PROMΕTHEUS | And YOU can Help Fight Fires!


Awards & Nominations

PROMΕTHEUS has received the following awards and nominations. Way to go!

Global Nominee

The Challenge | And YOU can Help Fight Fires!

Build a fire-monitoring and crowdsourcing tool that will allow local fire managers to respond to wildfires.


This project concerns fire prevention and management with the assistance of GIS, Remote Sensing and a variety of data. The main goal is to give citizens the opportunity to help rescuers by informing the authorities and taking care of their safety.




TEAM MEMBERS: Dimou Athanasios, Lazios Thomas, Soubry Irini, Tzioumaklis Dimitrios


Fire was from ancient times an energy source and a way of protection from the forces of nature. It is known from Greek mythology that Prometheus fought with Zeus, during the Titanomachy, and after he saw the hardship of men, he stole the fire and gave it to them. He taught them how to manage and benefit from it.

From the moment though, where fire becomes uncontrollable, it causes devastating destructions which affect human lives, economy and nature. Fire is a dynamic phenomenon and a function of many factors. The specific characteristics of each region give different intensities, therefore this phenomenon cannot be generalized and treated in a uniform way. The inability to cope with it, is caused by the lack of knowledge of the complete morphology of the environment where it is evolving. This consequently leads to the lack of strategic response systems.

Our team has developed a tool designed to show people rational management of fire and protection from it. As rural engineers we decided to use cutting edge technology in order to develop a system that will assess the location of prominent fire, where it will spread when it breaks out, how the response units will move strategically and how the citizens can be protected. It also syncs all the available resources as to treat the fire timely and cost-effective, it enables the citizens to contribute to this effort and at the same time to be protected. With this tool we give the possibility to use heterogeneous temporal data, to handle ashes and smoke spread, to estimate the movement of future floods as well as landslide risks – as soon as the damage of the fire is assessed.

Worth mentioning, prior to the explanation of the project, is the strategic framework of crisis management, where treatment algorithms are built upon the concept of time-distance and proximity. The notion of the unimaginable dynamics of fire can be degenerated in a cohesive way and structured in a spatial-strategic way, as the individual advantages of the involved parties are viewed as multiple properties in order to optimize the solution in the shortest possible time. In the aspects of biomimicry, this approach includes a function where a light beam chooses its path based on less time, even if the distance to travel is greater (minimum time principle). Similarity can be found within the ant species Wasmannia auropunctata, which choose the optimal route for food search based on the speed at which they will reach their destination (Ant Colony Routing).




Most part of the data can be derived from open source data, such as NASA data, OpenStreetMap files, and from world-wide websites with similar available open data (see Data Sources). This will significantly reduce the cost of installing the system and will speed up the building process.

Picture 1: GIS DATA ( )

1) The data that will be used for the GIS system will be vector as well as raster:

  • NASA FireSat images*
  • NASA Real Time satellite images
  • Fire danger forecast maps (available after remote sensing processing)
  • Use of Landsat 8, Sentinel Europe, NOA, ASTER and MODIS images

* A constellation of more than 200 thermal infrared imaging sensors on satellites designed to quickly locate wildfires around the globe, fully operational system in space by June of 2018.

Satellite images of high resolution are the most convenient for our application.

2) Lidar Data, with appropriate processing for the calculation of forestry density.

3) Topographic (surveying) Data that will be of vector type:

  • Digital Terrain Model (DTM) rasterized and created with GIS from vector data from surveying methods, contour lines with 5 m distance, TIN, land slopes and directions
  • Hydrographic network (polyline) which includes data from the main rivers, lakes, seas, lagoons etc.
  • Road network (polyline), it consists of 5 different type of data, highways, provincial roads, municipal roads, rural roads and trails.
  • Geological data (polygon) contains data of the subsoil (material and type of soil, ground water sources)
  • Land uses (polygon), classification of areas according to the type of land use (forest, urban areas, fire protection areas etc.)
  • Points of interest are the locations of hospitals, firefighting vehicles, water extraction points, airports, etc.

4) Climate Data is of raster type and includes maps produced by satellite imagery and remote sensing processing in conjunction with GIS:

  • Temperature maps (min-max-average)
  • Wind maps
  • Groundwater maps
  • Solar maps

5) Meteorological Station Data:

  • Pressure
  • Temperature
  • Humidity
  • Wind speed

6) Descriptive Data includes two categories:

  • A chart (database) with information about previous fires with the location (coordinates) as well as the date and extend of the destruction
  • Location names (toponyms) of the region for orientation purposes

Descriptive information beyond the last two categories will also exist in the rest of the data, as the advantage of geographic information systems is the combination of spatial (geographical) and descriptive information in order to have the best decision-making system in a rational and strategic way.


The control center is the heart of the system and at the same time the brain of the whole undertaking. Like in the human body, the brain is the only organ that works having in mind the future, in the same means, the control center is building information and directing movements to prevent and manage the crisis and post-crisis. These operations are done in a GIS environment, in the way it is structured and explained above, while it is constantly supplied with new information.

