Patricia Raquel Navarro Tichell’s research while affiliated with Polytechnic University of Valencia and other places

Abstract video: https://alperkanyilmaz.com/decision-making-using-artificial-intelligence-in-civil-engineering/ The conceptual design decisions have the largest influence on a building project’s safety, value, and environmental impact; hence they are commonly assigned to a “senior engineer” to make use of his/her experience. However, the senior engineers can be biased towards solutions inside their area of expertise, which often prevents them from finding the best solutions among alternatives that must consider complex inter-related, and multi-disciplinary parameters. The engineering community could benefit from a rapid and high-quality decision-making method or tool to increase the speed and quality of its high-impact design choices. There are valuable studies in the literature exploiting Artificial Intelligence (AI) to improve the structural design process; however, most of them focus on the final design stage (e.g., Building Information Modeling), and the rest requires an existing project database (e.g., architectural drawings, already decided material types) to propose a small number of initial design alternatives. In this article, we present the development and validation of a genetic algorithm tool based on Non-dominated Sorted Genetic Algorithm II (NSGA-II) that can be used to analyse a wide range of safe, economical and low-CO2 options for the conceptual design of buildings. The design space starts from a design brief (with only the information about the site characteristics and project objectives). The solutions are explored with the material, grid size, floor type, lateral resistance, and foundation system variables. In a short computational time (< 2 min per run), users are provided with a Pareto graph of a large set of feasible solutions (in terms of cost, embodied CO2 emissions and free space) that an engineer would not be typically able to evaluate within a traditional conceptual design process. For future applications, the methodology presented in this paper is flexible to include more engineering materials (e.g., timber, masonry, structural glass), complex architectural forms and merge other disciplines in decision making (e.g., building physics construction management, fire safety).