Recently, a "Faculty Research Day" was held in the Faculty building of Civil Engineering at the Technion - Israel Institute of Technology. During it, an award-winning competition was held in which each graduate student presented his/her research through a poster. Among the winners of the competition, AIPDORCS Research won the Audience Award.
AIPDORCS Research Abstract:
Despite significant developments in the field of computing in construction, schematic engineering design – such as creating structural layouts based on architectural models – lags technologically. Schematic design is still manual and based on rules of thumb and experience. This stage is the most important and creative stage in the design process of buildings and greatly affects the final engineering solution. Even after rigorous analysis, a non-optimal initial design will often lead to a suboptimal structure.
In the construction industry, there are still relatively few applications that use artificial intelligence (AI) in the initial design phase. An advantage of AI is the ability to learn procedures from data describing cases. For example, using machine learning, it should be possible to receive as input many structural models, designed by engineers, and to learn how to generate and to evaluate new building designs. Thus, it should be possible to simulate the experience of engineers accumulated over years of practice. In this study, we aim to illustrate this idea by devising an artificially intelligent application capable of performing initial design of structures, in a fashion similar to, and perhaps even better than that of human engineers.
To read the original LinkedIn post of the National Building Research Institute click the button below.
Very nice!