Explore a range of project management research internships to complete as part of your degree during the semester break.
The following internships listed are due to take place across the Summer break.
Applications open 15 September and close at midnight on 4 October 2026.
Supervisor: Dr. Sujuan Zhang, Xuanqi Li
Eligibility: Distinction average, Qualitative data analysis, Desk research skills, Interest in project studies and organization theory.
Project Description:
Data centers form the information backbone of an increasingly digitalized world, with demand rising rapidly alongside artificial intelligence, distributed manufacturing, and autonomous vehicles. In Australia, data centers could account for around 6% of grid-supplied electricity by 2030 and 12% by 2050.
Their rapid expansion, however, has also intensified social and environmental concerns. While data centers can support energy transitions through renewable energy procurement and energy storage, they also raise issues around water competition, unevenly distributed costs and benefits, and local impacts such as noise.
Data center projects therefore sit at the intersection of digital and sustainability transitions, whose anticipated synergies do not always materialize in practice. Drawing on paradox theory, this study examines tensions emerging in the delivery and operation of Australian data center projects (drawing on case studies in Australia) and how project actors navigate these tensions.
Requirement to be on campus: No
Supervisors: Dr. Sujuan Zhang, Xuanqi Li
Eligibility: Distinction average, Qualitative data analysis, Desk research skills, Interest in project studies and organization theory
Project Description:
Project cancellation is treated as the final chapter of a project’s life—its death. However just as stories of past heroes continue to be told after they are gone, cancelled projects can reappear in media discourse after their physical future has ended. They linger like project ghosts: no longer materially alive, yet still present in public narratives.
This study takes this phenomenon as a starting point and investigates the discursive life of cancelled projects. Drawing on global collection of cancelled projects that repeatedly resurface in media reporting, it asks two key questions: why do these project “ghosts” persist, and how do they persist?
The study frames the discursive life as a form of project legacy. It argues that project legacy extends beyond organizational lessons learned to include how cancelled projects are socially constructed in public narratives. In doing so, the study reconsiders when a project truly dies.
Requirement to be on campus: No
Supervisors: Prof. Stewart Clegg, Dr. Xinyue Zhang
Eligibility: Excellent written skills in English, with the ability to produce clear and well-structured academic writing.
Project Description:
This project investigates how digital narratives shape public engagement with major Australian infrastructure projects. Working within Australian Research Council (ARC) Discovery Project DP260103106, the intern will identify and systematically collect publicly available online material relating to selected major infrastructure cases. Sources may include project websites, government publications, news media, community forums and public social media posts. The intern will develop search terms and a transparent data collection protocol, organise and document the material, and conduct preliminary thematic and narrative analysis. The research will examine how project organisations, governments, communities and activists frame project purposes, risks, benefits, legitimacy and future impacts. The internship will contribute to a comparative digital ethnography of infrastructure governance while providing practical experience in qualitative research, digital data collection and analysis.
Requirement to be on campus: No
Supervisor: Dr. Xiaoshan Zhou
Eligibility: Students should have a background in computer science. Experience in PyTorch, diffusion models, and/or 3D human modelling is preferred.
Project Description:
Long-horizon human activity video generation remains challenging due to limitations in maintaining temporal, spatial, and physical consistency across consecutive actions. When multiple human activities are generated and composed into a continuous sequence, although the semantic state of the activity may have changed, the corresponding body dynamics may not complete a physically and kinematically plausible transition. These inconsistencies can lead to abrupt pose discontinuities, foot sliding, body instability, and visible artefacts in the generated video. To address these challenges, students will extract 3D human skeletal models and motion representations from videos, identify and correct implausible action transitions using kinematic and physical constraints, and integrate the refined motion sequences with diffusion models to generate videos with improved temporal continuity, physical plausibility, and visual consistency. The research group has access to substantial GPU resources to support model development and experimentation.
Requirement to be on campus: No
Supervisors: Dr. Xiaoshan Zhou
Eligibility:
This project welcomes students with an interest or background in one or more of the following areas:
Project Description:
Reinforcement meshes and rebar cages are critical components of reinforced concrete structures. Their inspection requires engineers to verify dimensions, reinforcement quantities, bar spacing, and installation quality against design requirements. This process currently relies heavily on manual measurement, photographic documentation, and report preparation. This project aims to develop a prototype multi-agent system to automate key stages of reinforcement inspection. Students will collect experimental data using depth cameras and 3D LiDAR point clouds, and construct a structured database linking site observations with design drawings, specifications, and engineering guidelines. Computer vision and point-cloud processing methods will be used to extract geometric features and relevant reinforcement parameters. Specialised AI agents will then coordinate data processing, retrieval-augmented information access, compliance checking, anomaly identification, and automated report generation. The expected outcome is an integrated proof-of-concept workflow that connects field data acquisition, design verification, and digital reporting to support more efficient, consistent, and traceable quality inspection.
