Advancements in the development of deep learning and computer vision-based approaches have the potential to provide engineers with the ability to improve the performance of their projects. This research aims to design and develop a deep learning and computer vision-based framework for engineering projects. The research objectives are: (1) considers current progress on computer vision and deep learning; (2) identiy the research challenges that can materialize with using deep learning to in engineering; and (3) provide a signpost for future research in the emergent area of deep-learning within the context of engineering.

There is high demand for heavy equipment in civil infrastructure projects and their performance is a determinant of the successful delivery of site operations. Although manufacturers provide equipment performance handbooks, additional monitoring mechanisms are required to depart from measuring performance on the sole basis of unit cost for moved materials. Vision-based tracking and pose estimation can facilitate site performance monitoring. PhD research projects will develop regression-based deep neural networks (DNNs) to monitor equipment with the aim of ensuring safety, productivity, sustainability and quality of equipment operations

Civil infrastructure projects are associated with many site accidents and severe injuries. Proper safety training that builds upon hazard perception of site crews will help to increase safety performance and reduce site accidents. Towards this aim, this PhD research will develop a framework to facilitate a context aware safety training for site processes. This framework is then evaluated for its applicability by creating safety training scenarios using building information models (BIM), virtual reality (VR) and Augmented reality (AR).

The construction industry is one of the least automated industries that feature manual-intensive labor as a primary source of productivity. Whether it’s new commercial construction, renovation or demolition, robots don’t yet play a significant role in any step of a building’s lifecycle.

The application of interactive robots in construction are being tested at Monash University to change thie current situation in the industry.

As a highly unautomated industry, construction robots will have a major impact on productivity and safety of the industry. As construction companies look to automate more and more tasks, demand for construction robots will grow steadily. This project aims to address this demand.

Design for construction and assembly (DfCA) optimises the process of product design and development in construction. Main objectives of DfCA are to improve tangible performance measures in construction and assembly such as structural, environmental and temporal attributes. Among construction subsectors, off-site prefabrication provides significant potentials for implementation of DfCA. Similar to manufacturing, product design in off-site is followed by production operations, assembly and installation. Adopting DfCA principles such as reducing the number and weight of subassemblies has been proved effective in optimisation of construction and on-site assembly.

There is need for research to quantitatively analyse impacts of DfCA in construction settings, which leads to optimisation of decision making. On this basis, the current research views and models DfCA as an optimisation problem with defining design, construction and assembly requirements as constraints. In a novel modelling initiative, structural, environmental and temporal attributes are quantified and linked to the achievement of DfCA objectives and satisfaction of constraints.

Infrastructure projects provide required physical and organizational facilities for communities. Information flows in such projects are complex especially when a hybrid of on-site and off-site processes is in progress. With infrastructure projects still experiencing budget and time overruns, there is need to re-examine the information flows.

There is need for research to develop robust frameworks for information flows in infrastructure projects. On this basis, a three-dimensional view of construction production including portfolio, process and operation aspects is required to improve performance measures in design, construction, operation and maintenance. Such improvements include but are not limited to minimised rework and re-entrant flows, flexible construction, enhanced multidisciplinary collaborations, and efficient maintenance. This research contributes to the body of knowledge by examining information flows in complex infrastructure settings.

Robust supply decision making is critical to the advanced manufacturing of prefabricated products. Mainstream research has focused on minimising cost overruns in off-site construction supply networks by optimising purchasing decisions. However, decision parameters such as strategic preferences to include or exclude certain suppliers and utilisation of multi-supplier configurations are yet to be formulated and analytically solved. This project aims to enhance supply network performance with smaller overall investment.

Toward this aim, the project develops and tests several research hypotheses on optimisation of supply decisions and configurations. Real-world prefabrication projects serve as the test bed to demonstrate the effectiveness of the analysis and analyse cost implications of supply related decisions. The modeling method and results contribute to optimal decision making in advanced manufacturing of prefabricated products.

Hybrid infrastructure projects are defined as triads of on-site/coordination/off-site project dimensions. Interaction of uncertainties in such settings result in deviations from project objectives by causing time and cost overruns, safety issues, quality deficiencies, technical problems, and lack of client satisfaction. To address these, a holistic approach in identifying and analysing risks in hybrid (multi-dimensional) projects is required. Towards this aim, the current project develops and tests several research hypotheses using real-world data from infrastructure projects.

Practical implications of triadic risk analysis in hybrid infrastructure projects will inform optimum decision making in project settings. An integrated approach to risk management will decrease the chance of deviations from project objectives.