carloscastilla - Fotolia

SAP HANA application helps with travel risk management

Prescient Traveler is a new travel risk management application that harnesses the power of the SAP HANA platform to deliver real-time alerts to business travelers.

Traveling these days can be a risky business. People who travel for pleasure can simply not travel to risky areas, but business or government employees may not have the same luxury. A new application called Prescient Traveler helps businesses and government agencies with travel risk management.

Built on the SAP HANA platform, the application helps travelers by sending alerts if they venture into dangerous areas, and it can tell them about the areas before they get there. It helps the security stakeholders in companies by keeping track of their travelers' locations and alerting them if trouble occurs.

Chicago-based Prescient started as consultants in the national security space with contacts primarily in the U.S. government, and in 2014 began to produce products for the private sector, according to Mike Bishop, Prescient's chief systems architect. The first of these was Prescient Comply which provides due diligence analysis and services to companies for things like mergers and acquisitions or third-party vetting.

SAP HANA key to providing analytical horsepower

SAP HANA is the key to providing the kind of analytical horsepower that can deliver results in real time, Bishop explained. "If you're aggregating data about vendors and trying to figure out who's reliable, having the ability to probe that both from the historical sense, but also real-time events -- what affects your vendors, what makes a company worth doing business with -- you can see how HANA might be valuable in that sense."

The company then turned its considerable analytic experience to developing a risk mitigation application for travelers. The resulting product, Prescient Traveler, provides a platform that helps travelers gain knowledge of areas they need to travel to. For example, they can use it to learn about local customs and taboos, or discover which areas to avoid and why -- a benign shopping area may become more dangerous at night. Users get this information on a mobile app or dashboard, but the back-end analysis is handled by Prescient's HANA implementation.

I determined that there would be no faster way to reach the fidelity of analytic product that we can through HANA.
Mike BishopPrescient's chief systems architect

This back-end system is where HANA does the heavy data crunching analysis of travel scenarios. Data from a variety of sources, such as social media and local news outlets, is pulled into Hadoop and MongoDB and then fed into HANA for processing and analysis. HANA's speed and ability to handle large amounts of structured and unstructured data allows Prescient's data analysts to assess threats and send out alerts to the travelers' mobile devices in real time.

"Our analysts use a tool we built called Content Builder that conditions all this threat input -- essentially the foundations for threat assessments at a country or city level -- for them to produce no-go zones and cautionary zones," Bishop said. "That's what shows up on a map on our mobile app and the dashboard."

Geospatial analysis tracks traveler locations

In addition to integrating data from social media, RSS feeds and news sources, Prescient does geospatial analysis in Esri ArcGIS that keeps track of a traveler's location in real time. The mobile app is location aware and sends the traveler's latitude and longitude to the system, where HANA calculates in real time where the traveler is in proximity to areas that have been identified as dangerous. It then kicks off an alert based on user-defined settings. "On the mobile app I can say that if I'm within 200 meters I want to get a heads up, and that notification is actually being generated by calculations in HANA," Bishop explained.

The amount of data that Prescient Traveler deals with is vast. There are about 350 different information categories, and those categories all have different data sources. RSS feeds are one category, and Prescient currently has 1,672 worldwide RSS feeds from sources like government agencies and local news outlets. This unstructured data is routed into HANA, which then does natural language processing for sentiment analysis and fact extraction based on Prescient's object model.

"Our object model is seeded by our subscribers and our own definitions of what is a threat domain or a category of information that we want to track, for instance crime or terrorism," Bishop explained. "There's databases or keywords for each one of those and lexicons to say that if someone's talking about crime in an RSS feed, we figure out what crime they're talking about and if there's any geocodable information in it. We parse that out and it becomes an incident in HANA, and that begins to form a statistical backdrop of where things occur in the world and where certain criteria are met. For instance, we can say that this is happening now or this happened in the last 24 hours; we'll push that forward to an end user."

HANA makes the cut after extensive testing and research

Prescient Traveler was developed from an entirely new greenfield HANA implementation, according to Bishop. Design work began in September 2014 and lasted a few months, during which Prescient conducted extensive testing and research that included examining other platforms. After this, it was determined that only HANA had the capacity and scalability to handle the kinds of workload that the system required.

This did not come easily or without considerable soul searching. "I determined that there would be no faster way to reach the fidelity of analytic product that we can through HANA," Bishop explained. "It took me a few months to come to terms with that because of the cost. And I don't think anyone will dispute that it's a fairly expensive platform, but the fact that we're an OEM partner allows us to sell that analytic power at a very modest price."

Prescient Traveler currently runs on SAP HANA SPS10 on premises at Prescient's primary data center in Chicago, with an autonomous failover on-premises implementation at a disaster recovery site in the Western U.S. Having it on two different regions of the national grid ensures high availability and disaster resilience, which is critical to the application's success.

"We built this entire Prescient Traveler platform to not only serve the day-to-day business needs for safety and business continuity, but also for contingency operations when there's truly a catastrophic crisis in certain areas of the world," Bishop said. "Even [if there's a catastrophe in] our own HQ area here in Chicago, we can continue operations at our DR sites. The dashboard that the security stakeholders use is available in a fully decentralized fashion; they can use it from anywhere in the world."

In addition to the safety alerting functions, Prescient Traveler has a component called Secure Concierge that lets users assess the safety of local vendors when they are traveling, such as hotels or transportation services. Using the same HANA-based due diligence analytics that powers Prescient Comply, Secure Concierge provides real-time results.

Prescient Traveler is available for enterprises on a monthly, per user basis with significant volume discounts. For now it's structured primarily for multinational firms, but the company plans to offer single-trip subscriptions to individual travelers in the future.

Next Steps

Find out how SAP HANA stores data.

Learn how to prepare for an SAP HANA installation.

Read what the experts are predicting for SAP in 2016.

Dig Deeper on SAP HCM