How big data saves lives

How big data saves lives

SHARE

The article shows how Big Data can be used to save human lives by forming the base of an early warning system, which warns and notifies the drivers about a foggy zone ahead and helps them to avoid crashes and accidents

Big Data has become a buzzword today. The recent explosion of digital data has led organizations, either private or public, to a new era of more efficient decision making. At some point, businesses decided to use that concept in order to learn what makes their clients interested, with phrases like ‘sales funnel’ analysis, ‘actionable insights’, and ‘positive business impact’. So, it stands to reason that Big Data was viewed through green (read: money) colored lenses. Somewhere along the line, however, someone realised that collecting and processing data doesn’t have to be for business purpose only, but also could be used for other purposes to assist law enforcement or to improve policing or in road safety. This article explains how Big Data has been used in the fields of policing and road safety in order to improve the decision making process. It presents a study where Big Data can be used to save human lives via the implementation of an early warning system which warns and notifies the drivers early about a foggy zone ahead and helps them to avoid pileups and accidents. This system collects Big Data from the real situation about the roads, traffic, etc. using advanced technologies such as sensors, smart phones and mobile devices. Then, this data is sent to a cloud-based server using advanced networking technologies in order to be analysed. The system notifies the drivers earlier and helps them to select an alternative safe road to reach their destination. The system has been implemented and applied in the Emirate of Abu Dhabi in the UAE.

Based on historical evidence, driving in heavy fog conditions is one of the most serious causes that lead to massive highway accidents. For example, the Abu Dhabi – Dubai Highway (E10) faced two record accidents in recent times. The first accident was in March 2008 in which more than 200 vehicles were involved in a mass collision. The second was in April 2011 and it involved 130 vehicles. These two massive accidents, and several other relatively minor ones, were due to poor visibility conditions caused by dense fog. Vehicles driving at high speed suddenly enter road sectors covered by dense fog without warning and are then implicated in mass collisions. After the two tragic events, the Government of Abu Dhabi and especially the Abu Dhabi Police Department tried to find a solution for this serious problem. This solution aims to improve road safety in Abu Dhabi when drivers face poor visibility conditions caused by dense fog.

Abu Dhabi City is considered to be a smart city with a well-integrated and sustainable transport infrastructure. Such smart, integrated and high-tech infrastructure generates a huge amount of data via several devices and systems such as sensors, CCTV, mobile devices, remote sensing devices, real-time monitors, etc. Unfortunately, this available Big Data is not fully analyzed for decision making purposes.

In this research, it is intended to take full benefit of this available Big Data to find a solution for the above mentioned problem. Our pro-active solution is a Big Data-based ‘fog early and real-time warning system’ which analyses the collected data and sends early real-time warning signals to all drivers who are about to enter poor visibility sections of the highway, of the dangers ahead, using radio signals or cell phone based short messages. Warning signals can also be displayed to the drivers using Changeable Message Signs (CMS) installed along the highway.

Fog Warning System

The main components of the system which are summarized above are presented below:

Input: Big Data collection component: This component has the main function of detecting the formation of fog and determining the geographical locations of the boundaries of foggy zones. There are several methods or devices for determining the conditions and locations of poor visibility sectors of the highways and these include (fog-related data):

Fog sensors: Installation of fog sensors on light poles, radar stations or cell phone towers along the highway. Data from the sensors is collected by an appropriate communications network and fed to the fog data analysis component. The fixed positions of the sensors will limit the data collection only to the areas where the sensors are deployed.

Fog Warning System

Police cars on patrol: In this project, it is intended to benefit from smartphone technology and to use mobile devices as data collectors in order to detect the boundaries of the foggy zones. In this case, the patrol officer would approach the boundary of the fog-afflicted zone and send a signal to the server while at the boundary point. This is done by simply pushing a single screen icon on a smartphone set which is programmed to send a data through the 3G data network. This data contains the coordinates (x and y) of the boundary of the foggy zone to the processing component of the system (second component). This technique is very efficient because it is dynamic (not restricted by the location as in case of sensors) and can cover any zone in Abu Dhabi area.

