Sustainability Videos & Lecture Series

Michel Morvan: Thank you for your introduction, and also for the invitation. I have to add one word—I mean two words. The first one is I’m not any more at Veolia; I left last week. You can see that I’m not at Veolia because I don’t wear a tie and a suit, so I’m becoming a normal person again, which is fine. The other thing is that I’m very happy not to answer the question to know if God is green or not, because I have absolutely no idea about this. My speech will mainly be without PowerPoint because I’ve spent four years doing corporate PowerPoint presentations. By the way, I realize that I don’t really like that; I prefer to speak to people. At the end of the presentation I will show you one or two very short videos to show you the kind of things we have been doing in the last couple of years.

As Sander mentioned, I will discuss this question of modeling approaches for sustainable urban development for the three perspectives he describes. The first one is the academic one, with academic questions in mind. Of course, the question of sustainable urban development is very important for academia. The second perspective I will keep in mind during all my talk will be the big corporation perspective on this question. Of course, I will use and share with you the Veolia perspective on it. Third, as also Sander mentioned, I created a couple of years ago a start-up company that is developing now—I am not any more inside of the company, but, of course, I am very interested in what they are doing. They are specialized in developing tools for modeling complex systems in very different domains. I’m going, also, to elaborate a little bit about this point.

These three perspectives, for me, are very important for the subject we are going to talk about. The sustainable urban development—sustainable development of cities, of territories—is something that we can address. Not only one perspective can help to solve and understand this problem. If we want to solve and understand this problem--how can we help cities to develop, and to develop in a sustainable way so that the people can live in a good, and healthy and nice environment, and so that the economic growth can go in these cities? You cannot just hope that you are going to solve that just from the academic point of view, or just from the big company point of view, or just from the political point of view. This is one of the characteristics of what Sander mentioned, and this is the things we have about this global system science.

This global system science is the science that deals with these very big and global questions—global systems in which you have a lot of complexity, but in which you have a lot of very heterogeneous stakeholders playing together in the context of open system. It’s not every system, I’m sure, that you can work with—it’s something very open at very different stages. What I think is that these different perspectives that I’m going to keep during all my presentation are very important to understand the kind of things we have to do in the future if we want, in a way or another, to help these cities to become more livable and well-livable places.

Another important thing about this global system science in which I can say that this approach is part of is that—at least in this case, but I think in many others—we want to, in a way or another, to work with decision makers. It can be politicians, it can be—I mean, many kinds of decision makers, but one of the big things is to say, “How can we—scientists, academics—help the decision makers to make the good decisions?” Sometimes we feel—probably you felt that before, that you understand the problem better than the decision maker understands the problem. Sometimes it’s true; I mean, sometimes, it may be wrong, but sometimes it’s true. The question is how can you, from where you are, go to the decision maker, and make a change here and convince the decision maker that this should be better in this way, rather than this way, because of this and that?

This is not very easy because these people—very often, they don’t have a lot of time, and you have to convince them, and you have a huge background in your domain, and you just have three minutes, or 10 minutes, or 20 minutes—something like that. This is something that is important for this global system science perspective; in particular, in this question of cities. This is something that I personally experience while working with Veolia. We launched a program to help cities develop globally, to help cities to think of their future, and their sustainable future on the long-term perspective. I was in contact with the mayors of different cities, and the big question was, “How can we convince them to listen to what we are explaining them?” This is not so easy; I will elaborate on this a little bit later.[Pause]

In the title of my talk there is—I’m talking about sustainable urban development; so, what do I mean by this? What is the perimeter of the things I’m trying to do and trying to present? What I mean by urban, in this talk, is a city or a set of cities—a metropolitan area, something like that. Sometimes people think about systems of cities, which can be the same, but can also be a little bit different in the way that these can be considered in an upper scale, in a way, with different cities in the world interacting together. I’m not going to talk about this situation that is very interesting in itself. I will mainly consider the question of: Okay, there is a metropolitan area, there is a territory—not too big—and someone in the territory has specific questions, and problems, and wants to make decisions. These decisions will have impact on the long-term, and the person wants to know what will happen, based on the decision that has been taken.

