With respect to our ​previous ​project (moveON) and other own proposals, this new project means a big jump forward to dealing with important challenges of cities, drawn from the ​six axes ​defined in Europe in the ​smart city and communities ​initiative.

The different domains involved and the need for cross-intelligence to address of all them reveal a vast new area for research and innovation. We hope to provide ​many ​useful applications and services at the end of this project (see next subsections), as well as setting the bases for implementing the desired ​holistic ​vision to solve problems in a city.



There is no doubt that Economy is one of the driving forces in the world, for example, the America’s 50 largest metropolitan areas generate around 82% of GDP. One of the most appreciated features of cities in our modern world is that they provide a unique place for ​business and knowledge creation. In the quest for an holistic view of the Smart City, we propose to create an enriched geographical ​map ​, where each building is assigned a color depending on the economic health of the companies living in the building, or even quantitative money values related to their activities. This ​“economy map” could provide a snapshot of the city, useful to define the economic profile of neighborhoods, re-thinking their needs, or analyze their evolution.

Other important use of the proposed map is to find correlations with other high-level aspects of the cities. One of these aspects is mobility (here we enter another holistic integration). One could expect that more mobility be found around wealthy buildings, as a consequence of the good activities of the building. Then, we also plan to study the correlations between traffic and economical development of a city. ​Detecting this kind of relation can be used by the city managers for planning/providing the expected facilities.


The European Commission estimates that road congestion costs 1% of the EU’s GDP per year (€100 billion euro in 2015) while in modern cities, mobility contributes to a 23% of the pollution of the city. Smart Mobility ​optimization ​emerged to reduce the pollution generated by traffic, with great results (up to 50% reductions). A way to reduce the negative impact of traffic consist in optimizing the routes. Every route has associated costs: time, money, pollution. Consequently, more than one ​objective ​must be optimized simultaneously (e.g. time vs. pollution). In addition, data used in the optimization is not precise (it contains errors) and it varies during a trip. One of the lines of this project focuses on this issue: our idea is to provide routes customized to the needs of citizens and modify them according to the current state of the roads and traffic: ​dynamic routes.

Eventually, all cars need to ​park ​, especially in downtown areas. The problem is to optimally allocate the public parking spots in the city to the drivers, while taking into account their preferences. Then, as the time to find a parking spot is minimized (for each driver), the ​overall ​greenhouse gas emissions and the fuel consumption are reduced. We already explored an innovative solution to this problem ​with ideas from biology: epigenetic algorithms. However, sometimes public spaces are not available, then a good alternative is to look for ​off-street parking​. Taking advantage of ​open data ​, usually ​published by local councils, we propose to develop a ​web site plus an app to visualize the current and historical occupancy rate. Furthermore, we propose a set of different ​predictors ​as a tool for predicting future occupancy rates.


Governance is listed in most studies on SC as one of the most important axes for development in the near future. Taking decisions based in data analytics and with the goal of optimizing the urban resources is the only way to excellence. In this axis, urban safety refers to the integration of technology and the city in a cyber-physical system to reduce crime and terror threats and achieve readiness and quick response to arose emergencies. We focus on providing tools to the police and other security forces to improve their activities. The first one is an ​intelligent assistant for the police car​, initially in charge of controlling the emergency lights of the car and other base services. Here, a voice control will be important with some intelligent system to automatize these actions.

Another important topic is ​urban crime prevention. Since information about the events in which the police has participated had been collected during the last years, it can be used to develop different services. An initial one is the building of a ​“security map” displaying the number and characteristic of relevant events in each zone. In addition, the use of machine learning, deep learning, and other intelligent techniques over the data collected by the police departments can provide us of crime patterns. Some uses of these patterns could be the increment the number of agents in the predicted risky zones, some modification of patrol routes to make safer the district. or the detection of unusual scenarios, which can represent changes in the crime behaviour for an early prevention.


