Structural Design Recommendations in the Early Design Phase using Machine Learning

TL;DR

This paper introduces ApproxiFramer, a machine learning system that automatically generates structural layouts from building sketches in real-time, aiding early-stage architectural design with structural insights.

Contribution

The paper presents a novel ML-based system for real-time structural layout generation from sketches, enhancing early design exploration and collaboration between architects and engineers.

Findings

Achieved 2.2% average error in column position prediction

Demonstrated system's feasibility for orthogonal metal structures

Enabled real-time structural feedback during early design phases

Abstract

Structural engineering knowledge can be of significant importance to the architectural design team during the early design phase. However, architects and engineers do not typically work together during the conceptual phase; in fact, structural engineers are often called late into the process. As a result, updates in the design are more difficult and time-consuming to complete. At the same time, there is a lost opportunity for better design exploration guided by structural feedback. In general, the earlier in the design process the iteration happens, the greater the benefits in cost efficiency and informed de-sign exploration, which can lead to higher-quality creative results. In order to facilitate an informed exploration in the early design stage, we suggest the automation of fundamental structural engineering tasks and introduce ApproxiFramer, a Machine Learning-based system for the automatic generation of structural layouts from building plan sketches in real-time. The system aims to assist architects by presenting them with feasible structural solutions during the conceptual phase so that they proceed with their design with adequate knowledge of its structural implications. In this paper, we describe the system and evaluate the performance of a proof-of-concept implementation in the domain of orthogonal, metal, rigid structures. We trained a Convolutional Neural Net to iteratively generate structural design solutions for sketch-level building plans using a synthetic dataset and achieved an average error of 2.2% in the predicted positions of the columns.