Automation of the conceptual design process in construction industry using ideas generation techniques

Abstract

Early design stage in Construction projects is a crucial part of the sophisticated longterm design process. Here many fundamental and critical decisions are being taken. The more successful and innovative solutions are developed during this stage, the more advanced, effective, and less costly design is gained. Those solutions can be found by using different tools for ideas generation and the Theory of Inventive Problem Solving (TRIZ) is believed to be one of the most effective and well-structured problem-solving techniques. In the digital century it is reasonable to link modern construction design software with ideas generation techniques in order to enhance and automate design creativity. Nowadays Building Information Modelling (BIM) became a popular stream in the construction design. Existing BIM software have range of instruments enabling designers to bring all their knowledge and experience into projects but, however, such software does not support users in searching for nontrivial conceptual ideas for design.

That is why the ideas generation stage is still a separate, not automated and humandepended part of the construction design. Thus, there is no professional software which would automate the decision-making process at early design phases including search for not only optimal and reliable solutions but also for inventive ones. On the other hand, BIM and graphical programming for design are state-of-the art in the modern construction design and Computer-Aided-Invention (CAI) software is becoming more popular in different disciplines. Merging this with existing inventive techniques could add artificial intelligence into the design software and enhance and automate design creativity in the conceptual design stage.

In the dissertation a method for automation of the conceptual design process of framed bearing structural systems has been developed, implemented and tested. The method consists of three key steps: shape creation, function analysis and contradiction analysis. First, the theoretical grounds for the development of the proposed methodology are justified in the thesis. After that, suggestions are made on the ideology of the method and it is technically implemented using modern design software. Finally, the method is tested on a case study and evaluated using a survey. The results show that the proposed approach reduces design time and increases effectiveness of the process. Further it is proposed to generalise method and increase its functionality.