of microsimulation models in identifying impacts of social policy changes on small sectors of the population. It does this by simulating the income effects of the increase in the qualification age for National Superannuation on the population sector aged sixty to sixty - five. Although the thesis demonstrates the effectiveness of microsimulation models, the project uncovered a number of areas where currently available data are not sufficiently adequate for the methodology to be utilised to the full. The thesis finishes by suggesting a number of areas where further development could be productive and assist in improving the quality of social policy analysis.
Action-selection in RoboCup keepaway soccer : experimenting with player confidence : a thesis presented in partial fulfilment of the requirements for the degree of Masters of Science in Computer Science at Massey University
(Massey University, 2006) Neilson, Samara Ann
Through the investigation of collaborative multi-agent domains, in particular those of robot soccer and robot rescue, and the examination of many popular action-selection methodologies, this study identifies some of the issues surrounding entropy, action-selection and performance analysis. In order to address these issues, a meaningful method of on-field player evaluation, the confidence model, was first proposed then implemented as an action-selection policy. This model represented player skill through the use of percentages signifying relative strength and weakness and was implemented using a combination of ideas taken from Bayesian Theory. Neural Networks. Reinforcement Learning, Q-Learning and Potential Fields. Through the course of this study, the proposed confidence model action-selection methodology was thoroughly tested using the Keepaway Soccer Framework developed by Stone, Kuhlmann, Taylor and Liu and compared with the performance of its peers. Empirical test results were also presented, demonstrating both the viability and flexibility of this approach as a sound, homogeneous solution, for a team wishing to implement a quickly trainable performance analysis solution.

Massey Documents by Type

Browse

Filters

Settings

Sort By

Results per page

Search Results