Location Matters: Why hyper-accurate location data and analytics matters for Insurance
Download as PPTX, PDF0 likes526 views
The document discusses the importance of hyper-accurate location data and analytics in the insurance industry, highlighting trends such as the rise of insurtech and climate change data for risk modeling. It emphasizes the use of emerging technologies like AI, telematics, and computer vision to improve data integrity, decision-making, and risk assessment. Additionally, it details specific climate datasets and their applications for better understanding and managing risks associated with natural disasters.
Location Matters: Why hyper-accurate location data and analytics matters for Insurance
- 1. Location Matters Why hyper-accurate location data and analytics matters for Insurance Andy Chun | Regional Director – Prudential Corporation Asia Ryan Crompton | Managing Director – Risk Frontiers Daniel Tatro | Solution Architect – Insurance, Precisely Gerry Stanley | Senior Product Manager, Precisely Matthew Eagan | Editorial Director, Innovatus Media
- 2. Agenda Insurance industry trends – AI, data and more – Andy Chun Leveraging climate change data for risk modelling – Ryan Crompton The rise of Insurtech: emerging technologies for Insurance – Daniel Tatro Data sets for the insurance industry – Gerry Stanley Q&A Ryan Crompton Managing Director Risk Frontiers Gerry Stanley Senior Product Manager Precisely Daniel Tatro Solution Architect Precisely Andy Chun Regional Director Prudential Corp Asia
- 3. Empowering Better Risk Decisions RISK FRONTIERS Proprietary and Confidential
- 4. Risk Frontiers Core offerings: • Catastrophe Loss Modelling • Climate Risk • Resilience RISK FRONTIERS Provide innovative science-driven research, analysis & solutions to build safe & resilient communities Proprietary and Confidential
- 5. Climate Science Engagement RISK FRONTIERS Proprietary and Confidential • Partner Organisation of ARC Centre of Excellence for Climate Extremes (CLEX) • Climate Measurement Standards Initiative – Scientific Committee Member • Macquarie University climate risk projects • Bushfire & Natural Hazards CRC
- 6. • Observational & Climate Projections RISK FRONTIERS Climate Data Proprietary and Confidential
- 7. Climate Datasets • Observational • ERA5 reanalysis: ANZ & SEA, from 1979, 0.25 deg (~30 km) resolution • ERA5-Land reanalysis: ANZ, from 1981, 0.1 deg (~9 km) resolution • Climate Projections – CORDEX • AUS-22 & SEA-22 • Historical (1979-2005) & RCPs 2.6 & 8.5 (2006-2099) • 0.25 deg (~30km) resolution • Bias corrected to ERA5 RISK FRONTIERS
- 8. • Current parameters relate to extreme temperature & precipitation • A wide range of climatological & meteorological parameters can be calculated from the reanalysis & climate projection data • Outputs at native resolution Extreme temperature indices Definition Very hot days Annual count of days with max temp > 40°C Hot days Annual count of days with max temp > 35°C Very hot nights Annual count of nights with min temp > 25°C Hot nights Annual count of nights with min temp > 20°C Cold days Annual count of days with max temp < 15°C Very cold days Annual count of days with max temp < 10°C Cold nights Annual count of nights with min temp < 5°C Frost nights Annual count of nights with min temp < 0°C Highest max temp Annual max value of daily max temp Highest min temp Annual max value of daily min temp Lowest max temp Annual min value of daily max temp Lowest min temp Annual min value of daily min temp Extreme precipitation indices Definition Wet days Annual count of days with daily precip ≥ 1 mm Heavy precip days Annual count of days with daily precip ≥ 10 mm Very heavy precip days Annual count of days with daily precip ≥ 30 mm Max 1-day precip Annual max 1-day precip total Gridded Climate Parameters RISK FRONTIERS Proprietary and Confidential
- 9. RISK FRONTIERS Heavy Precipitation Days Aus – Historical Frequency of days with precipitation > 10 mm (ERA5-Land)
- 10. RISK FRONTIERS Heavy Precipitation Days Aus – Projected Frequency of days with precipitation > 10 mm (CORDEX, RCP8.5)
- 11. RISK FRONTIERS Hot Days SEA – Historical Frequency of days with maximum temperature > 35 deg C (ERA5)
- 12. RISK FRONTIERS Hot Days SEA – Projected Frequency of days with maximum temperature > 35 deg C (CORDEX, RCP8.5)
- 14. The Rise of Insurtech The global insurance industry is experiencing a massive technological shift.
