The technology behind identity graphs and smart cities: PwCehind identity graphs and smart cities: PwC

Summary

Operating a smart city requires advanced technology beyond the traditional identity management system solution.
Identity graph profiles create a web of data that ties people and things together based on relationships, common interests, patterns of behavior and other factors.
Smart cities can use the information they collect in various ways, such as leveraging analytics and data to engage visitors and improve business results for local merchants.

If you could create a city from scratch, where would you start? By laying out the streets? Figuring out the ideal spot for downtown? Turns out those are the easy parts. For our client, a major real-estate developer, the challenge of building a smart city quickly revealed itself to be as much of a data problem as it was a design problem.

The project in question: Turning 100+ acres of space into a unique, master-planned community and entertainment destination. The area will support everything from office space and residences to restaurants and concert venues. Built to be inclusive, walkable, and filled with natural parks and public space, the community, when completed, will support over 3,000 permanent jobs and generate $300 million of recurring economic impact.

But to get there, the developers should answer questions that no other planned community has faced. How will the city responsibly use data to understand its residents and visitors and respond to their needs? The answer may be found in the technology of an identity graph, which we’ll explore below.

Building a smart city, byte by byte

The development is designed to be an open community, where anyone can come and go as they please. There’s no requirement for visitors to provide identity information, but the developer would like to leverage data about the community to help maximize the value the city provides. Data collected through an identity management system could be used to increase the organization of facilities, help guide the planning of future events and streamline movement within the city.

Data could also be used for everything from planning the appropriate number of parking places to providing promotional offers in real-time based on a visitor’s preferences. Ultimately, it’s about giving people what they want, when and where they want it — all while improving the financial outlook for the community and its resident businesses.

Rethinking traditional solutions to make it more possible

A traditional identity management system might involve issuing ID cards or radio frequency identification (RFID) tags that have to be scanned when entering a building. Over time, management could piece together a rough profile of an individual, stored in a traditional relational database.

There are a number of major limitations with this concept. Card-based solutions require specific user interactions (scanning a card), don’t tie into online activities or multiple devices, and collect no real information about the person behind the card. That might work for a small-scale shopping experience, but it’s often too narrow and limiting for a smart city.

What the developer needed was a solution that solved these issues. It needed to be seamless and work across both the physical and virtual worlds. The system would need to tie together a sneaker purchase made via a smartphone app to a night at the movies a week later. And it needed to be robust enough to handle large volumes of data while being flexible enough to separate personally identifiable information (PII) from non-PII information to comply with privacy rules and meet user expectations.

Our proposed solution, PwC's Smart Venues, uses our proprietary identity graph, Signal Graph, that is built on Microsoft Azure.

Why sustainability matters to your smart venue

Lead the future in sustainability with Smart Venues

Explore this infographic for six ways next-gen technology can drive sustainable outcomes in your physical space.

Learn more

What’s an identity graph?

In general, an identity graph is designed to collect information about a user’s behavior and preferences, whether they’re offline or online (no matter how many devices or unique logins they use), and consolidate that information into a single, holistic profile.

Identity graph profiles aren’t exclusive to human users. Venues such as restaurants, for example, can have their own graphs. Location or event-based graphs can be connected to user graphs, creating a web of data that ties people and things together based on relationships, common interests, patterns of behavior and other factors.

What’s included in the identity graph? With Microsoft, PwC developed a domain model consisting of:

Purchasing activity (regardless of payment method)
Social media presence and activity
Location information (such as GPS data)
Media consumption habits
Offline movements and attendance at various venues
Interactions with IoT devices
General preferences about everything from dining habits to music tastes
Connections to other users (spouse, relatives, friends and so on) and locations (such as residence and workplace)

Once this domain model is loaded into Signal Graph, the data collection streams and storage components are deployed on top of Azure.

Technologies built to help power smart cities

Building a smart city requires technology like Microsoft’s that can offer security, flexibility and speed. To help illustrate, IoT sensors capture a wide range of customer and business data, giving the developers the ability to run complex, real-time analytics on nearly any aspect of city operations. Navigating through this enormous amount of data requires a sophisticated, customer-tailored query interface and engine supported by streaming anomaly detections.

Microsoft’s tools such as Event Hubs (real-time ingestion), managed Cassandra (time-series store) and Azure Kubernetes (compute platform) allowed PwC to deploy Signal Graph without worrying about scale and data integrity. This provided the ability to focus on helping solve customer problems through constructing appropriate domain models, queries and UIs.

Putting the identity graph to work

There are various ways that a smart city can use the information it collects, applying analytics to leverage data in ways that captivate residents and visitors while improving business results for merchants and the municipality.

We’re just getting started with all of this, and we now stand at year one of a ten-year project. It’s impossible to know what the future will bring as this community begins to grow, but the graph database sets the development up to be as future-proof as possible, allowing for new features to be added and old ones revamped, the graph itself growing and evolving just as the city grows and evolves.