Engineering Science

Project

The objectives of groundfish surveys are to provide fishery-independent abundance indices of fish species available to bottom trawling and to collect biological samples of selected species. Groundfish surveys usually cover relatively large areas and the stations at which the fish are caught are numerous and often widespread. Further, the variability in the estimate of abundance - known as the coefficient of variance (CV) - needs to be under a given threshold. To reduce the CV more stations must be added to the survey. This increases the cost of the survey as it takes longer to complete and uses more fuel.

This project will investigate the trade off between CV and survey cost - given a budget which stations should be visited and what CV can be achieved, or if a certain CV must be achieved how much will it cost? In exploring this trade-off a key complication lies in determining how much surveying a given set of stations costs, as finding the order in which to visit the stations is a difficult optimisation problem. Therefore this project will also examine solving the optimisation problem efficiently and resolving it with one, or several, stations added, or removed.

Project

Conceptual modelling is a crucial part of the simulation process in which a model is abstracted from the real world situation and results in the conceptual model, a description of the simulation model including the inputs, outputs, and components. Hierarchical Control Conceptual Modelling (HCCM) is a particular form of conceptual modelling designed to make it easier to model systems with complex control logic. A standard has recently been developed for HCCM that describes the requirements of an HCCM conceptual model.

This project will work with and enhance practical conceptual modelling tools that have been developed to assist in the use of, and improve compliance with, the HCCM standard. The tools enable electronic representation of the conceptual model, including activity diagrams. There is potential demonstrate the effective use of the tools and to extend them to enable various output viewpoints and formats.

Project

Heart failure presents a significant social and economic burden worldwide. Identification of raised cardiac filling pressure is critical for the correct diagnosis of heart failure, but the existing gold-standard assessment procedure is invasive and not every patient is eligible for this procedure. While it is possible to estimate these pressures using non-invasive techniques such as echocardiography, the diagnostic accuracy can be as low as 25Ôªø %. Are the current guidelines adequate, or can we do better?

Using a unique dataset of invasive pressure measurements and echocardiography images collected from over 200 patients across New Zealand, this project will explore new methods for non-invasively estimating filling pressure.
Students will join a multi-disciplinary team of bioengineers, clinicians, and imaging specialists. This project will suit students with an interest in translational research, medical image analysis, and developing tools for clinical decision-making. Prior medical imaging experience is not required.

As part of a NZ Government funded programme on biomechanical modelling of the heart, outputs of this project will contribute to the development of more efficient and effective strategies for the diagnosis and monitoring of heart failure.

Project

Practical demonstrations and devices have always been key to learning and understanding the fundamentals of cardiac disease. In this project, we aim to create an array of devices and haemodynamic models by combining 3D printing, machining, programming, and electrical circuits. These devices and models will enable the next generation of bioengineers to be inspired, while increasing health literacy in the community.

Students will join a multi-disciplinary team of bioengineers, clinicians, and device manufacturing experts. This project will suit students with an interest in translational research, device manufacturing, and hands-on experience in an engineering environment. Prior CAD, machining, and electrical knowledge are not required.

Project

Supra-systolic oscillometric waveforms, captured by devices such as the USCOM BP+, offer a non-invasive and cost-effective alternative to invasive pressure measurements, holding the potential to improve patient outcomes and reduce healthcare costs in the diagnosis and monitoring of heart disease. However, these waveforms require reshaping to resemble invasive, brachial pressure waveforms, and subsequently transformed into central pressure waveforms. The current reshaping model leaves room for improvement. This project aims to enhance the reshaping process using an existing substantial dataset of BP+ waveforms and corresponding catheter pressure waveforms from patients at Auckland City Hospital.

The main objectives of this project are to design and test various reshaping functions using empirical data;, train and cross-validate the reshaping functions using regression/machine learning techniques;, and model approaches to narrow down the possible forms of the reshaping function.

This project would be an excellent opportunity for a student looking to gain hands-on experience in biomedical research and who is interested in contributing to the advancement of medical device technology.

