By Carlos Gershenson, Researcher at Universidad Nacional Autónoma de México
For the first time in the history of our planet, more than half of the human population lives in cities. This fraction is expected to grow to 70% by 2050. Cities are offering better opportunities and lifestyles, motivating one million people per week to migrate to urban areas. For instance, Copenhagen currently grows by one thousand residents each month! However, people moving into cities also generate several problems, such as traffic, pollution, crime, and other social problems. If we want a sustainable future it is urgent that we must search for better solutions than the ones we have now.
There have been several companies promoting the concept of “smart cities”, where technology can be used to improve urban infrastructure. Still, many of these projects attempt to do what traditional technology does: predicting the future to deal with it before it is too late. Nevertheless, traditional technology is limited in the solution of urban problems. This is because traditional technology tries to optimize systems for an expected future. However, urban problems are changing constantly. For example, it is not possible to predict the position of vehicles in a city with a high confidence for more than two minutes in the future. How can we attempt to optimize traffic flows if we do not know where vehicles will be? I am not suggesting that smart cities are not desirable. I am suggesting that smart cities are not enough to face the increasing complexity of urban problems.
Inspiration from Living Systems
An alternative to traditional technology is offered by living technology. Living technology exhibits the properties of living systems, such as adaptation, robustness, learning, evolution, and self-organization . Living technology does not require predicting the future of a system, since it is constantly adapting to the current situation, just like living systems. Notice that living technology does not imply that technology is alive, but that it is similar to living systems.
For example, a city with “living” traffic lights would respond to the immediate demand at each traffic light, coordinating global traffic with local rules without requiring a central controller, nor even direct communication. I lead the Self-organizing Systems Lab at the Universidad Nacional Autónoma de México, where we explore methods to design and control distributed systems that can find solutions by themselves. This is useful when the problem to be solved is unknown or dynamic. In the case of traffic lights, it is not possible to predict the future position of vehicles. Still, traditional methods attempt to optimize phases for expected traffic flows, even when these are changing constantly. If there is a varying density, vehicles cannot go at the expected speed, leading to inefficient performance. In contrast, self-organizing traffic lights can adapt to the changing demands. Since we deal with so many variables and situations, we require computer simulations to study, test, and understand alternative traffic light controllers. In simulations we have performed , traffic lights are able to self-organize using simple rules to reduce average waiting times of vehicles by half, contributing not only to faster commutes, but also to less fuel spent and thus reduced emissions. With such an adaptive system, drivers never have to wait behind a red light unless there are other vehicles crossing.
Living Public Transport
Many countries try to favor public transport over private vehicles, as it can be more efficient and sustainable. However, growing urban populations increase commuting times and in some cases to the regular collapse of public transportation systems. Increased travel times do not imply only wasted hours, but also affect negatively the health of commuters and in general decrease the quality of life.
In another study, we have shown how public transportation systems (buses, trains, metro, trams, BRT, and even elevators) can self-organize to improve their efficiency beyond the theoretical optimum . Traditional theory tries to minimize passenger-waiting times with equal intervals between vehicles, also known as headways. On the one hand, an equal headway configuration is inherently unstable: small delays feedback into the system causing greater delays until the system collapses. On the other hand, equal headways force vehicles to idle at empty stations or to give partial service to some stations for delayrecovery. If equal headways can be forced, they do give minimum waiting times at stations, but not total travel times. We proposed a self-organizing method inspired by ant colony communication. Ants do not always communicatedirectly, but sometimes leave a perfume (pheromone) that other ants can smell. They can use this communication via the environment (called stigmergy) to coordinate efficiently and adaptively collective tasks. In a similar way, our vehicles leave markings in their environment that other vehicles can detect and use to respond to the local demand at each station, balancing load and demand. Passengers may wait more at stations, but once they board a vehicle, they will reach their destination in less time than with an “optimal” traditional method. Vehicle intervals are irregular, similar to a healthy heart. This is because hearts also have to respond to changes in the demand of different parts of the body (while walking, sitting, jumping). Ill hearts are regular and cannot adapt. Most of our transportation systems are akin to ill hearts, but now we are learning how to heal them.
Logistics, telecommunications, governance, safety, sustainability, culture, and other aspects of cities; they all pose dynamic urban problems where traditional technology is limited due to the inherent unpredictability of the problems. Living technology can adapt by itself to changing demands, offering more viable solutions. We are already witnessing novel applications of living technology in cities, and we will see more. I believe that the only obstacle that might prevent its propagation is our own conformism and apathy that would prevent the exploration of innovative solutions. If we can overcome this, the quality of life will be visibly increased in the coming years.
 Bedau M., Hansen, P. G., Parke, E., and Rasmussen, S. (Eds) (2010). Living Technology: 5 Questions, Automatic Press/VIP 2010.
 Gershenson, C. (2005). Self-Organizing Traffic Lights. Complex Systems 16(1): 29-53. Gershenson, C. & D. A. Rosenblueth (2012). Self-organizing traffic lights at multiple-street intersections, Complexity 17(4):23-39.
 Gershenson, C. (2011). Self-Organization Leads to Supraoptimal Performance in Public Transportation Systems. PLoS ONE 6(6): e21469.
Featured image courtesy of Stuart Miles / FreeDigitalPhotos.net