I use the word “sustainable”—just a little story. Sander mentioned the fact that I was an Eisenhower Fellow, and at that time I had the chance to meet a lot of people in the U.S. during a period of, more or less, two months. I went to the Senate and I met the senior advisor of Senator Boxer. Senator Boxer, at that time—I don’t know if she is still a senator—but at that time she was senator of California. I was presenting to the senior advisor, I was thinking about different things; I was thinking about cities, and sustainable cities, and all that stuff. Then she told me, “Okay, what you say is great, I’m sure Senator Boxer will be very interested by what you’re explaining about sustainability, but please, in this country, don’t use the word ‘sustainable.’” I said, “Why?” and she said, “Oh, if you use the word ‘sustainable,’ people are going to think that you’re a communist.” Well, I am not a communist; I could be a communist, but that’s what she told me.

I learned from that, and every time I speak about sustainability and sustainable—of course, not here, but when I’m talking to other people—I really mention that of course, with sustainability I mean environmental aspects—waste, water, energy, transportation. When I was at Veolia—and also health questions, all this kind of stuff; how can we do things so that it’s sustainable, so that it’s not going to kill everything, and everything is not going to die because of what we have done? Of course, it concerns economy—you should take decisions you want your economy to be sustainable; you don’t want your economy to collapse rapidly. Just to mention that sustainability—for me, of course, and in this situation—is related to environmental questions, but not only—and concretely.

When were were discussing, and when we were working with Veolia with My Earth, environment was not their major concern; their major concern was economy. Their major concern is, “How can I attract people to live in my city? How can I have the people in my city live well? How can I increase employment?”, and all these kinds of questions. This is very important, because I’ve heard a lot of people working on sustainable cities, and what they have in mind is only the environmental part, which is very—I mean, I have absolutely no problem with that.

Another point I’m going to mention is that nothing can be done in a city without the citizens; every decision you’re going to take, even if you take it at a very high level in the city, you will have to imply it—at a moment or another—you are going to have to imply, and to have additions from the citizens. Unfortunately, for many reasons we can all understand, it’s not possible to ask for a handful of citizens to increase their energy bill to save the planet. Maybe one percent of the citizens can do that, and have the means to do that, but it’s really not possible. If you want to do something, and this thing to be efficient, then you have to have that accepted by the people who are the first person concerned, who are the citizens. This is something that we always have to have in mind.

The main driver, for a lot of people, is what they pay at the end of the month, and so we have to take that into account. For example, if you say, “Oh, I want to develop alternative energy. I want to have a centralized energy production in my city, because this is great.” If at the end of the month, of course, the price of the bill at the end of the month for electricity increased, people are not going to use it—to accept it. If you have to increase, for example, for the kilowatt hour, at the same time you have to give to the people the possibility and the ways to reduce their consumption; otherwise, it just won’t work.

This is very basic, but this is something that all the My Earth—the people in charge—have in mind. You cannot tell them—I explained at the beginning that you have to convince the people, you have to convince the decision makers to act in a way or another, and if you want to convince them you obviously—and this is very obvious what I say—but you have to obviously take into account what they really want. In this case, this is mainly driven by economy rather than by environment. Of course, tomorrow, if there is a cap and trade thing about greenhouse gas, and things like that, things may change; but this is what I really, concretely realized when I was working with the city.

If you look at the reason why, and what we do when we use these modeling techniques—for the moment, they just talk about city’s decisions, and, of course, sustainability. The subject of the talk is to talk about modeling. The idea that we have and we think—and when I say “we” it’s a lot of different kinds of people—is that using modeling tools can be very, very useful to reach the target I just described. Unfortunately, modeling a city is not something that is very easy, so I’m going to say some words about that. Before doing that, let’s think about, why do we want to use models of cities, or of territories, or geographies? Well, the first thing can be—if you’re in a pure academic approach—can be, “Well, we want to understand what’s happening; we want to understand the way the city is developing. We want to understand the way the cities are related one to the other,” and these kinds of things. This is the mainly academic approach.

We may want to use these kinds of models to see what the future will be, or to try to imagine what the future will be. For example, a lot of people think in a couple of decades there will be 80 percent people living in cities, compared to 50 percent today. This will change a lot of things in the world, and maybe, using models, we can try to see what will be the long-term consequences on the planet, and on the world, of this kind of change. This can be another reason to use models.

A third thing that justifies the use of models and has been really, really developed and advertised by big corporations is the concept of smart cities. Very often when you have this concept of smart cities, what you have behind is to use very sophisticated tools, including models and modeling of the cities, to be able to have decision makers to take decisions on kind of a real-time situation. Let’s say there is a problem in my city today—what can I do, what will be the impact today, tomorrow, in the next couple of days, of what I’m going to do? There is a pipe that explodes in the street, then it changes everything for the traffic—what can I do, what would be the consequences of that? This is, for example, what is done by IBM; the IBM Smart City program is mainly dedicated to working on these kinds situations.

