Search results

The integration and automation of the whole design and implementation process have become a pivotal factor in construction projects. Problems of process integration, particularly at the conceptual design stage, often manifest through a number of significant areas, from design representation, cognition and translation to process fragmentation and loss of design integrity. Whilst building information modelling (BIM) applications can be used to support design automation, particularly through the modelling, amendment and management stages, they do not explicitly provide whole design integration. This is a significant challenge. However, advances in generative design now offer significant potential for enhancing the design experience to mitigate this challenge.

The approach outlined in this paper specifically addresses BIM deficiencies at the conceptual design stage, where the core drivers and indicators of BIM and generative design are identified and mapped into a generative BIM (G-BIM) framework and subsequently embedded into a G-BIM prototype. This actively engages generative design methods into a single dynamic BIM environment to support the early conceptual design process. The developed prototype followed the CIFE “horseshoe” methodology of aligning theoretical research with scientific methods to procure architecture, construction and engineering (AEC)-based solutions. This G-BIM prototype was also tested and validated through a focus group workshop engaging five AEC domain experts.

The G-BIM prototype presents a valuable set of rubrics to support the conceptual design stage using generative design. It benefits from the advanced features of BIM tools in relation to illustration and collaboration (coupled with BIM's parametric change management features).

This prototype has been evaluated through multiple projects and scenarios. However, additional test data is needed to further improve system veracity using conventional and non-standard real-life design settings (and contexts). This will be reported in later works.

Originality and value rest with addressing the shortcomings of previous research on automation during the design process. It also addresses novel computational issues relating to the implementation of generative design systems, where, for example, instead of engaging static and formal description of the domain concepts, G-BIM actively enhances the applicability of BIM during the early design stages to generate optimised (and more purposeful) design solutions.

This paper aims to explore the emerging relationship between Industry 4.0 (I4.0) digital technologies (e.g. blockchain, Internet of Things (IoT) and artificial intelligence (AI)) and the construction industry’s gradual transition into a circular economy (CE) system to foster the adoption of circular economy in the construction industry.

A critical and thematic analysis conducted on 115 scientific papers reveals a noticeable growth in adopting digital technologies to leverage a CE system. Moreover, a conceptual framework is developed to show the interrelationship between different I4.0 technologies to foster the implantation of CE in the construction industry.

Most of the existing bodies of research provide conceptual solutions rather than developing workable applications and the future of smart cities. Moreover, the coalescence of different technologies is highly recommended to enable tracking of building assets’ and components’ (e.g. fixtures and fittings and structural components) performance, which enables users to optimize the salvage value of components reusing or recycling them just in time and extending assets’ operating lifetime. Finally, circular supply chain management must be adopted for both new and existing buildings to realise the industry's CE ambitions. Hence, further applied research is required to foster CE adoption for existing cities and infrastructure that connects them.

This paper investigates the interrelationships between most emerging digital technologies and circular economy and concludes with the development of a conceptual digital ecosystem to integrate IoT, blockchain and AI into the operation of assets to direct future practical research applications

Successful implementation of infrastructure projects has been a controversial issue in recent years, particularly in developing countries. This study aims to propose a decision support system (DSS) for the evaluation and prediction of project success while considering sustainability criteria.

To predict sustainable success factor, the study first developed its sustainable success factors and sustainable success criteria. These then formed a decision table. A rough set theory (RST) was then implemented for rules generation. The decision table was used as the input for the rough set, which returned a set of rules as the output. The generated rulesets were then filtered in fuzzy inference system (FIS), before serving as the basis for the DSS. The developed prediction tool was tested and validated by applying data from a real infrastructure project.

The results show that the developed rough set fuzzy method has strong ability in evaluation and prediction of the project success. Hence, the efficacy of the DSS is greatly related to the rule-based system, which applies RST to generate the rules and the result of the FIS was found to be valid via running a case study.

Use of DSS for predicting the sustainable success of the construction projects is gaining progressive interest. Integration of RST and FIS has also been advocated by the seminal literature in terms of developing robust rulesets for impeccable prediction. However, there is no preceding study adopting this integration for predicting project success from the sustainability perspective. The developed system in this study can serve as a tool to assist the decision-makers to dynamically evaluate and predict the success of their own projects based on different sustainability criteria throughout the project life cycle.

Cracks are prevalent signs of pavement distress found on highways globally. The use of artificial intelligence (AI) and deep learning (DL) for crack detection is increasingly considered as an optimal solution. Consequently, this paper introduces a novel, fully connected, optimised convolutional neural network (CNN) model using feature selection algorithms for the purpose of detecting cracks in highway pavements.

To enhance the accuracy of the CNN model for crack detection, the authors employed a fully connected deep learning layers CNN model along with several optimisation techniques. Specifically, three optimisation algorithms, namely adaptive moment estimation (ADAM), stochastic gradient descent with momentum (SGDM), and RMSProp, were utilised to fine-tune the CNN model and enhance its overall performance. Subsequently, the authors implemented eight feature selection algorithms to further improve the accuracy of the optimised CNN model. These feature selection techniques were thoughtfully selected and systematically applied to identify the most relevant features contributing to crack detection in the given dataset. Finally, the authors subjected the proposed model to testing against seven pre-trained models.

The study's results show that the accuracy of the three optimisers (ADAM, SGDM, and RMSProp) with the five deep learning layers model is 97.4%, 98.2%, and 96.09%, respectively. Following this, eight feature selection algorithms were applied to the five deep learning layers to enhance accuracy, with particle swarm optimisation (PSO) achieving the highest F-score at 98.72. The model was then compared with other pre-trained models and exhibited the highest performance.

With an achieved precision of 98.19% and F-score of 98.72% using PSO, the developed model is highly accurate and effective in detecting and evaluating the condition of cracks in pavements. As a result, the model has the potential to significantly reduce the effort required for crack detection and evaluation.

The proposed method for enhancing CNN model accuracy in crack detection stands out for its unique combination of optimisation algorithms (ADAM, SGDM, and RMSProp) with systematic application of multiple feature selection techniques to identify relevant crack detection features and comparing results with existing pre-trained models.