Requirement to be on campus: No
Supervisor: Dr. Xiaoshan Zhou
Eligibility: Students should have a background in mechanical engineering, robotics, mechatronics, industrial design, or a related discipline, with experience in 3D printing and/or VR/Unity.
Project Description:
This project aims to develop a robotic platform capable of locomotion and bimanual manipulation in unstructured environments. Students will use 3D printing to design and fabricate a mechanical locomotion system that can move across a ladder-like structure, with an adjustable step length that enables the robot to reach successive support positions with varying spacing. They will develop the mechanical design and integrate motors, actuators, sensors, microcontrollers, and basic control algorithms. The robotic structure should maintain balance and stability during locomotion and support a mounted dual-arm manipulation system without tipping, overturning, or undergoing excessive structural deformation. Working as part of a team, students will also use VR-based human demonstrations and imitation learning to teach the dual-arm robot to perform coordinated wrapping, twisting, fastening, and securing actions.
Requirement to be on campus: No
Supervisor: Dr Jin Xue
Eligibility: Experience with Python or R; quantitative analysis; statistics; GIS or spatial data desirable; interest in urban analytics and human mobility.
Project Description:
Online reviews shape how people choose and evaluate public, recreational, religious and commercial social spaces. Yet paid, incentivised, or coordinated activity may distort measures of experienced quality. This project will develop an AI-enabled framework for identifying potentially manipulated reviews. The student will combine large language models (LLMs), embeddings and anomaly detection to examine duplicated language, rating-text inconsistency, temporal bursts and reviewer- or place-level patterns. An agentic AI workflow will assign specialised agents to analyse linguistic, behavioural and temporal signals, while retrieval-augmented generation (RAG) will ground assessments in platform policies, manipulation strategies and similar reviews. Because observational data cannot prove a review is fake, classifications will report uncertainty and be evaluated using weak supervision and manual validation. The project will test whether excluding or down-weighting suspicious reviews changes place- and area-level quality estimates. Outputs include a reproducible pipeline, a validation dataset, visualisations, and a research report.
Requirement to be on campus: Yes *dependent on government’s health advice
Supervisors: Dr Jin Xue
Eligibility: Literature studies; Qualitative analysis; Academic writing; Critical thinking; Theorisation
Project Description:
How can project leaders make high-stakes decisions on major projects when the available data are incomplete, fragmented or even contradictory? This research project investigates how artificial intelligence and knowledge-management systems can address this issue and help project, program and portfolio professionals improve decision quality, forecasting and project outcomes. By working on qualitative interviews and case studies involving practitioners from government, Defence and industry, it will offer you the opportunity to engage with real-world perspectives on project controls, risk, organisational culture, data quality and the responsible use of AI. You will contribute to interview preparation, transcript review, qualitative data organisation, thematic coding and the synthesis of research findings, while developing valuable skills in applied research, critical analysis and evidence-based problem-solving.
Requirement to be on campus: Yes *dependent on government’s health advice
Supervisor: Dr. Jin Xue
Eligibility:
Project Description:
This project will investigate how Large Language Models (LLMs), graph-based retrieval, and reinforcement learning will transform organizational meetings into an intelligent decision-support system. Students will build a Conversation-as-Data Retrieval-Augmented Generation (CAD-RAG) pipeline that will follow meetings over time, convert audio into structured and searchable organizational memory, and generate minutes, summaries, evidence-based analyses, and actionable recommendations for leaders. The system will integrate speaker diarization, automatic speech recognition, knowledge extraction, temporal knowledge graphs, and graph-based retrieval. A reinforcement-learning component will learn from expert feedback and decision outcomes to improve evidence prioritisation, action recommendations, and adaptation to organisational preferences. The system will be benchmarked against HippoRAG and GraphRAG using the AMI Meeting Corpus and a curated question-answering and decision-making benchmark. Evaluation will combine LLM-as-a-Judge and human assessment. Students will gain hands-on experience in LLM agents, graph AI, reinforcement learning, and human-centred decision intelligence while developing a publishable prototype with strong real-world potential.