Another use for these kinds of short-term modeling tools that I’m mentioning here is also to give—and this is done in some places—to use these modeling tools, and these representation tools of the city to give citizens tools to decide what they are going to do; how they are going to behave. This is also something very important—keep that in mind—but this is mainly on a short time scale. I mean, if I do that, this will impact me in this way in the next weeks, or the next months, or maybe until the end of the year; but this is not a very, very long-term.

The last—I mean, this is not the last, but—the fourth point I’m going to mention—the one I’m going to describe today—is the long-term perspective. What you want to do is to be able to model, represent the city, in a way or another, so that you can use your modeling tool to make simulations that are going to give you the big trends coming from the decisions you have taken. Okay, I decide to increase the length of my metro line—what would be the consequences of that in 10 years, 20 years? I decide to put in my city a district cooling system, or district heating system—what is it going to change in the life of citizens? What is it going to change on the long-term in my energy consumption?—this kind of stuff. Of course, this is not—as I presented here—this is not very hard to do on the first approximation. There are modeling tools that can help you to do this kind of stuff. What I’m going to describe is something a little bit more complex than these very simple examples.

In this fourth situation, which is the long-term modeling perspective, what you want to do is not only to say, “Okay, I have”—for example—“something to model my transport network. I increased the size of my transport network, what are the direct impacts?” What is very important, and very hard, is to be able to understand the indirect impact. The tools that are mainly used, when they are used, are really only focused on direct impact, while the big question—I’m going to give examples about that—is about indirect impact.

First very simple example: if you say, “Okay, there is too much greenhouse gas. I am in a country”—for example, France, where the electricity is produced without greenhouse gas emission. We have nuclear power plants; I don’t know if it’s good or bad, but no greenhouse gas emissions. I’m going to create a new metro line. Then it will have impact because people are going to take the metro rather than taking their car, and then it will have an interesting impact. Then you can try to measure that, and take your decisions based on that; of course, this will not be the only parameter that you’re going to look at, but assume that this is a parameter. If you look precisely, what is going to happen when you increase your metro line? You’re going to completely reshape the city by doing that. You’re going to increase the price of the land around the metro, and you’re going to have new companies working there because it’s easier for the people to go and work, and you’re going to reshape everything. At the end of the day, the impact on CO2 or on greenhouse gas can be all different magnitudes above the impact that you just imagined directly.

These are the kind of things that you have to be able to understand and to manage. These are the big points the decision makers face today. To be honest, I met a lot of them, and when you start the discourse about these kinds of things, they immediately say, “Oh, yes, it would be fantastic if it was possible.” Then I say, “How do you decide now?” They say, “I don’t know, if I think if it’s good I do it. I pay Makenzie, and then I do it.” When you start to know them a little bit more, they start to say, “What is important to me is to be able to explain afterwards that, at that time, I couldn’t take a better decision.” They would be very happy to have tools to be able to do that, and that’s the kind of things that we have worked on.

Just a last point about these kinds of use of modeling, and just to precisely think—I mentioned the IBM-like approach on short-term, using, of course, big data; using modeling tools, using a lot of things. The other one, the long-term modeling approaches for scenarios. To clarify what I mean when I said, “This is close, but not the same”—it’s just to say in one situation, it’s like when you try to do a weather forecast—you want to know the weather in the next ten days. On the other side, you’re trying to understand the climate—what it’s going to be in 10 years, 20 years, 30 years. In both cases, you need data, you need models—you need a lot of things, and there are a lot of similarities, but this is not exactly the same. You have these two big lines in the people working on this kind of situation.

Just to let you know how I arrived to work on this program, it was just when I went to Veolia—when I’d been hired, in a way, by the dark side of the force—going from academia to big industry. It happened that a couple of weeks before I got to the company, the CEO of the company met with the mayor of the federal district of Mexico City. Of course, he had a lunch, and he was presenting to the mayor all of the great and fantastic things that Veolia was doing for waste, for water, and for transportation; explaining that Veolia was the first in the world, and all this kind of stuff. At the end of the lunch, the mayor said, “Oh, this is great”—which doesn’t mean anything, of course—“but my big problem is that I absolutely cannot know how my city is going to develop in the next 20 years. This is a big question for people living here.” By the way, it was a big question for him because he wanted to be a candidate for the president’s election, and all this kind of stuff; he wanted to show the great things he had done. He was not elected in the end, I don’t think.

