Optimization in Shell Structure
Finding the optimal shell solution from structure analysis integrate evolutionary computation.
In order to learn how the result of geometries that generating from the specific questions related to structural behavior by looking through the lowest and the highest fitness from each generation.

Genome and Fitness

Defining the parameters to calculate as Genome for instance create three sets of points which can be move and connect with line for each set then loft all the lines to create the surface for structure analysis.
When we have displacement and mass values from analysis, it can be use as Fitness.
In this case to get the efficient and optimal shell,
we have to get the lowest of displacement and mass values.
Displacement

Generation
In 3 hours galapagos can generate 161 generations for each generations contain 50 population and we can select the best results which are red boxes
and the worst results which are the blue boxes to compare each other in term of structure.

Galapagos calculate the solution randomly from the first generation until it get the best result and try to random the solution agian around the best result.

Deformation
Selecting the highest and lowest results from three generation in order to analysis why different geometries have given different results.
Deformation of the geometry related to displacement value, if high displacement then high deformation , and mass related to area of the geometry.
From diagram below, it show how geometry related to the value for instance the first diagram show that geometry has high degree of curvature meaning it has large area and it’s easy to deform.

Utilisation
At the same time, displacement value related to utilisation and less utilisation value giving the efficient structure.
Galapagos try to balance the values between displacement and mass.

Genetic Workflows: Evolving Solutions is a project of IaaC, Institute for Advanced Architecture of Catalonia, developed at Master in Advanced Architecture in 2015/2016 by: