Parametric Analysis, Generative Design and Machine Learning in Architectural Practice

Dr. Victor Okhoya holds a Doctor of Design in Architecture from Carnegie Mellon University with a specialty in Computational Design, and has worked with architectural computational technology for more than twenty years. He has published widely on architectural computation technology and related subjects, has held leadership roles in digital design technology at leading architectural practices, and has trained hundreds of industry professionals on the use of digital design tools.

Marcelo Bernal, Ph.D., holds a doctoral degree in Architecture from the Georgia Institute of Technology with a specialty in Computational Design, and has more than two decades' experience in research and architectural practice. He has held professorial positions at the Universidad Técnica Federico Santa María (UTFSM), Chile, has published widely on architectural design and related subjects, and is the Past President of the Iberoamerican Society of Digital Graphics, SIGraDi.

Preface xi

Acknowledgments xv

Part One: Back ground and Definitions 1

Chapter 1

COMPUTER-AIDED DESIGN AND BUILDING INFORMATION MODELING 3

Computer Aided Design 5

Building Information Modeling 9

Chapter 2

COMPUTATIONAL DESIGN AND PARAMETRIC DESIGN 19

Computational Design 19

Parametric Design 30

Chapter 3

WHAT IS PARAMETRIC DESIGN? 41

Definition of Parametric Design 41

The Need for Parametric Design 47

Project Case: Parametric Analysis 51

Project Case: Generative Design 60

Part Two: Parametric Analysis 73

Chapter 4

METHODS OF PARAMETRIC ANALYSIS 75

Performing Energy Simulations 75

Performing Daylight Simulations 82

Daylight Analysis Process 83

Performing Parametric Analysis 90

Evaluating Parametric Analysis 97

Chapter 5

PARAMETRIC ANALYSIS FOR ENERGY AND DAYLIGHT PERFORMANCE 105

Energy Analysis Project Case: Hotel Residence 105

Project Case: Student Housing 112

Project Case: Student Residence 120

Project Case: High School Building 132

Part Three: Generative Design 139

Chapter 6

METHODS OF GENERATIVE DESIGN 141

Space Partitioning Algorithms 146

Site Subdivision Algorithms 155

Form Making Algorithms 160

Chapter 7

GENERATIVE DESIGN AT THE URBAN DESIGN SCALE 171

Project Case: Business Park 172

Project Case: Mixed Housing 179

Chapter 8

GENERATIVE DESIGN AT THE BUILDING SCALE AND INTERIOR DESIGN SCALE 195

Generative Design at the Building Scale 195

Generative Design at the Interior Design Scale 206

Part Four: Machine L earning 219

Chapter 9

MACHINE LEARNING FOR ENERGY ANALYSIS 221

Machine Learning Algorithms 223

Project Case: Residential Tower 230

Chapter 10

DEEP LEARNING FOR SOLAR ANALYSIS 239

Types of Deep Learning Algorithm 241

Generative Adversarial Networks 246

Project Case: Urban Solar 252

Bibliography 257

Index 259