a genetic algorithm approach to space layout planning optimization hoda homayouni

Post on 19-Dec-2015

237 Views

Category:

Documents

5 Downloads

Preview:

Click to see full reader

TRANSCRIPT

A GENETIC ALGORITHM APPROACH TO SPACE LAYOUT PLANNING OPTIMIZATION

Hoda Homayouni

Outline

Space Layout Planning Introduction to Genetic Algorithm GASP: Structure

Space Layout Planning

Why computers?

Complexity of the large problems

Shortcomings of human mind in complex problems

Excellent rational and search ability of computers

Challenges

Ill defined problemsQualitative constraintsUsability for architects

Introduction to Genetic Algorithm

Why Genetic Algorithm??

Hill Climbing

local

global

Why Genetic Algorithm?

Multi-climbers

Why Genetic Algorithm?

Genetic algorithm

I am not at the top.My high is better!

I am at the top

Height is ...

I will continue

Why Genetic Algorithm?

Genetic algorithm few microseconds after

Components of a GA

Survival of the fittest

Encoding technique (gene, chromosome)Initialization procedure (creation)Genetic operators (mutation, crossover)Evaluation function (environment)Selection of parents (reproduction)

Genetic Operators

Crossover

Genetic Operators

Mutation

Parent Offspring

GA Phases

reproduction

population evaluation

modification

End

Stop?

Yes

No

Genetic Engineering

Genetic Algorithm Space Planner

GASP: Structure

Hierarchical growth approach

….

Building Layout

Room1 Room 2 Room N

Room Level Operations

Crossover

Room Level Operations

Fitness function

Area fitness Perimeter fitness Concavity fitness Proportion fitness

Creation of rooms in GASP

Click icon to add picture

Building Level Operations

Initialization

Building Level Operations

Crossover

Building Level Operations

Evolving Genes

Building Level Operations

Mutation

Building Level Operations

Fitness function

Area fitness Perimeter fitness Proportion fitness Adjacency fitness

Creation of a building layout in GASP

Click icon to add picture

Results

Results

Results

Results

Future Work

More fitness functions Interactive environment Multi-story layout problems More heuristic Methods

The End

Click icon to add picture

References

[1] http://galeb.etf.bg.ac.yu/~vm/GenAlgo.ppt [2]

http://web.umr.edu/~ercal/387/slides/GATutorial.ppt

Parameter Settings

Parameter Settings

Parameter Settings

Parameter Settings

Parameter Settings

Parameter Settings

top related