December 7, 2015 at 11:06 AM
Data analytics is certainly making its impact felt in our collective progress as a species, with the technology being applied across a wide range of human activity. A report by the United Nations entitled Humanitarianism in the Interconnected Age identifies four challenges surrounding data analytics in helping tackle global challenges such as access to education, natural disaster management and disease control and prevention. These requirements centre on the development, acquisition, analysis and sharing of the increasing numbers of new information channels to find solutions to global challenges. In this blog, we'll look at three ways data analytics is helping save lives by allowing us to understand, anticipate and manage events or situations that pose a threat to human life.
1. How call records are minimising loss of life in the wake of disasters
Aftershocks, structural damage or even disease-outbreak pose a continued threat in the aftermath of an earthquake. Minimising the toll on human life means anticipating population movement as it occurs in real-time following such an event. Now, Call Detail Records (CDRs), traditionally used by telcos for customer billing and call analyses, are being used to aid in tracking the movement of individuals or groups affected by disasters. As mobile phones register every billable event such as a phone call, text-message or email with cell towers, they reveal the user's location and movement, enabling authorities and aid agencies to track, locate and rescue people who may otherwise have gone undiscovered. Over time, CDR analysis also reveals how groups of people move following a disaster, helping various agencies understand where and how to focus post-event aid efforts. The 2010 Haiti earthquake is a good example of how CDR data assisted aid agencies in providing time-sensitive aid to victims of the earthquake.
2. How social media aids early detection systems and casualty assessments
Twitter is known for jumping into action when disaster strikes and recently beat the United States Geological Survey (USGS) to the warning-bell when reporting on earthquakes. The USGS, tasked with the monitoring of natural disasters, realised that by augmenting the information they gathered though their network of sensors with that of the social platform, it stood to improve its ability to detect, warn and report on natural disasters. With 2000 sensors placed across the globe, the need to supplement data fed into the USGS' systems was clear – making millions of online Twitter users spread across the globe both an unexpected but abundant source of information. Early warning signs such as tremors are reported by Twitter users and once trends are detected, agencies such as the USGS are able to collate this information with that of their sensors to quickly confirm the validity of reports and kick the warning process into action. Also, Facebook launched its “Safety Check” page earlier this year which allows victims of disasters to report on their whereabouts and condition to all members within their groups, assisting people in communicating their status and getting help to those who need it most.
3. Reducing emergency response times with predictive analytics
The nature of their jobs dictates that emergency personnel work under the constant realities of life and death. In situations of this type, time is of the utmost essence, but ambulances often navigate high traffic volumes over large distances, which can severely undermine their ability reach those stricken in the shortest space of time. In addition, rescue staff are often involved in collisions themselves, compounding the issue even further. Now, emergency call-centres are using predictive analytics to make rescuers' jobs easier by using real-time maps fed by various information sources to help centre agents provide ambulances with the most efficient routes to their destinations. Information sources can include geographic information systems, wireless communications, and global positioning systems along with data sourced from other aid agencies. Applied to predictive modelling, these data sources helped Jersey City in the USA reduce their response times from 8 minutes and 59 seconds to under 6 minutes on average – a significant reduction when considering that human life hangs in the balance.
But addressing the needs of the wider populace will take time. Although these examples show great promise of what data analytics can do to improve our ability to reach, find and help victims of natural disasters, they still fall short in helping everyone. With most of the infrastructure to support such efforts focused in urban, middle class areas, more can be done to ensure that rural populations have access to the technology needed to aid them in times of need. Also, it can't be assumed that data analytics will automatically lead to better decision making. Instead, it requires humans to unravel insurmountable volumes of information in near real-time to understand - and act upon - the warning signs resident in data. But we've taken considerable leaps toward making our planet a safer place, thanks to the power of data.
Image credit: http://thehackernews.com/