{"id":145095,"date":"2026-03-17T14:22:03","date_gmt":"2026-03-17T06:22:03","guid":{"rendered":"https:\/\/www.curtin.edu.au\/research\/?post_type=hdr-r-projects&p=145095"},"modified":"2026-03-17T14:22:03","modified_gmt":"2026-03-17T06:22:03","slug":"nonlinear-scheduling-optimisation-for-green-hydrogen-production","status":"publish","type":"hdr-r-projects","link":"https:\/\/www.curtin.edu.au\/research\/hdr-r-projects\/nonlinear-scheduling-optimisation-for-green-hydrogen-production\/","title":{"rendered":"Nonlinear scheduling optimisation for green hydrogen production"},"content":{"rendered":"\n
\"\"<\/figure>\n\n\n\n

Project Overview<\/strong><\/p>\n\n\n\n

Nonlinear Integer programming is widely used to solve complex decision problems in scheduling, logistics, supply chain management, and optimal control. These problems suffer from the well-known curse of dimensionality, whereby the size and scale of the feasible space grow rapidly, often outpacing the available computational power. This project aims to exploit recent breakthroughs in variational analysis, previously restricted to the realm of continuous optimisation, to develop new solution methodologies for large-scale mixed-integer nonlinear programming problems. In recent years, variational analysis has provided the theoretical foundation for most modern methods in continuous nonlinear optimisation, such as bundle methods, proximal methods, and gradient descent-type methods. Yet, very little of this theory has crossed over to the discrete domain. This is because the notions of gradient and sub-gradient, the cornerstones of variational analysis, are commonly believed to have no meaning in discrete space. Recent developments by and Dr Hoa Bui have shown that many key results in variational analysis can be formulated in general frameworks that allow for the application of bundle methods in discrete optimisation. Their approach has then become the state-of-the-art for solving several important classes of discrete quadratic programming: diversity problems, Euclidean max-sum problems etc. This opens the door to potential applications of many continuous optimisation methods in discrete optimisation, creating a solid foundation for developing fast gradient-based algorithms and a better understanding of algorithms \u2018convergence analysis.<\/p>\n\n\n\n

This project will focus on further developing new mathematical optimisation algorithms for a range of discrete nonlinear optimisation problems in Green Hydrogen Production, to enhance overall productivity and reduce green hydrogen production costs. Optimisation problems in this domain are highly nonlinear and of massive scale. The project will leverage recent breakthroughs in integer programming and nonlinear optimisation to create efficient computational algorithms for overcoming this complexity. These algorithms will provide critical insights into optimal operations strategies for potential Australian hydrogen scenarios. The new theoretical developments will contribute to bridging the gap between discrete and continuous optimisation, two fields that are normally studied disparately.<\/p>\n\n\n\n

The selected PhD candidates will be embedded within the Operations Research group at Curtin Centre for Optimisation and Decision Science, who have high quality research profiles and rich experience working with industry.<\/p>\n\n\n\n

Aim<\/strong><\/p>\n\n\n\n

This project will use recent breakthroughs in variational analysis to develop new exact mathematical optimisation algorithms for discrete nonlinear optimisation problems. New solution approaches will open door for embedding nonlinearity in formulations of many real-world problems in industry, especially in Green Hydrogen.<\/p>\n\n\n\n

Objectives<\/strong><\/p>\n\n\n\n

The project outcomes will include scientific articles published in leading journals such as Operations Research, Mathematical Programming, and the SIAM Journal on Optimization, as well as a library of state-of-the-art optimisation algorithms (in Python, Julia, or C++) designed to solve large-scale nonlinear optimisation problems. We expect new nonlinear formulations of many existing industrial optimisation problems, leading to more accurate and higher-performing optimisation models.<\/p>\n\n\n\n

Significance<\/strong><\/p>\n\n\n\n

Over the last 50 years there have been tremendous advances in linear integer programming, culminating in mature commercial solvers such as CPLEX. But progress in nonlinear integer programming has lagged, and there is now a substantial performance gap between linear and nonlinear integer programming methods. Currently, the most common strategy for dealing with nonlinearity is computational demanding, making it prohibitive for tactical decision making where solutions in seconds or minutes are required. This project will help bridge this significant gap by leveraging advanced optimisation theory to design new efficient algorithms for solving large-scale nonlinear discrete optimisation problems.<\/p>\n\n\n\n

Ideal Candidate<\/strong><\/p>\n\n\n\n

This opportunity will provide two full-time on-campus PhD scholarship in the Curtin School of Electrical Engineering, Computing and Mathematical Science. <\/p>\n\n\n\n

This project is open to domestic and international<\/strong> applicants.<\/p>\n\n\n\n

Scholarship<\/strong><\/p>\n\n\n\n

The successful candidate will receive a stipend of $38,440 per annum pro rata (tax free), this amount will be indexed annually.
The duration of the award shall be for three years with a possible extension of up to six months (maximum), assessed on a case-by
case basis.
In addition, this scholarship will include a tuition fee waiver for successful international candidates.<\/p>\n\n\n\n

Applications close\u00a0April 10th 2026<\/strong><\/p>\n\n\n\n

Enquiries<\/strong><\/p>\n\n\n\n

If this project interests you, contact Project lead\u00a0Dr Hoa Bui\u00a0<\/strong>via\u00a0Expression of Interest<\/strong><\/a>.<\/p>\n\n\n\n

<\/p>\n","protected":false},"author":102,"featured_media":0,"template":"","faculties":[51],"research_areas":[5298],"class_list":["post-145095","hdr-r-projects","type-hdr-r-projects","status-publish","hentry","faculties-science-and-engineering","research_areas-data-science-machine-learning-and-ai"],"acf":false,"featured_image":false,"_links":{"self":[{"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/hdr-r-projects\/145095","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/hdr-r-projects"}],"about":[{"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/types\/hdr-r-projects"}],"author":[{"embeddable":true,"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/users\/102"}],"version-history":[{"count":0,"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/hdr-r-projects\/145095\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/media?parent=145095"}],"wp:term":[{"taxonomy":"faculties","embeddable":true,"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/faculties?post=145095"},{"taxonomy":"research_areas","embeddable":true,"href":"https:\/\/www.curtin.edu.au\/research\/wp-json\/wp\/v2\/research_areas?post=145095"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}