Project

This project will enhance OpenSolver (http://opensolver.org), a popular open source optimiser for Excel developed by Engineering Science that has been downloaded over 640,000 times and is used by companies and universities around the world.

This project will make a number of improvements to OpenSolver. These will include improved approaches for modelling non-linear problems, improved support for explicit modellling using an A-matrix, adding support for new solvers (including HiGHs), and developing new tools for solving problems with more than one objective using Excel. Solving these latter problems requires showing users a set of possible solutions, and letting them easily understand the trade-offs in these solutions. We will need to develop good solution algorithms and new visualisation tools to let the user explore the solution set.

If time allows, this project will also make enhancements to SolverStudio (http://solverstudio.org), a companion project that provides a complementary set of modelling tools using languages such as Python and Julia.

The student needs good skills in programming. The scope of this project will be adjusted to match with the student's skills. An understanding of linear and integer programming, VBA and/or C# would be helpful, but are not required.

Project

We are progressing a National Science Challenge project that involves developing analytics and optimisation tools that can help identify missing shareholders. Many Māori organisations have a growing number of shareholders and face the very real challenge of keeping track of these shareholders. This is not easy, as demonstrated, for example, by the over $4,000,000 of dividends that the Taranaki-based organisation, Parininihi ki Waitotara (PKW), has owing to missing shareholders. Similar problems are faced by the Māori Trustee who is tasked with managing Māori land, where each land parcel can have 100's of owners.

We are developing analytics tools that use public information to help organisations such as PKW better understand and locate its missing shareholders. The challenge we face is to develop mathematical Bayesian inference models to process large amounts of often messy data to identify links between people. Our work exploits advances in analytics and new graph-based database approaches.

We are seeking students with good programming and/or mathematical modelling skills to work with us in developing these tools, and creating interactive online visualisations and other tools for interacting with our data and models. This includes working with and extending our Unity games engine 3D visualisation tools, and preparing these for delivery to PKW. The scope of the project will be adjusted to best match the skills of the researcher with the wider project goals.

The successful applicants will join our small research team and gain skills in analytics, inference and visualisation along with practical experience in using modern industry-based software tools that can be applied to a wide range of problem domains.

We are seeking two students to contribute to this research project.

Project

Multicriteria Decision Aiding is applied when making decisions where a finite number of options are available and need to be evaluated under multiple criteria. For instance when we decide between different options for travelling we may trade off the cost, travel time and emissions generated by our trip. There is usually not a single option that is best with respect to all criteria and a preferred trade-off needs to be identified. There are various well-known approaches to tackle these decision problems such as multi-attribute value theory, analytic hierarchy process or outranking methods. Andrea teaches these methods in ENGSCI 755, but prior knowledge is not required for this summer project. In 2020, we developed a decision support toolkit for some of these decision making approaches with a graphical user interface as a stand-alone application. This tool will be used in teaching and delivery of ENGSCI 755. While there are different software packages available they generally do not include different decision making approaches but instead focus on a single one, which means students in ENGSCI 755 have to learn about multiple different software packages. Ultimately we will develop a single stream-lined package that allows ENGSCI 755 students to learn about different multicriteria decision aiding methods within one package. In this summer project we will continue development of the existing software to include other decision making techniques. We will also apply this to work through some case studies.

Requirements: Prior knowledge of decision making methods is not required, and the student working on the project can learn about these over the summer as needed. Working on this project does require a confident programmer. The existing tool is implemented in Python.

Project

A multiobjective shortest path problem is a shortest path problem the aims to identify good tradeoff solutions of a problem with two or more objectives, for example trading off cost and time. Multiobjective shortest path problems are commonly solved when finding good public transport routes (aiming for low cost, low time, few transfers between services), or in logistics (aiming for low cost, time, and emissions). In this project we will improve multiobjective shortest path algorithms by testing some ideas for speedup techniques for single objective (standard) shortest path problems, such as landmarks, contraction hierarchies, A* search, etc. Different techniques will be adapted to the multiobjective case, implemented and tested on a large testbed of problem instances.

Requirements: Existing code is implemented in C++. The student working on this project needs to be a confident programmer who is able to work with the existing code.