Creating smarter fire services using data science

AI of Things    19 September, 2017
Here at LUCA, we are passionate about Big Data for Social Good, or BD4SG. In our previous post, published during the first ‘Big Data for Social Good in Action’ week, one could see the power of data science in analyzing the Zika outbreak of 2016, fighting deforestation and aiding hurricane relief. This article focuses on the world of firefighting and explores some of the problems firefighters face, data-driven solutions to these issues, and the wider implications of these solutions.

The Challenges

Firefighters work in a highly pressurized environment where time is short and lives are often at stake. During the journey from the station to the site of the fire, information arrives from many sources including tablets, Sat-Navs, manuals and radios. Bart van Leeuwen, a Dutch firefighter and founder of Netage, suggests that this information overload is one of the main problems that fire services face. He often uses the example of the Anne Frank house in Amsterdam to highlight this point. The average journey time is 59 seconds, and in this time, the team must read five pages of reports on the building.

Even if we assume that a human being can absorb this quantity of information in such a short timeframe, no amount of preplanning can make you 100% prepared. The Black Swan Theory refers to Western explorers who, having discovered black swans in Western Australia, decided that they were not in fact swans, because they understood all swans to be white. The subsequent theory that Nassim Nicholas Taleb developed describes a surprising event that is rationalized in hindsight. ‘Black Swans’ are commonplace for firefighters, as well as doctors and police officers.

The Solutions

How can we overcome these issues in order to create smarter emergency services? Van Leeuwen argues that harnessing data is the key. Data is being created and collected at an exponential rate but an app that simply contains this information is no longer sufficient. Rather, access to the data is what is needed, so that data scientists with a passion for Social Good can work with the data to discover insights that the emergency services themselves may not see. ‘Open Data’ would facilitate this, and van Leeuwen argues that it must be a two-way transfer. As such, the Amsterdam fire department publishes their data in an open format at the same time as requesting access to data. One example is the ‘Firebrary’, a library of technical terms developed so that everyone can be on the same (web)page.

Amsterdam, the city using data to create smarter fire services
Figure 1 : Amsterdam; the city using data to create smarter fire services.

‘Open Standards’ for data would bring consistency and make working with Open Data even easier. If the world were to adopt Open Standards, ‘Linked Data’ would become an even more powerful tool. A great example of Linked Data is Wikipedia, where one link on a page leads to another page, containing more information and more links. Put simply, Linked Data unlocks information. In Amsterdam, the fire department posted live tweets of fire incidents which contained links to detailed data of the event and key terms on the Firebrary. Firefighters could therefore know the full details of the incident in a matter of seconds, and could even be warned if they were about to face a ‘Black Swan’

Mapping the environment in which the
emergency services operate is also possible using Big Data. In the case of the
fire department in Amsterdam, this involved using data of all past fires to map
high-risk areas and using metrics such as economic background to develop
insights on these patterns. These insights can help with performance
measurement as well as deciding where stations should be located. Data science
such as this can be equally applicable to the police and ambulance services. In each of these sectors, traffic mapping is vital since reaching the site of the issue in
the fastest time possible is key. Tools such as
LUCA Transit (which analyzes over 7 billion daily events to provide
insights on traffic routing, volume and more) use Big Data to achieve this. You can read about some case studies

Protecting people from fires arguably
starts with prevention. Fire alarms, carbon monoxide detectors and sprinkler
systems are common ways of keeping our homes and offices safe and all of
these pieces of equipment emit data signals. “Nest” is a company that develops
tools for smart homes and their “Nest Protect” is an example of ‘Connected Data’ in action. The key
selling point is that the device can send an alert to your phone even if you are not at

The Nest Protect
Figure 1 : The Nest Protect

The potential uses of Big Data in the
emergency services are not limited to these examples, but they provide a
snapshot of the amazing potential that data has in benefiting society. The UN
has set 17 Sustainable Development Goals for the year 2030, and at LUCA, we
fully expect data science to play an important role in achieving them.

Don’t miss out on a single post. Subscribe to LUCA Data Speaks.

Leave a Reply

Your email address will not be published.