Incorporate Video Analytics and IOT into
Your Enterprise Data Strategy
Public Safety and Smart Cities - Asia Pacific
Whether its enterprise solutions, data
warehouses or big data platforms, the data strategy is very similar.
Information is 'harvested' from existing data sources and is ultimately
reconstructed to provide business insight.
Outside of greenfield projects and major
upgrades, the data strategy is usually considered in the final mile. It's also
regularly positioned as an assumed benefit that is shadowed by operational and
strategic benefits already achieved.
Enter the Street Light Effect
David H Freedman coined the term "The Streetlight Effect" which people only tend to
look where it is easiest (i.e. under the streetlight even though they lost
their keys in the dark). We tend to use the data that was created for other
purposes and assume it can be restructured to provide another dimension of
Some examples of where this is done today:
Customer Behaviour- is derived purely from transaction history and assumed
demographics (not on the actual behaviour of the shopper). A customer can spend
30+ minutes browsing and evaluating a shop's contents. However, we only really
understand which products were extracted from shelves at the checkout.
Employee Retention- is measured on tenure, exit interviews and performance feedback
Campus/Office Utilisation - is generally measured
by overall campus access and/or predefined schedules e.g. meetings times,
lecture hall bookings etc
Performance Optimisation - Are measured by
throughput, total cost of manufacture and qualitative checks at points of
Prediction of Future events - Are generally derived from
historic events and, leverage regression techniques or spatiotemporal models
for location-based prediction.
All of these have one similar characteristic.
These data sources are already in place and serve a different mission to
business insights listed previously. As a result, we tend to search 'in the light'
because there are no other available data sources.
Enter Video Analytics as Real-Time Data Source
What if you had a whole new class of data
source that can be readily implemented to bridge these gaps in your data
Hitachi's Video Analytics Suite
Creates Real-time Data Sources from Video Streams
Video Analytics can provide these new data
sources. Data is extracted from video streams and ingested into your data lake.
These can be leveraged to paint a greater picture of real-world interaction. By
combining these new sources with existing data sources, we can provide insights
as determined by your required business outcomes.
Consider Other Real-World Sources from IOT
Other new data sources for business insight
may include solutions such as face recognition, geo-location based Apps and IOT
sensors. These can provide identity, presence and real-time sensor feeds.
Now Leverage This Source to Transform Your
Customer Behaviour- let's redefine the customer
journey by starting when they create a shopping list on an in-store app with
geolocation. The App also provides a 'pick list' for when the customer arrives
at the supermarket. By measuring the customers an ‘Activity Visualiser' (actual
route, deviation from suggested route, dwell time and many others) a new
profile of shopping behaviour along with store utilisation is provided. This
results in a significantly richer insight than transaction data alone.
Employee Retention - identifying a 'mega-trend' such
as employee churn at 18 months is easy to locate but understanding why can be
more difficult. Facial recognition to understand how employees utilise the
campus. A 'Company App' can be deployed to give insight as to how employees are
interacting with each other. Employees at risk will tend to disassociate. One
organisation could correlate the reduced amount of supervisor/employee time
with employee churn (supervisors thinking less time spent was a positive
reinforcement yet had the opposite effect)
Campus/Office Utilisation - Understand real-time campus
utilisation with people counters in/out of each room. Facial recognition in
situations where you want to verify attendees in a room.
Performance Optimisation - By correlating visual
activity analysis with production check points; exceptions, defects and
slowdowns can be associated with the actual real-life events. This supports
real-time optimisation techniques in contrast infrequent process reviews. This
provides lead indicators that can be modelled for predictive intervention (e.g.
maintenance, replacement etc).
Video Analytics and Industrial
IOT to Business Insight
Don't be afraid to look in the dark
Existing data sources are instrumental in providing
intelligence, but, don't be afraid of constructing new sources to enrich,
evolve and (potentially) disrupt your business insight. What other business
insights and enablement would you expect from video analytics??