He said, “At Veolia, since you have water, waste, energy and transportation, which are four big systems in my city, you should be able to help me to predict what will happen if I take a decision. Because, of course, everything is going to impact everything, and how will my city be in 20 years if I decide to do that, or if I decide to do that?” In the case of Mexico City this is a big deal—for example, the decision to restructure the infrastructure of city of Mexico is huge. There is a huge problem of waste; there is not enough room in landfills. There is a huge problem of water. For example: 60 percent of the water that is sent to the network is lost by leaks, so this is the total consumption of the city of Madrid that is lost in Mexico every year.

This is very interesting; every decision that you take—for example, the decisions about the waste—will have a huge impact on health; because if you don’t have any more waste everywhere in the streets, then you know that the children are going to have less sickness, and so it will have an impact on the economy. He said to the CEO of Veolia, “Oh, certainly you should help me for this problem.” Then, of course, the CEO, what did he say? He said, “Yes, of course we can.” Then he came back to Paris, and went to me, “Oh, I said yes, we can—what can we do?” That was my first discovery in a big corporation. That was interesting. Of course, what did I say, I said, “Oh, yes—I am going to invent something,” which proved that I can learn quite fast.

Just before entering in Veolia, I used to work in modeling biological systems. I am going to ask you to stop thinking about city for five minutes, and to try to think about this: this is my thumb, and if you look at your thumb, you’re going to remark something interesting: that, more or less, it looks like mine. In a way, if you really think, it is quite surprising. What is also very surprising is that this thumb will, more or less, look the same in ten years. However, in ten years, no one cell that is here today will be here in ten years. Probably no one atom that is here today will be here in ten years; however, the shape, I’d say, will be the same in ten years, and it will be the same as your thumb.

If you want to try to understand where this shape comes from—if you start to discuss with biologists—you realize that part of them are going to say, “Oh, no problem—it comes from genes. You have genes in your cells, and these genes act, and at the end of the day, you have the shape. Because you keep these genes, and they do their job—and then, there is the shape.” Of course, this is not enough, and they would say, “These genes, they interact one with the other, and sometimes one is active, and it inhabits the other one, and you have a network of genes. Of course, this is the result of the actions and dynamics of the network of genes.”

Then, another would say, “Oh, yes, this is true, but these genes, what they do—they act on morphogens, which are biological substances that have the opportunity to move from one cell to another, to change the behavior of the cell—the way it grows, the shape; the physical properties, the mechanical properties of the cell. If you want to understand the shape, you have to understand the genes, the network of genes, the morphogens, the impact of genes on morphogens, the impact of morphogens on the network of genes, and you have to understand—of course—what happens at the bond area of the cell, and—of course—the cell-cell interactions, and—of course—you have to understand the properties of the tissue in the organ,” and so on, and so forth.

If you want to understand the shape of your thumb, or of mine, or whatever, you have to be able—you cannot try to understand that, and to do that, just by looking at one level. If you just look at the cell level, it will not be enough; if you just look at the gene level, it will not be enough. You have to think all of these different systems, integrated one into the other, and you have to take all of them—more or less—together to try to understand what happened. It happens that the kind of things I was working on before going to Veolia, I was working not on the thumb but on the plant, and trying to understand the morphogenesis—the way the shape arrives—and using modeling tools for doing that.

What I said to the CEO after—not immediately—maybe we can try to realize that if we look through the window, and we look at our city, we are going to realize that this is not so different from your thumb. If you look at the city, you have a lot of different systems that interact together; systems at very high scale, systems at very low scale. You have individuals interacting with individuals, you have social networks, you have energy systems, you have transportation systems. All these systems, they work together, and maybe if we want to answer these kind of questions about the city, and the future of cities, then maybe the good idea would to be to use the modeling approach, and to use the tools that had been developed for biology, and to try apply these tools to the cities. That’s what we try to do, to be just convinced that the same kind of things happen in your thumb that also happen in your city—which is what we call emerging phenomenon—that probably a lot of you who have who have heard of or who work in complex systems know very well.