Communicating to people in the near area presents a big opportunity to develop apps, by locating unknown users around a source point of information or by using a list of registered citizens waiting to get serviced. This happens in ​transport centers​, but also in office ​buildings ​, and in arbitrary locations of a city. We propose a new model of notifications called ​Smart Alert Notification System or SANS. By managing an ad-hoc Wi-Fi communications infrastructure through discovery of unknown near smart devices (like cellular phones or tablets) and by contacting known users (previously registered), and connected to 4G/5G providers. SANS will be able to send customized messages ​to groups of users in order to alert to important events like evacuation ​alarms​, train/bus/flight departures, publicity, etc. The implementation of our near field alert system could benefit to millions of people who daily make use of similar public services. We plan also to gather information and make data analysis on the previous history of events. We will explore a potential industrial application. We also propose ​WAI (Who Am I), a personal assistant app which tracks the position of a person, this application could be offered by the local city administration to promote cultural events or to warn people in case of an emergency in the area. Additionally, when many people will have the app, they may act as a crowd sensing of the city, to get useful information about a particular area, transforming simple data into information and then knowledge.


The quality of life in cities is determined by many factors, such as the air quality or road traffic. For example, according to the World Health Organization, 3.7 million people prematurely die worldwide per year because of air pollution, and road traffic congestion in EU costs nearly 100 billion Euro (2500€ per driver). Having geographical knowledge ​about such factors will help the institutions to take actions (e.g., planting trees, regulating the road traffic) to improve the life of the citizens. Additionally, this information could help the citizens to take life ​decisions​, ​e.g. where to buy a house according to the pollution or the noise. Therefore, we propose to develop KyS (Know your Streets), a service as a platform that will perceive the status of the streets by collecting, processing, and analysing real time data. KyS system will be based on the deployment of a low cost sensor network. The ​KyS motes will include a set of sensors to measure the ​air quality and pollution (CO​2 ​, NO​2 ​, particle matter, and dust), the ​noise​, the ​temperature​, the ​humidity​, and the ​citizens flows (using Wi-Fi and Bluetooth).

On another note, the volume of produced data is continuously growing, making even more difficult to address optimization problems that impact directly on our living quality, such as ​routing or ​energy plans for home saving​. For example, Cisco projected that by 2019 the volume of data will reach 507.5 zettabytes per year, and the cloud traffic will quadruple. Therefore we will need to find innovative ways of processing data, while reducing the data traffic: ​edge computing. By 2020 it is projected that 5.6 billion IoT devices will use edge computing for data collection and processing. Considering this, we propose AIWY, a ​ubiquitous intelligent system capable of flexibly solving complex problems of smart cities using the processing capacity of heterogeneous devices in a distributed way.


The European authorities, supported by the ​Circular Economy ​theory (turning waste into a valuable resource), established the goal to reuse and recycle 50% of certain waste materials from households by 2020. In addition, waste ​collection services work under uncertainty, i.e. the amount of waste the citizens will generate and when they will throw it away is unknown, but certainly show some patterns. Then, collecting information about actual filling capacity of containers in real time and learn is very important. We will use a fill-level sensor to collect essential information for an intelligent algorithm which will learn and predict future amount of waste. This ​prediction ​will help us to solve the waste collection problem efficiently. Optimizing dynamic routes, real time maps of filling levels, and time series prediction will demand from us accuracy and efficiency.

On the other hand, ​energy ​is a crucial factor for environment, as for economic competitiveness and employment. Nowadays, most part of the energy comes from fossil-fuel based energy systems, generating a negative impact on climate change. Therefore, the EU agreed to reduce global emissions by at least 30% by 2020. According to Eurostat, in the EU 50% of the energy is consumed by industries and households, then understanding the energy ​consumption patterns present a great opportunity to optimize the provisioning, and as a “side effect” ​minimize the greenhouse gas emissions. Thus, by using ​machine learning techniques to characterize household buildings and industries by their energy consumption data, and by using ​metaheuristics​, ​deep learning and additional information (e.g. weather forecast) we propose to ​predict ​future consumptions. Setting a baseline to optimize power distribution, as well as an entry point to build a business intelligence platform, offering a wide range of value added services, ranging from custom rates, to energy savings plans.