- 15. Emerging Technologies Robotic Process Automation • Metaphorical software robots (bots) are trained to perform repetitive tasks • Examples: • Claims validation • Document acceptance Artificial Intelligence • Make decisions which normally require human level expertise • Designed to make decisions, often using real-time data • Using sensors, digital & remote inputs, they combine information, analyze instantly, and act Real-Time Data • Dynamic Weather • Social Media Sentiment • Dynamic Demographics Imagery • High resolution, aerial imagery • Vertical, oblique, infrared & 3D • Access to spatio-temporal blue/gray imagery Computer Vision • Field of AI that trains computers to interpret and understand the visual world • Accurately identify and classify objects such as buildings or swimming pools • Building size, roof pitch/area Telematics • Devices track driving habits and passes them onto the insurer • Consumers receive feedback to help them improve their driving skills • Premiums decrease as safe driving increases
- 16. Common Challenges Addressed Identifying Commercial Settings • What parcels are associated to any given address? • Do buildings span multiple parcels? • Do parcels fall within a building? Co-Tenant & Adjacent Risk • What other types of businesses operate: • At the same address • Within the same building • On the same parcel • Next door Common Ownership • Who owns the property that contains the address/building? • Do they also own more property nearby? • Tell me about those additional properties
- 17. Real World: Boots Off The Ground 1. Establish Ground Truth 2. Get the Big Picture 3. Work Smarter • A single point is not enough • Precisely geocoding, parcels, buildings and ownership to accurately defines the commercial setting • Proceed to step 2 • With the “ground truth” from Precisely, request imagery from a major provider • Allow underwriters and inspectors to visually inspect • Proceed to step 3 • Computer vision identifies and classifies features: • Buildings • Parking Areas • Greenspace • How does the roof “look”? • Good? Great!
- 18. Data for Insurance Enrich your data
- 19. Connect today’s infrastructure with tomorrow’s technology to unlock the potential of all your enterprise data Integrate Understand your data and ensure it is accurate, consistent and complete for confident business decisions Verify Analyze location data for enhanced and actionable business insights that drive superior outcomes Locate Power enhanced decision making with expertly curated, up-to-date business, location, and consumer data Enrich The four elements of data integrity D A T A I N T E G R I T Y
- 20. Data Portfolio World Boundaries Premium Postcode and Administrative Boundaries Community Boundaries Residential & Commercial Neighborhoods Cities, metros and Municipal boundaries Time Zones Risk Data Telco Data Dynamic Weather StreetPro® Classic StreetPro® Display StreetPro® Navigation StreetPro® Navigation Premium StreetPro® Traffic Enterprise & Desktop Routing and Drivetime Consumer Data Insights Demographics Purchasing Power Consumer Expenditure Geodemographics Consumer Vitality Crime Context Settlement Spaces Canada Wealth World Premium POI World Premium POI – Consumer Business Summaries Geofences Community & Venue Geofences Commercial Geofences Retail Destinations Retail Pitchpoint Address Fabric Property Features Property Attributes Building Footprint Parcels Precisely Boundaries Precisely Streets Precisely Demographics Precisely Points of Interest (POI) Precisely Addresses
- 21. Risk Bundle Bushfire Flood Address level risk attributes Leading risk rating plus associated risk and location information. Coverage • Australia • New Zealand Very High (5) High (4) Medium (3) Low (2) Negligible (1) Very High (5) High (4) Medium (3) Low (2) Negligible (1) Unknown (-1) Earthquake Storm • 500-year ARI peak ground acceleration • Soil zonation • Hail • Lightning • Tropical cyclone
- 22. Multi-peril Data for Bangladesh • Flood • Drought • Physiography
- 23. Address Fabric Australia • Building level geocode accuracy using the primary building location from GeoVision • Inclusion of the PreciselyID and the G-NAF PID • Improved awareness of location in relation to risk, particularly flood and fire • Improved understanding of service delivery point and primary building location G-NAF Address Fabric
- 24. Property Boundaries Building outlines Hydro lines G-NAF points
- 25. PreciselyID – Australia Suburbs & Localities and Postcodes GeoVision Buildings Multi Risk Bundle PreciselyID StreetPro suite WPPOI suite G-NAF Premium Complete Demographics suite CadastralPlus
- 26. Dynamic Future
- 28. Dynamic Location Data Household Number of people, marital status Ages, genders, ethnicities Health Education Home Housing type and tenure Location Neighbourhood Pre-defined segmentation Geodemographic based Affluence Lifestyle Socio-economic $£€ Economics Income, disposable income Purchasing power Spend by category Employment Unemployment MOBILE TRACE Place: Where do people go? Persistency: How frequently do they go there? Period: How long do they stay there? Path: Where do they go next?
- 29. Transition: Complexity to Insights
- 30. Create a dynamic profile for a location Inflows of Population: Mobile Trace Origins of Flows to One Destination Chart 1: Resident profile Chart 2: Resident profile plus inflow
- 31. Questions?
Editor's Notes
- #23: There is need for High Quality dataset for Flood, Draught & Physiography. Nityo needed our multi-peril data (essentially historical flood and draught data) for flood risk management (FRM). FRM utilizes our data and provides accurate, reliable, and up-to-date information on flood risks and events. Multi-Peril Bundle (3rd Party Data) This dataset of flood, Draught & Physiography
- #28: Fully API Driven, no dataset to deliver Integrated with PreciselyID to allow property level weather data and above Client defines thresholds of the data they want to see e.g. Winds over 30mph Detailed Weather data back 3 years, archive data back 10 years Track and analyze Cyclones, Convective storms, extreme cold or heat events and wildfires in the USA Weather Incident Reports, Daily Weather Forecast emails and Severe Weather Briefing options coming soon