Firstly, the control center continuously receives new satellite imagery data (ex. FireSat), and with appropriate remote sensing processing, it creates thematic maps of drought and vegetation. Combined with land topography data as well as climatic data, it creates hazard maps. The fact that this data is temporal, yields to optimal assessment results. Thus, the control center can properly manage its fleet by using geospatial criteria in order to optimize their performance when the assessment if verified. It is also possible for the algorithm to yield possible terrestrial interventions that would help manage the crisis, for example, fleet guard buildings, water transport, fire protection zones, etc.

Secondly, the control center has to manage and filter the information that is related to fire outbreaks. This information contains two categories:

a) Updating with the use of satellite imagery (in order to be able to detect fire the area has to be 10-15 m wide in the best case). Taking into account the delay of transferring and processing information, the fire range will have grown significantly, so no further control, but alert movement is required.

b) The second category is crowd-sourced data which includes telephone calls and geo-located images.

As far as crowd-sourcing is concerned, there are two type of users, simple and authorized. Therefore, filtering and verification of information is required, especially for the simple users. For the authorized users, a simple confirmation question through the application is enough, whilst for simple users the verification process includes a framework of smoke sensors, live cameras and UAV systems. Furthermore, a threshold of fire incidents at a specific region is chosen in order to activate immediate mobilization of rescue teams. This is mandatory in order to avoid false alerts. The use of geo-location allows for faster check of information entering the fire extinguishing control center. If the location is far from terrestrial access, the use of UAV systems (drones) is considered necessary. They can give the exact location of the fire and its size in real time. Also, when there is an outburst of data over a minimum period of time (ex. 150 entries/min), another algorithm is used to facilitate the fire control network.

It is worth noting that the use of the application’s geolocation is very important because the data must be grouped to understand the location and the number of events. Also the production of a single coordinate pair to be used by the application as an indication (information simplification) is very crucial. Ideal tools for the achievement of this could be the Delauny Triangulation, Voronoi charts or Voronoi dynamic charts as well as a combination of these or a selective process depending on the number and density of the points’ distribution. A satisfactory grouping condition can be calculated based on time that has elapsed since the fire started, using an incremental method (Nearest Neighbor Interpolation).

With telephone calls, only descriptive information is provided and therefore the exact location of the fire at an early stage is difficult. The database with place names (toponyms) can give an initial satisfactory approach of the position. To tackle this problem, it is recommended to use terrestrial tools, such as smoke sensors and live cameras.


The structure is divided into two sections, the module for simple users and authorized ones. This separation is done for a smoother flow of information.

Picture 2: Simple Users App ( )

Simple users are given the opportunity to be informed of real-time events, as well as forecasts of fire and location estimations of it. In addition they can inform the control center about a fire that has begun near them. At the same time, they have access to information about points of interest like hospitals, transport points, evacuation route information, etc. Furthermore, if they are encountering a direct incident, like being surrounded by fire, they can alert directly the control center with the press of a button. This data follows another processing and control path for immediate assistance. Finally, the cartographic background is given along with all the available data, such as risk assessment, fire outbreaks, fire movement, traffic speed, points of interest and of course optimal escape paths. These are updated frequently by the control center which has an oversight of fleet traffic and crisis management. In this way users can navigate quick and easy to a safer destination.

Picture 3: Authorized Users App ( )

Authorized users are given more options through the application. Also the level of detail of the information is bigger in order to serve better their purpose. Thus, they are also able to know at any given time the details of each fire and more accurate estimation of its movement, the position of the whole fleet, water abstraction points and their load. They can also access classified government information such as underground military facilities, minefields and so on, in order to protect these areas and not approach them. The application is in direct communication with the control center, which it can update and be updated from.


Picture 4: Processes ( )

Once the position of the fire has been confirmed, the system must immediately calculate and estimate the traffic. This is possible by combining all the available data. Fire movement is a function of soil morphology, wind speed and direction, obstacles encountered and, of course, the type and amount of fuel.

In order to be able to calculate this, it is necessary to create an equation that will be structured with weights, depending on the category of data. Depiction of humidity and drought maps can give an initial estimation of traffic, provided there is a large deviation from the average. Altitude, gradients and direction play a key role both in the development of local wind patterns that are created, but also in the type and amount of fuel that exists. Land uses, forest registration models and, of course, Lidar data can yield the full amount and type of fuel. In addition, information on fire protection zones and other structures (ex. road network) can be obtained from land uses, which can be a deterrent to fire transmission. The hydrography of the area contributes to the estimation of the movement and the underground hydrography to the temperature of the soil.

With the above data it is possible to model the estimation of fire movement, as well as the level of combustion and intensity. Thus, the size and extent of this can be calculated by time models. Also, the system is able to calculate the fleet size required for its extinguishment. Its strategic approach is feasible using multi-criterial spatial resolution methods, and it is therefore possible to calculate the timescales required for each group, optimal fire approach paths, optimal instrument selection, and even possible time needed to deal with the phenomenon.