Here, the shape is an emerging phenomenon; it is something that emerges from a lot of different interactions and systems. In cities you have, very often, the same kind of thing. I have an example that is, for me, quite interesting. I’m not going to say that we have modeled it with our system, because it happened one century ago. In Paris, one century ago, there is a phenomenon of social segregation that arises in the city, which means that the poor people left the city of Paris to go to the suburbs. It was in a relatively short time period—a couple of years. Historians and sociologists worked a lot to try to understand why it was so fast, and what at that time. Of course, you can imagine the impact on the urban geography has been huge in the next century, because people not living in the same place, not working in the same place; it has an impact on transportation networks, on traffic jams, on energy needs—many, many, many social impacts.

Of course, if you try to understand what happened, you realize that this is the consequence of a new technology; this is the consequence of a new technology that arrives in the city at that time. This new technology is called elevator. If you have a building in the city with no elevators, you have the rich people living in the first floor, second floor, and the poor people living in the sixth floor. Because every day, if you have to go to take the bread—if you’re in Paris and you forget something, you have to go down six floors, and go up six floors. That’s how the city was; when you put elevators in the city you have deep change because, of course, the apartments at the top were more expensive, at least, certainly, more expensive than what they were before. Of course, the poor people cannot, anymore, pay, and so then they have to leave.

This is the first emerging phenomenon; the new technology implies a consequence that was not thought before. What is interesting it to understand why the elevators arrived at that time; elevators arrived at that time in the city because the electric network arrived at that time in the city. Because when you use an elevator, what you want is to be sure that you have a good energy distribution system, so that you’re not going to stay in between two floors. If you realize that it means that the development of the electric network, which is good, has had—as consequence—the fact that you have a social segregation, which we can talk about, but is probably not very good. This kind of situation, it’s very important to be able to anticipate in a way or another.

If you look at that, you realize that this is not very different than what happens in biology, if you have thought and worked to model biological systems. This is the same kind of thing, and the idea was to say, “Okay, let’s use the tools that I have developed in biology to do this kind of modeling.” Of course, this started four years ago, so I’m going to show you, in a couple of minutes, one or two simulations and kinds of things we can do with this. I just want to say, also, some words about complex systems and how all these problems I just discussed—in particular, the cities—how they belong to this complex systems approach that is now, of course, more and more developed. What do I mean when I say complex system—what is a complex system for me, and hopefully for a couple of other people? It’s a system in which you have entities, whatever they can be, that interact with local rules, so that in the end of the day you have global behavior that cannot be directly understood from the behavior of each entity.

I gave you the example of a set of cells, where the entities can be cells, and genes, and morphogens—all this kind of stuff. Entities can be cars, can be individuals, can be buildings; can be networks in this inner city. You put all that together, and you have the dynamics on that, and these dynamics lead you to emerging phenomenon that are not very easy to understand from the basic entities. If you look at that, you realize that, very often, these systems that we call complex systems, they are composed by many different subsystems which are coupled, one with the other. Very often, these subsystems are heterogeneous. For example: consider a city, or consider a thumb—you have the cells, you have the genes; you have the set of cells, which is the tissue, and all that of a different nature. In the city you have individuals, you have networks, many different networks—water network, energy network, energy decentralized production sites, you have a lot of very different things—and all of that is interconnected; implicated one into the other.

You have very heterogeneous systems, and if you go deeper, and look deeper, you realize that these systems are not just, for example, a different space scale; an individual and a transportation network are not playing at the same scale, but also a different time scale. You have it originating in the way they are, and it originates in the characteristic time—the way they change—so you have this huge mess. This is this huge mess that makes the city, being the city that you know, or the thumb being the thumb that you know. From all that is coming the structure, and the emerging phenomenon. These terms, of course, are very—the first impression, you can say, “Oh, I don’t want to deal with this stuff,” and then you realize that all the living systems are like that, and that almost all the systems that you know are like that. It can go from very natural systems, like the living systems, to very artificial systems, like a car.

In a car, for example, you have 14 different levels of hierarchy, implicated one into the other. Of course, big planes, and all this kind of stuff—so artificial systems, natural systems—and you have systems that are, in my perspective, very, very interesting, which are the hybrid systems. Like the city—if you look at a city, you realize that it is completely hybrid. You have things that you can design in a city, and this is also a living system, because you have people; you have ecological systems living in the city. Here, you have a hybrid system, but if you think, these kinds of systems, they are everywhere. They are interesting, also, for another reason: that they have a lot of very good—at least, very interesting—properties. Adaptability is one property that you can find in many of these systems; not all of them, but many of them. You have the robin’s nest, and the fact that you have—very often at the same time—robin’s nest and fragility; the same system at the same time is very often very robust and very fragile.

