Building Orientation is a Machine Learning tool developed to orient a given set of buildings to specific target objects or objects. In this exercise, a portion of land with a stream flowing across its topography is considered and the algorithm is trained to rotate the longer lengths facade towards the stream in order to create maximum views from the buildings, which leads to more number of apartments with the view of the steam.


Three types of building typologies are considered. The goal is to orient and sort buildings in the specific height based on its position on the given topography and its distance from the water body.

Traning of the algorithm is done by considering –

  1. Shortest and widest building closer to the water body.
  2. Medium height and size buildings after that.
  3. Tallest and slender buildings on the maximum distance from the stream.


This is done in order to achieve maximum flats with a view towards the stream. This also helps in cross ventilation in cooling the flats.

Step 1 – Training algorithm with conditions.

Step 2- Population the topography with buildings to sort

Step 3 – Applying the trained algorithm to a new set of buildings


Building Orientation is a project of IAAC, Institute for Advanced Architecture of Catalonia developed in the {Master} 2019/20 by 

Students: Pratik Borse and Faculty: Mateusz Zwierzycki