Politecnico di Torino - Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY

+39 011 090 6100 info@tech-share.it

suSTruct.ai – Conceptual design of buildings with artificial intellgence

architectureBuilding design assistantConceptual design methodconstructionMulti-criteriaStructural systems


“suSTruct.ai” is a conceptual design tool based on artificial intelligence (AI) and genetic algorithms (GA) that assists engineers in the decision-making process of building design. The design scope focuses on mid-rise buildings with a rectangular floor plan and equally distributed span distances. The design variables that will need to be determined are structural materials, building and grid dimensions, and types of foundations.

Technical features

The conceptual design phase is the most critical part of the development of functional and efficient buildings. Due to the wide range of possibilities for design projects, it is a long and challenging task to find the most suitable solutions. Consequently, the current approach to the conceptual design of buildings often results in a waste of materials and a high environmental impact. SuSTruct.ai’s approach involves multi-objective optimization that considers three main conceptual design goals: minimizing structural cost, maximizing free space, and minimizing environmental impact. suSTruct.ai was validated using design examples found in the literature. It provides the user in a short calculation time (<2 min. Per execution) with a graphical representation of the best feasible and optimized structural solutions, their cost and equivalent CO2 emissions.

Possible Applications

  • Civil Engineering companies;
  • Medium-rise buildings with a rectangular plan and equally distributed span distances.


  • Reduced cost and environmental impact of new structures;
  • High impact of early decisions on the structural system;
  • Convenient user interface;
  • Customisable software, meeting any company’s specific demands and requirements to give them a competitive advantage in bids and competitions.