For example, if you consider a human body, and if you consider this as a system composed of a billion cells and bacteria, and a lot of different things. If you shake me in different systems, things are going to be more or less wrong, at least for me—for you, I don’t know, but for me, things are going to be okay, but wrong. Okay? You can change me in a way or another; I can eat a lot of different things, drink a lot of different things. Let’s say, globally, the behavior will be okay, but if I just eat one milligram of poison, the behavior of the system is going to change dramatically very rapidly. What we say is that it goes to another asymptote, which is a nice way to say that I’m going to die.

These systems also have very interesting properties in themselves; for example, the most interesting part of the life or evolution of the system is not at the limit. If you know mathematicians, very often when they study a dynamical system—a system with dynamic rules—what they try to understand is how is it going to behave at the limit? If I do that, how will it be at the limit—how will it be at the very long-term period? If you look at me, what I’m going to be at the limit is probably not the most interesting thing you can observe about me. Probably the most interesting thing for me will happen between age of zero and age of something in the near future. After the limit, the system is not very interesting. It means that if you’re a theoretician, it means that if you want to study these kind of systems, rather than do what you have done—which is to try to look at the asymptote, to try to look at the very long-term future, at infinity—you have to look at a very specific period at the beginning of the system life.

This is something that is interesting, and if you are a mathematician, it means that rather than looking at living cycles, or strange attractors, or whatever, you have to look at transience. This is—in transience—that the very important behavior of the system happens. This is very important for two reasons: first, because—as I just said—the global system itself is going to have these properties, but there is another reason for that. When you are going to study the subsystems—remember I said these subsystems are coupled, one to the other. When you have systems that are coupled one to the other, what happens is that one is going to perturbate the other all the time, and they are going to perturbate each other all the time. It means that very often, no system has the time to go to its limit; so, this is another reason for studying the transience.

In my opinion, that is very important, because if you look at the work that has been done in dynamical system studies—I’m not saying that nothing has been done on transience, but the main things that have been done on transience, for example, is to look at how long do they last until you arrive to the limit. For example, I just mentioned that on the theoretical perspective—I know that you are not all mathematicians, far away from that, so I’m not going to develop more—but I think there is something very important here that can have important implications. Another very interesting question is about unpredictability. There are beautiful theorems about the fact that you cannot predict what happens for these kinds of systems, which does not mean that you cannot work with them. This is another very, very interesting.

If you want to deal with these systems, the problem is that if you look at them from a scientific—let’s say hard science perspective—you realize that it’s very hard to have good results; you cannot solve the systems, in a way. You have the model—sorry, you cannot solve the models—you create a model for your system, and it’s not possible to have a beautiful equation saying, “Oh, this is how the model is going to be in 10 years, or in 20 years.” What you have to do is to use modeling tools and to use simulation—of course, being very careful with simulations—but trying to see and to test different scenarios to understand what’s going to happen to your system; when it’s going to evolve. Then you can try different situations; “What if I do that? What if I do that?”

This is the kind of thing that we have done. Why can we do that now—just a little, very short digression—why now, and not 20 years ago, or 30 years ago? I think there are three important points. The first one is that if we want to do that—if we want to be able to describe these complex systems—this can be the case for living systems, for social systems, for cities; also, hybrid systems. What you need is to have a lot of data to create your model. What happens is that now we have access to a lot of data. This was the case, of course, with biology, when there were the genome projects—to obtain a lot of data—but this is also the case for social systems. Now, you have the possibility to know what people think because they talk together all the time on the Internet, and then you can keep track of that. Of course, this is the case for cities, so we have a huge amount of data.

We have a huge computational power, which means that we can make simulations that were just impossible to do 20 years ago. We start to have new modeling tools that can help us to do this kind of stuff; that’s why it’s possible to do that now. There is a risk with that, and I go back to something that is very important, that is: how can I convince the decision maker? Don’t forget, in this tool, this is my goal; at the end of the day, I want to convince the decision maker. The risk with these kinds of tools—and we have really, really to be very careful—is that the decision maker thinks that this is the truth, that what the model is going to describe is the truth; is what is going to happen. Of course, this is not the case, it’s just we have to use modeling tools as tools. It’s not the truth, it’s not a way to know the future; it’s just a way to know, kind of, trends. “Well, if I do that, it can have this impact. It doesn’t mean that it’s going to have this impact, but be careful—this can be one of the impacts. If I do that, rather, then it can have another impact.”

