In renovation operations, whether they take place on the scale of an apartment, a building or a district, the energy and environmental dimension plays an increasing role. This raises technical questions that are sometimes difficult for decision-makers to grasp. Can the use of artificial intelligence be a game-changer?
Artificial intelligence (AI) is increasingly used in these operations, which reminds us that renovation is a space for innovation. Four areas can be distinguished.
The first is to classify in order to prioritize. AI is embedded in coarse-mesh diagnostic tools. For example, it will make it possible to identify, in a group of 500,000 public buildings and on the basis of their energy bills compared to their dimensions, the 1,000 or 10,000 for which one euro invested will have the greatest impact in terms of reductions in CO2 emissions or energy consumption.
The second area is the creation of optimized renovation plans, upstream of a construction site: they make it possible to improve the layout of an apartment, the renovation of a building or the renewal of an urban area.
Third area, at runtime: the census. The techniques of computer visionwhich are based on images from various sources (scans, drones), make it possible to identify everything that is on a site, but also to determine the percentage of progress or even to identify risks.
The fourth area is everything that comes into play when the building exists: predictive maintenance and intelligent management. Based on various sensors, AI can anticipate breakdowns or better understand the preferences of building users. Some systems of smart building go further by allowing interactions between the users (or inhabitants) of the building with conversational agents, or even by making predictions on the needs of the residents and thus generating personalized service proposals.
In terms of renovation, it is today the second area, design, which is seeing the most promising advances.
With the rise of eco-design, on the one hand, and fabrics with increasingly stringent standards, on the other, this design work has become extremely complex in the last twenty years. What role does AI play in this complexity?
At the building level, technical standards are mastered by professionals and the contribution of AI is not yet major. It will nevertheless make it possible to carry out simulations of energy performance, environmental impact or climate resilience according to various objectives and constraints, which can contribute to optimizing renovation solutions from an economic and environmental point of view. In a changing environment, where the prices of energy and materials evolve a lot like at the moment, the use of simulation makes it possible to reopen the field of possibilities and to find paths outside the habits of professionals. Automation also makes it possible, more prosaically, to speed up the work.
At the neighborhood scale (more rarely the entire city), where things get more complicated, AI plays a more important role. Both for questions of resources: the industrial professions involved are practiced by powerful economic players, often at the forefront in terms of technology; and for questions of needs: it is mainly at this scale that new variables appear, which were not integrated before and have no obvious “technical solutions”, transposed into practices.
These questions are in particular those of the environmental impact of renovation activities and the resilience and adaptation of infrastructures to climate change. They are added to those, more classic, which have governed decision-making up to now: cost of works, comfort, energy consumption in the economic sense (purchasing power of households, costs for managers), consumption of energy in the ecological sense (emissions). A renovation plan today is a very complex matter!
How does AI fit into human processes?
Basically, we are in a logic of decision support. Let’s take an example. In France and in Europe, many renovation operations concern neighborhoods or buildings belonging to the “social” stock, built for the most part in the thirty years following the end of the last World War. These are therefore collective decisions, involving public actors and, of course, public money. The processes are cumbersome, the decision-makers numerous, the stakes high, because the physical degradation of these neighborhoods goes hand in hand with social and sometimes political problems. Moreover, they are most often thermal sieves, and the poor quality of the building particularly exposes the inhabitants to energy shocks (increase in price, discomfort). Finally, the design of neighborhoods and buildings makes them vulnerable to the consequences of climate change, particularly heat waves and floods. There are car parks in these neighborhoods which are heat islands, and those on the banks of rivers may be subject to a risk of flooding which was negligible when they were built.
To sum up, renovation is here a subject that is both urgent and complex, which brings up new issues that are not easy to master for public decision-makers. One of the key issues is therefore to accelerate and optimize collective decision-making. This is precisely what AI allows.
Renovation is both an urgent and complex subject, which brings up new issues that are not easy to master for public decision-makers.
How ? By allowing, at early stages of the discussion, to automatically generate optimized representations, plans for example, which are not “the” solution but which allow the discussion to move forward. These representations are, what is called in management science, “boundary objects”: they are sufficiently concrete to constitute an interface between social worlds and actors with different perspectives.
On a decision-making process that lasts a dozen months, if we take a pessimistic scenario, the AI could thus make it possible to win one or two. It is significant, without being a radical upheaval. We can also bet that the decision will be of better quality: it is informed in a way that allows everyone to understand the reasoning of others, and on the basis of representations that are easier to discuss than columns of figures. AI, by making it easy to generate proposals, is a facilitator and optimizer of human decision-making.
Can AI “participate” in the discussion?
Not in the sense of a computer speaking at the table…but a discussion dimension can be built into design work that involves AI. The field that interests us within the framework of the RenovAIte R&D program on which I work, is what is called Adversorial Resilience Learning (ARL), a concept that makes it possible to model and train artificial neural networks by putting them in competitive situations. We take two virtual entities, which will literally throw themselves against each other. One will play the role of defender; it will be a modeling of the urban ensemble from BIM models (Creation of information representation systems : digital models that do not represent geometric shapes, but objects) and other known elements. The other will play the role of the attacker: these can be all crises (cold, heat, flooding, with risks to comfort, to heating systems, risks of cracks or impregnation, budgetary constraints, etc.). After several hours, this competition will allow each entity to develop the best strategies and come up with a set of proposals for optimized urban renewal plans for decision-makers.
It already works: this concept has been used in Germany in the design of intelligent energy distribution networks (smart grid), to identify malfunctions on the network and generate the appropriate automatic responses.
Interview by Richard Robert