Requirement to be on campus: Yes *dependent on government’s health advice
Supervisor: Dr. Jin Xue, Xinyi Xu
Eligibility:
Project Description:
Australia’s renewable energy transition will require major expansion of solar, wind and associated infrastructure. A key challenge is understanding how this future development may change landscapes, and where those changes may be concentrated.
Students will work with satellite imagery, GIS datasets and renewable energy scenarios. AI-assisted image analysis will be used to compare before-and-after satellite images of existing renewable energy sites and estimate changes in land cover and landscape condition. The project will also explore how large language models (LLMs) and multimodal AI can support geospatial analysis by linking information from imagery, spatial data and metadata, and assisting with parts of the GIS workflow.
Students will gain hands-on experience with satellite imagery, geospatial datasets, GIS, remote sensing and AI-assisted spatial analysis, and learn how AI-derived information can be incorporated into GIS workflows to support large-scale environmental and energy planning.
Requirement to be on campus: Yes *dependent on government’s health advice.
Supervisor: Hoonyong Lee
Eligibility:
Project Description:
Monitoring crowd density is important for improving safety, comfort, and operational management in public environments such as transport hubs, events, and large buildings. Traditional crowd monitoring methods often rely on cameras or manual counting, which may raise privacy concerns or require significant infrastructure.
Recent studies suggest that wireless signals such as WiFi can provide an alternative sensing approach for estimating crowd density. The presence and movement of people influence WiFi signal characteristics, making it possible to infer crowd levels through signal analysis.
This project explores the feasibility of using WiFi-based sensing to estimate crowd density in outdoor environments. The student will investigate how wireless signal data can be used to infer the presence and concentration of people, and explore simple machine learning or statistical methods for estimating crowd levels.
The project will provide hands-on experience in sensing technologies, data analysis, and smart environment monitoring.
Requirement to be on campus: Yes, *dependent on government’s health advice
Supervisor: Dr. Hoonyong Lee
Eligibility:
Project Description:
Remote work can be convenient, but it may be difficult to recognise when working relationships become uncomfortable. If such issues go unnoticed or remain unresolved, they can negatively affect work over time. Wearable sensors can capture changes in physical responses in these situations and may help provide timely support.
This project aims to design and develop a mobile intervention system using wearable sensor data. The student will plan how the system will process collected sensor data and deliver brief intervention messages to users’ mobile phones. Based on this plan, the student will develop and test a functional prototype.
The project provides hands-on experience in wearable sensor data processing, mobile message delivery, software system design, and prototyping.
Requirement to be on campus: Yes, *dependent on government’s health advice
Supervisor: Dr Hoonyong Lee
Eligibility:
Project Description:
Falls remain among the most common and costly safety incidents in workplaces and everyday environments. One promising prevention strategy deploys autonomous robots to explore these spaces and detect potential fall hazards before anyone encounters them.
Quadruped robots are strong candidates for this role. They can traverse most of the spaces people walk through, and by continuously logging their own motion, they can detect a hazard from the abnormal pattern it produces as the robot crosses it.
This project tests a quadruped robot's motion in a virtual simulation environment as it traverses different surface conditions at different speeds and gait patterns. The goal is to demonstrate the feasibility of using a quadruped robot to detect fall hazards, laying the groundwork for real-world field validation.
Requirement to be on campus: Yes, *dependent on government's health advice.
Supervisor: Dr Hoonyong Lee
Eligibility: Basic programming skills (e.g. JavaScript, Python, or similar), some experience with 3D modelling. Interest in virtual reality, safety, or human-centred design
Project Description:
Why do people fail to act safely even when they clearly recognize a hazard? This project explores the idea that risk grows when a person's capacity to regulate movement declines while attention stays intact.
Using VR simulation, the project will induce a decline in self-regulation and test how it affects balance and walking. VR eye-tracking and wearable motion sensors will capture attention (gaze) and movement control (gait and balance) together.
The project will provide hands-on experience in building and piloting a VR scenario, developing a sensor-analysis pipeline, and reviewing the literature on which movement and gaze measures reflect regulatory control.
Requirement to be on campus: Yes, *dependent on governement's health advice.
Last updated 12 September 2026.