This is the way we have to think about, and I think it’s very, very important, particularly in the global science perspective, not to sell to the decision maker the fact that we’re magicians, and that we’re going to solve all their problems with computers. I mean, I’m saying that, I’m laughing about that—this is something that can happen, so I think it’s important to have that in mind.

Last thing before I show some short examples of what we have: one of the big problems when you want to do modeling for these complex systems is exactly what I said before about the systems—that you want to model systems, but these systems are heterogeneous. They’re a different time scale, different space scale; there are hierarchies, and it’s a big mess. That’s why I mentioned that we created, a couple of years ago, this start-up company who was a spin-off of the research lab from the tools that have been developed from the biology. The idea was to say, “What is a complex system, in general, and how can I provide tools for modelers so that they can model very, very heterogeneous systems.” They can couple them; it can be modular. You can try to test different kinds of modeling quite easily, and so that you can have all this kind of stuff in a toolbox, so that you can describe your model the way you think it is, rather than describe your model the way the modeling tools want you to describe it.

The idea is not to say, “I’m going to have a partial differential equation tool, and you are going to model your city with my partial differential equations,” or, “You’re going to work with agent-based modeling, and I have a tool to develop agent-based modeling.” Then you have to describe the city with agent-based modeling; maybe your city is part agent-based modeling, part P.D.E.—partial differential equation—part something else, and you have to put all that together. That’s what we have done with the start-up. These tools have been used for biology, but also for the cities’ modeling. I’m going to use this laptop, now. I used to be a computer scientist, but a theoretical computer scientist, which is—

Sander:[Laughter] That’s a wonderful introduction!

Michel Morvan: That was a long time ago. Just a very, very simple example—I have a problem, to be honest, which is I’m not allowed to show you the real thing that has been done for the cities, and the real things that have been done for the clients—for Veolia—for I.P. reasons, and a lot of things. What I show you here is a simple example that has been made by the company that’s called The CoSMo—the start-up—The CoSMo Company, for one client who accepted. It’s not the best. The situation—this is the city of Grenoble, in the Alps.

Female Voice: Shall we make the whole screen?

Michel Morvan: I can do whatever you want. Okay, like that? Okay, so this is the city of Grenoble, and the mayor of Grenoble asked the company the following thing. He said:

“Okay, here is the city today. We know that the city is going to—there’s going to be an increase in population size; new people are going to come to the city. There is a place in the city that I want to develop that is here, and I have two ways that I want to use to develop this place, and have more people living here. The first one is to add new transportation system in this place, and the second one is to take some decisions to develop companies in this region. I want to know what would be the impact—if I do that—in the future.”

I’m just going to show you this very simple example. Once you do that, the first thing you do is that you decide exactly the territory you’re going to work with, then you divide the territory in different zones. That’s what the mayor of the city gives you; you have the different zones. For each zone, you have different characteristics, then—I have to be able to stop this, okay—you can see that you have different networks; different networks appear here. There was the tramway network, there was the bus network; these are the streets. You have all these networks interacting together, so you’re going to model each network, and you’re going to model the interactions of the networks. You’re going to model the characteristics of each little district here; you’re going to model the kind of population that lives here—where they live, the money they have, the way they work, and all this kind of stuff.

What I just want to show you here is a very simple thing. This is the simulation that is going to run, with all these different systems. Here, what is shown is two things: first, the congestion, or what happens on the streets. When it’s red it means that there are a lot of people. When it’s yellow it means there are not a lot of people, and the density of the zone—the darker it is, the denser it is. This is the evolution if the mayor doesn’t do anything; so business as usual, I know that my population is going to grow, I think it’s on a ten-year period. You look at what is going to happen. This is the simulation, of course—there are a lot of possible ways to see the data. To be honest, you cannot see anything on this, and this is made for.

The second thing, what you do—this is the zone that you are interested in, so you create a new transportation network here, and you also add new incentive for companies to go and develop here. You do the same, and you relay that to all the transportation networks, and you play the same game. Of course, when you look at that, you just see that it is going to evolve during the years. What is interesting is not the image, but it’s what is behind all the networks interconnected. Here I’m going to show the comparison between business as usual and what the mayor has done. If you look here, this is the difference in population size for each district that appears here. Here, when it’s red, it means that there is less population here after the adding of the new transportation network. Green means that you have more population here. What is interesting is just the following: that if you look in that, you realize that at the end of the day, you have something that you are not surprised, is that you have created more attractiveness here. This is fine; so you have more people, more density here. You have, of course, less density in other parts of the city, but what is interesting is here—you haven’t changed anything here.

