Optimisation of prefabricated modular-integrated residential construction using hybrid customisation methods

With a growing population, Australia's housing demand is expanding year after year. Nevertheless, in many parts of Australia, housing is unaffordable for Australians on average incomes, locking out new generations of Australians from their dream of homeownership. In addition, the Australian construction industry is dominated by traditional building systems that are expensive and inefficient compared to other methods. Thus, as a result of their affordability and productivity, prefabricated building systems are gaining popularity. The prefabricated building sector is relatively new and comprises a small proportion of the current construction industry in Australia. Current prefabricated systems do not address the needs of their clients for personalisation. Due to several reasons, current construction companies use prefabrication systems to work purely based on either volumetric or non-volumetric methods. However, these individual systems have developed their efficiencies in isolation without considering the greater potential of using hybrid methodologies. This research aims to develop an efficient prefabricated modular-integrated hybrid solution that will use the best elements of the prefabrication systems to facilitate serial production with enhanced performance at lower costs while satisfying the unique requirements of each household to guarantee customer satisfaction. It will use innovative methods, including modularisation and design for manufacturing and assembly (DfMA) techniques, advanced digital engineering tools such as Building Information Modelling (BIM), and computational design tools for design automation and multi-objective optimisation. The Architecture, Engineering, and Construction (AEC) industry is slowly transitioning toward using digital engineering methods for design, advanced manufacturing systems, and construction automation. This work uses the built-in intelligence of building information models to introduce a novel systematic, algorithmic approach to identify optimum module divisions for the decomposition of designs with minimum human intervention. The research explores the development of a rule-based smart building solution using a genetic algorithm (GA) based on design structure matrix (DSM), artificial intelligence (AI): fuzzy logic and generative design automation for optimal decomposition of the design for modules and smart assemblies, using innovative methods. In this process, a designer will be able to define the design parameters and design constraints of the module geometry, enabling the use of them in the modularisation algorithm. These modularisation rules enable modular coordination (MC) and module standardisation (MS), facilitating serial production for a project. Incorporation of the geometry engine and MC rules into BIM authoring tools will streamline the process and enable designers to automate various sophisticated design modelling tasks and optimise those designs to achieve goal-driven outcomes. The new integrated system will identify project-specific optimum module designs, lowering the assembly cost and handling time and product configuration for modules and smart assemblies of hybrid construction methods. The proposed hybrid system will use parametric unitised structural modules with smart non-volumetric assemblies and components to increase the flexibility of prefabricated residential housing. Furthermore, the new design integrated construction system allows the user to reverse engineer all the designs to achieve a flexible prefab solution.