As far as the best path calculation is concerned, again, this can be done using multi-criterial spatial resolution methods. In particular, using network analysis (linear elements, nodes, etc.) as well as elements of script theory where there are two know examples of optimal path finding through algorithms. The first is the route inspection problem or the Chinese postman problem, and the second is the travelling salesman problem which is recommended for selection of the path with minimum cost. In order to identify these, several algorithms have been created, such as the Bellman-Ford algorithm and others, but the Dijkstra algorithm is the most known and simplest of them all, and solves the problem of single source and shortest path.

These calculations are categorized in GIS into 3 categories:

  • Single pair optimal route single source and shortest path)
  • All pairs of optimal paths between pair A and B to point C

Appropriate processing of wind models, combined with topographic data and estimation of fuel size, can depict the movement of smoke and ashes. In addition, the terrain model combined with the hydrographic network can yield water flow, the amount of which can be calculated with meteorological data. Based on the amount of water and the intensity of the flow, combined with geological data, it is possible to estimate the areas where landslides may be experienced. Full mapping of the catastrophe is feasible using satellite images combined with land topography data.



Open source software in order to eliminate the purchase costs, but also to enable the configuration through programming.


Control center’s computing systems, cameras, control sensors and drones that are necessary for the supervision of the control department.

Human resources:

Specialized employees of relevant bodies for the use of the system’s software and the additional programming of it.


Possible funding for this application can be from the government, non-governmental organizations and also groups of organized civilians. The system can save resources directly through:

a) Licensing

b) Selling

c) Training new users

d) Extending the app to the management of other natural disasters

e) Selling data to third-party users (for example live cam data, sensor data)

In addition, through the vehicle fleet management strategy, the responsible bodies save money by the reduction of fuel and resources.


The benefits for a society, a state or the whole planet cannot be valued with economic terms. This application provides environmental protection (of nature and structures) and protects both civilians and firefighters. Even if one human life is saved through this system, humanity will have gained.


Tackling a problem is interwoven with the observer’s intelligence. Now we know that we don’t just look at the problem, but we see its solution. When we think we watch the flow of time and the future has already revealed itself. The circumstances are here and only the power of decision remains in order to implement a solution for all mankind. The problem of fire is not social, it never was, it is human. And human elements where hiding in the structures from the beginning of time. Only by following Prometheus’s steps and believing in people, in their hope that lays in their eyes and the power of their soul, only then will we be able to truly change this world!



1) NASA FireSat project:





6) Copernicus Open Access Hub:

7) Copernicus Global Wildfire Information System:

8) GeoServer:v

9) USGS Open Data and Tools:*%3A*

10) Earth Explorer USGS:

11) Chinese postman problem:

12) The Traveling Salesman Problem:

13) Bellman-Ford Algorithm:

14) Dijkstra Algorithm:

15) Prometheus-Greek Mythology:

16) Biomimicry: Ant specie Wasmannia auropunctata:

17) Voronoi charts:

18) Voronoi diagram and Delaunay triangulation:


19) Li L., Penga H., Kurths J., Yanga Y., Schellnhuber H. J., 2014. Chaos–order transition in foraging behavior of ants, PNAS, Vol 111, no. 23, pp. 1-6, doi: 10.1073/pnas.1407083111

20) Hyde P., Dubayah R., Walker W., Blair J. B.,Hofton M., Hunsaker C., 2006. Mapping forest structure for wildlife habitat analysis using multi-sensor (LiDAR, SAR/InSAR, ETM+, Quickbird) synergy, Remote Sensing of Environment, Elsevier, Vol. 102, pp.63-73, doi:10.1016/j.rse.2006.01.021

21) Rampl G., 2014. Crowdsourcing and GIS-based Methods in a Field Name survey in Tyrol (Austria), United Nations Group of Experts on Geographical Names, University of Innsbruck, Institute of Language and Literature, department of Linguistics Vol. 16, no. 9, pp. 1-7

22) Merino L., Caballero F., Martínez-de-Dios J. R., Maza I., Ollero A., 2011, An Unmanned Aircraft System for Automatic Forest Fire Monitoring and Measurement, Journal of Intelligent & Robotic Systems, Vol. 65, no. 1, pp.533-548, doi: 10.1007/s10846-011-9560-x

23) Feick R. D., Hall G. B., 2004, A method for examining the spatial dimension of multi-criteria weight sensitivity, Int. J. Geographical Information Science, Vol. 18, no. 8, pp. 815-840, doi: 10.1080/13658810412331280185

24) Dimou Ath., 2013, Creation of a model for estimating the energy footprint of moving vehicles with the exploitation of geoinformatics: the case of the municipality of Nikaia - Ag.I.Rentis Attica, Department of Geography, Harokopio University


SpaceApps is a NASA incubator innovation program.