The result of what you have done here is that here, you are going to increase density. It happened that this is not what they want because these are poor parts of the city; that the mayor wanted to reduce the number of people here. This is exactly the kind—on a very simple example—the kind of things that you can do when you use this kind of modeling. I’m just going, now, to show you another one. It’s funny, because as I told you, I’m not allowed to show you the real simulation. What I’ve asked my friends to do at The CoSMo Company was to send me a video of the public exposition. This video I can show you. Here is just to show the kind of tools, this is—previously, it was just a movie; here, it’s a person using the tool itself, showing what you can do with the tool.

It’s the city of Lyon; in the city of Lyon, here, you can see the different systems interconnected. I think this is the blue lines—I mean, there are many, many different things interconnected here. You have a great interface. What is interesting, for me, on this example, is just to show you the different kinds of interfaces that you can have. Why I want to show you—not because I want to advertise the company, first because the interface was not done by The CoSMo Company, but by another big player in 3-D representation. What I want to show you is two things; is that this is the kind of tools that we use to show to the mayor. This is the kind of thing that we provide to the mayor to help the mayor to take decisions.

I want to tell you two things about that: the first one is that using this tool we have had a very interesting experience with the mayor of Lyon. The experience is two-fold. The first point is that we have shown—the mayor of Lyon wanted a lot of things, like all mayors, but one of the things he wanted was to be able to know what will happen if he increases the size of the B metro line in Lyon, and create a new station—all modeling; many, many things in the model, many different systems—a lot of things—and at the end of the day a beautiful simulation that shows what will happen. We showed that to the mayor, and in a moment, the mayor says, “Stop, stop! I want to look at some place,” and he showed us some place, here. He said, “What does it mean that there are all these red lines here?”

It was close to the new metro station, and we said, “Oh, it just represents the traffic jams—the congestion,” and he said, “But everybody told me that creating this new station is going to reduce traffic jams.” We said, “Oh, unfortunately, in the long-term it’s not what happens,” and so we tried to see what was happening. What was happening was the following, it’s very simple: once you have the new station, the price of the square meter increases, so the price of the apartments increases around the station. The population is changing, and the new population has two cars rather than one, so you increase that—and they do not work at the same place.

This is just an example; what is interesting is what the mayor said. The mayor said, “This is the first time that I can see the impact of one of my decisions.” This is something important. I said that the experience is two-fold—the second point is just here. It’s just—to end, cause I think I just have three minutes—is this. You cannot see a lot, it’s just a photo. It’s a photo of the same event; you have here the mayor; here, this is the minister of I-don’t-know-what, who was visiting the booth. What happened was very interesting. This was my collaborator, who was showing what was happening, and explaining to the minister. At that moment, the mayor stopped him and said, “No, no—I am going to explain,” and then he started to explain to the minister what great things he was going to do with his city, and he showed the minister the problem with the traffic jams and the new station.

This is exactly the kind of thing we want to be able to arrive to. This is not only the thing we want to have, but we have to find ways so that the decision makers appropriate the modeling tool. This modeling tool is very, very complex; if you think of all the things that are in the model, all the things that you connect into the model, and you have, at the end of the day, to provide something. To be honest, it was by luck that it happened with the mayor of Lyon, but you have to think and to really—in a way—find a theory of, what do you have to show to the mayor so that the mayor can appropriate what you have done? This is a real question—I just showed that for fun, but this is a real question.

Just to conclude, these approaches have been developed to the city of Singapore, to the city of Mexico, to the city of Lyon; which is great, but very little. Only three cities, so everything has to be made, everything has to be invented. The second point I mentioned at the beginning, I just want to finish on that. This is a subject that can have—if we are able, and when I say “we” it’s a lot of players—if we are able to really provide new tools, really being able to help decision makers, we can, in a way, change the world, or participate to change the world. This is very complex; modeling can help to do that, but the key point is to have all the players playing together. There is no hope that only big corporations, or only academics, or only start-ups, or only cities, or only whoever, can solve this. This is something that has to be solved by a big number of different kinds of players. For me, this is the big challenge. Thank you very much.

[Applause]

[End of Audio]