Fog Warning System

Real-time incident detection tools/applications: Detection of slow traffic movement due to poor visibility conditions by the cell phone networks which tracks the movement of wireless devices inside vehicles driving along the highway. Such information is then passed on to the fog data collection and analysis system.

More than just fog-related data, other data not related to the fog can also be collected. For example, we can mention a huge and high-resolution spatial data (GIS Data) about the whole region of Abu Dhabi. This data is very useful as it contains information about the road characteristics and land use. Also, in order to make our system more accurate, data about current traffic situation, the number of cars on the roads, the spacing between vehicles, as well as the data about the incidents is collected instantly and directly from the field.

As we can see, the data which is collected is huge, is coming from several sources, is collected in real time by using advanced device and network technologies and has a variety of forms (structured and unstructured, Spatial and no-spatial, real-time or not). The characteristics of collected data comply with the feature of Big Data in terms of volume, variety, and velocity. All this data will be stored in a cloud-based machine or server in Abu Dhabi Police Department data center.

Processing: The data which is collected continuously and in real-time is transferred to a cloud-based computational server in the traffic management center (data and monitoring center). This server contains a component which automatically conducts the appropriate analysis of such data. The purpose of the real-time Big Data Analysis process aims to determine, on the GIS of the Abu Dhabi area, with reasonable degree of accuracy, the geographical boundaries of the poor visibility sectors of the highway caused by fog formation. It is important to mention that the accuracy of the virtual foggy zone is pretty good due to the volume, variety and the velocity of the data. The detected virtual foggy zones visualised in the virtual city of Abu Dhabi (GIS) represent the real foggy zones in Abu Dhabi City.

Fog Warning System

This component, saves the detected virtual foggy zones in its database for further analysis in order to define which area of Abu Dhabi is affected frequently by fog so the decision makers can take further actions and precautions concerning this issue.

Early Notification or warning: Once the boundaries of the fog-affected zone(s) are determined, this component will transmit warning signals to all drivers approaching the poor visibility sectors (real foggy zones) well before they reach such location in order to take appropriate precautions. More than that, if the driver has set his/her trip destination since the beginning, this component can propose an alternate road (on the map) to the driver. This road can be used by the driver to reach his/her destination while avoiding the foggy zones.

The system has several options for driver notification or warning:

Option 1: In this option the cloud-based server sends a notification messages to all the Variable message signs (VMSs) around or inside the foggy zone. The concerned VMSs will be defined by using the GIS data of Abu Dhabi area. This is the simplest and most straightforward option. In practice, this option has two shortcomings: the first is that the drivers may miss reading the sign at the appropriate time, and the second is the inadequate warning distance if the sign is too close or too far from the boundary of the foggy zone.

Option 2: Activating a channel on the vehicle radio that would automatically warn the drivers of the danger ahead. This will require that all vehicles are equipped with dedicated radio weather channels that are remotely activated, which is not true in practice. Furthermore, the warning signal is neither user selective, nor location selective. That is, all drivers will receive the warning signal.

Option 3: The transmission of SMS (data or voice) to the drivers’ smart or cell phone devices with the fog warning signal. In this case, the cell phone device will be pre-programmed to display the signal once received from the notification component. Only drivers approaching the boundaries of the poor visibility zone will receive such warning messages. The key issue here is that in order to receive such warning signal in time, the driver must have the cell or smart phone switched on all the time and it must be possible to program the device so that its GIS location is tracked and it is able to identify and interpret the warning signal. Nowadays, most modern smart phones have these capabilities. As we have mentioned earlier, for subscribed users the system can propose the alternative way to reach their destination, if it has been already mentioned by the drivers.

Fog Warning System

Fog Warning System