Cover Page

Architecture and Computer Science Set

coordinated by
Thierry Ciblac

Volume 1

Eco-generative Design for Early Stages of Architecture

Xavier Marsault



This book deals with “architectural eco-design”, a subject that is topical and fascinating, but difficult, because “in the drafting phase”, the designer has a wide range of options in front of him, but also a schedule, a certain number of constraints and rules that must be respected – and a barrage of uncertainties. This is why we have chosen to take “the generative approach” to help forge dynamic paths in these creative and promising spaces. In doing so, we address neither eco-construction nor eco-innovation, although they are closely connected to eco-design.

Architecture focuses just as much on the approach (the design process) as on the subject (analysis, study, construction, monitoring over time). Its subject of study is the structure, covering form, material, use and appropriate sustainability, all of which must come together to produce a building that is eco-efficient and pleasant to be in. It is discussed as the result of a project process based on the subtle balance between often contradictory decision-making criteria.

Elected representatives, managers, town planners, contracting authorities, architects and users navigate a constantly expanding universe of knowledge, often without the resources they need to understand its complexity and guide or explain their rationales, which are often a source of conflict. The fact that a large amount of knowledge is scattered and inaccessible to the people who have been affecting the environment through their choices for decades shows how urgent it is to produce tools that enable them to assess the impact of their decisions in a reasonable time frame.

A number of studies have questioned the conceptual, creative and innovative research phases of architecture and construction engineering. Most of these aim to define methods and develop tools that help with design and are likely to assist in the creation and production of more intelligently designed buildings. Digital instrumentation and support now have a vital role in this, and are the subject of regular studies and computing developments.

To meet the challenges of more environmentally responsible architectural production, for around 15 years, research has been enabling architects to access knowledge and tools from energy, environmental and constructive engineering, notably by means of digital simulation, a real interface between the engineer and the architect. Of course, we must acknowledge that specific tools for assisting with the design of efficient structures in the upstream project phase are only just starting to find applications outside laboratories. However, research strives to go further, proposing design support software environments that are better suited to the usual working methods of architects, attempting to preserve their autonomy and creativity.

The concept of efficiency, which is central to eco-design, runs through all the chapters. However, it is always “tricky to define efficiency in architecture, because it takes into account not only the objective and measurable qualities of an object, but also its relationship with its built or social environment, and the use to which it is put by users” [LAG 13]. Hensel proposes a redefinition of the concept of efficiency in architectural design, based on an analogy with biology [HEN 10]. We will return to this in the final chapter. This specific approach is different from previous ones, which either focused on questions of representation and meaning, or considered efficiency a synonym of function. According to current developments, efficiency is merely a level of requirement that must be reached retrospectively: energy efficiency, for example. But it could also be argued that the efficiencies that should be prioritized are those that have the greatest impact on the form and materiality of the structure.

In what follows, the four designations, namely criterion, objective, efficiency and fitness, denote one reality, seen, according to the case, from a qualitative or quantitative point of view.

What aspects should be prioritized in the upstream design phases? Which choices may be decisive, and what impact do they have on other aspects (formal, technical) that may influence the overall outcome of a structure?

We are going to examine the issues, possibilities and methods of eco-design, based largely on research and developments conducted within the French ANR EcCoGen project, which produced the EcoGen software program. Tackling the major difficulties of generative design head on in the interactive first stages of an architectural project (where the choices are the most decisive in terms of the overall and future efficiencies of the structure), EcoGen is interested in the behavior of structures in their constructed environment, through a generative and multi-criteria approach (morphological, energetic, atmospheric, functional, constructive) of eco-efficient design.

This book can therefore be considered partly as a summary of this project, which involved researchers from different laboratories, mainly the French CNRS’s UMR MAP 3495 (Models and simulations for Architecture, town planning and Heritage). Several texts from the final project report1 are cited to support and illustrate our arguments.

Finally, we will end with a discussion of the ambitious prospects combining some advances in the understanding of natural evolution with the desire to produce a truly bio-inspired theory of architectural morphogenesis. On this topic, the accounts provided in Chapter 3 should be linked to some of the bio-inspired prospects of Chapter 6.

I would particularly like to thank the following people for their contributions and thoughts: Philippe Marin, Renato Saleri, Hervé Lequay, Lazaros Mavromatidis, Florent Torres, Lara Schmitt, Nicolas Grégori, Jean-Claude Bignon, Gilles Halin, Estelle Cruz, Violette Abergel, Ronan Lagadec, Anaelle Quillet, Aymeric Broyet and Florian Mignot. I am also grateful to my “more distant” researcher colleagues: Grégoire Carpentier, Tibériu Catalina, Brian Mc Ginley and Thomas Jusselme, with whom I have had some valuable discussions.

This morning, arriving at his office, Paul knows that a new stage of his life as an architect is about to begin. Yesterday, he received the new tactile creative tool from Microsoft, “Surface studio”, equipped with a digital pen and control knob. He is one of the few in his profession to have this kind of equipment, although it is very affordable. His reason for turning to this modern solution, which encourages fluid design with tools that imitate freehand drawing, is called Minos. With this brand new software program, based on the integration of better technologies for designers, he knows that his ideas and creativity will transfer to the digital world like never before.

Minos registers Paul’s tiniest line, slightest curve or smallest volume sketch in real time. Little by little, it comes to understand the ideas behind his project by comparing them with local and cloud-based databases, and then delivers the results of multiple calculations by means of a voice assistant, along with graphics that support decision-making. It can therefore quickly give Paul advice, direct him to eco-efficient choices, suggest that he alters lighting, structural elements or openings, etc. However, in addition to supporting the project with this active intelligence, from the sketching phase onwards, Minos is constantly learning from the user and familiarizing itself with his ways of drawing and designing. After a certain amount of learning, it will be able to suggest innovative shapes to Paul, without jeopardizing his creativity. As Minos is subtle and knows how to be discreet, AI is a winner! It can even share this knowledge with other online users, if Paul authorizes it. It’s a real marvel!

Of course, just like the mythological hero after whom it is named, this software does not exist, or not yet. But are we very far from this kind of architectural design aid?