Coming of Age
Innovation and automation are being built on strong data foundation
There have been on-going discussions the past several
years about how Big Data and Artificial Intelligence
(AI) can be used to modernize physical security systems
and operations. The truth of the matter is that
digital transformation has already been taking place
for quite some time, albeit in small steps with what is now a very
fast-paced technology evolution.
What were once high-level theoretical discussions on mining
data from both traditional physical security operations and
scores of new and emerging IoT devices have now become a reality.
This next generation of innovation and automation is being
built on strong data foundations and data intelligence. And it is
changing perceptions on how physical security systems need to
be designed, implemented, and managed in this new era of datadriven
physical security solutions.
Perhaps the most significant take-away from the genesis of data-
driven physical security is the change in mindset from reactive
to proactive systems technologies – the ability to autonomously
process data in real-time to reduce friction in business process
execution by eliminating human intervention, and to deliver actionable
insights and even predictive analysis.
This has the potential to dramatically increase the overall effectiveness
and efficiency of physical and cyber security operations,
mitigating risk across the enterprise. As a result, physical
security technologies and operations will continue to leverage
real-time business intelligence across the enterprise to facilitate
real-time decision making.
THE BIG DATA INTEGRATION CHALLENGE
One of the biggest challenges faced by security professionals today
is the ability to harness and analyze the tremendous volume
of data being generated by physical security operations containing
spatial, sensor and transactional data and network of devices
to derive meaningful actionable intelligence. With the bulk of
such data being unstructured data, new data-driven solutions are
required to contextualize this disparate and fragmented information
in a seamless way.
One of the reasons this presents such a daunting challenge is
that physical security operations typically employ several different
point-of-control systems and associated sensors / IoT technologies
such as Physical Access Control Systems (PACS), video
surveillance, intrusion sensors, biometrics, visitor management,
dispatch, incident management systems and more. For the most
part, these systems are not highly integrated and provide siloed
views of activity which fail to provide a complete picture, precluding
intelligent insights from being drawn since all the data is
not being analyzed in the same context.
The first step in resolving the fragmented challenge is to bring
data from these otherwise disparate system technologies together
onto a special purpose unified platform. It requires discovering,
connecting, integrating, transforming, managing, analyzing, and
storing valuable data insights and enable the execution of applications
that are smarter and intelligent – net-net derive 5-10X value.
The next step is to understand how and where to use this
newly found data intelligence to improve business processes. For
example, can one apply data science to predict occupancy levels
of a building or a floor to better schedule employees coming into
a facility during today’s pandemic times? Can one use data intelligence
and predict what physical security IoT sensors or devices
will fail within 30 days, and make the retroactive repairs as opposed
to being reactive? There are numerous such promises of the
data science discipline.
Last, but not the least, is to make this all very simple for users
to consume. The data science, the machine learning and the arti-
ficial learning can only work if the intelligence and insights from
the myriad data sources are driven to the applications without
heavy lifting or understanding of computer and statistical science.
It should be very easy for users to incorporate these applications
into business operations. New developments in AI powered
software platforms can help alleviate both physical and cyber
threats by inferencing from large volumes of data, and quickly
triangulate small subsets of pertinent information, provide actionable
insights etc. This not only simplifies and improves physical
security operations while delivering tangible ROI and lowering
TCO, it also automates major security functions related to
Physical Identity and Access Operations, SOC Automation and
Cyber-Physical Security defense.
THE DATA-DRIVEN PLATFORM SOLUTION
New data driven solutions are now coming to market that promise
to address many of the inherent legacy system and data integration
challenges that plague the physical security industry.
A prime example is the recently unveiled Vector Flow platform,
which is capable of processing and analyzing vast amounts of
data from otherwise disparate security systems, data stores and
This innovative new solution derives actionable intelligence from physical security data using advanced Artificial Intelligence
(AI) and Machine Learning (ML) algorithms that empower a
whole new range of highly advanced automated physical security
applications that were previously unattainable. A truly converged
data driven physical security solution, the new platform presents
myriad opportunities for digital transformation and improving
several major business processes such as:
Physical Workforce Identity & Access Management (PIAM)
which unifies and streamlines identities, access and badges; on/
off boarding processes; physical access provisioning; access audits
and compliance to regulations; risks analysis and prescriptions
and mobile self-service.
Physical Security Operation Center (SOC) Automation for autonomous
false alarm reduction and reporting, unlimited device
monitoring and auto optimization, auto configuration of devices,
and auto detection of faulty devices, along with provisioning automated
virtual SOC assistants using AI/ML playbooks.
Cyber-Physical Security to enforce “defense in depth” using
advanced AI models that detect vulnerabilities in physical security,
IoT and building automation devices to prevent cyber surface
attacks and reduce organizations’ exposure to overall risk.
Vector Flow’s data-driven solutions are already implemented
at several Fortune 1000 companies. This includes a global telecommunications
provider with over 450,000 identities – the Vector
Flow team replaced the legacy PIAM application with a new
AI-enabled physical identity lifecycle application. The new solution
promises to save millions of dollars in direct costs over the
course of the contract while increasing overall security operations
productivity, compliance to regulations and delivering valuable
service to the enterprise.
In another example, a top Research and Pharmaceutical company,
focused on anti-viral drugs and treatments, deployed Vector
Flow’s AI-enabled solution to streamline SOC operations,
enable AI-driven device health prediction, reduce false / nuisance
alarm counts and streamline SOC functions by establishing and
measuring KPIs across all SOCs including
compliance with newer regulations such as AB
685 & SB 1159 for building occupancy during
This article originally appeared in the September / October 2021 issue of Security Today.