Blogs

Weather analytics help insurers weather the fiercest storm

Post Comment
Big Data Evangelist, IBM

Weather is a risk factor for everyone. Even the balmiest climes can be susceptible to the ravages of heat, drought, windstorms and torrential downpours. However, though we may never be able to predict local weather perfectly, we can insure ourselves against the risk scenarios that we think most likely—for example, consider how mortgagors in many low-lying regions are required to purchase flood insurance.

In five years out of one decade tracked by The Weather Company, weather-related damage accounted for more than $50 billion in insured property losses to the US property and casualty industry. Indeed, even a relatively low chance—say, 10 percent—that a hurricane will strike a location within the next few days is material enough for an insurer or retailer to take action to protect insured assets or stock up on storm supplies. Accordingly, insurance companies have long been particularly avid users of predictive analysis to gauge weather-related risks.

Among the principal uses of weather data analytics in the insurance industry are the following:

  • Underwriting
    Altogether, weather disasters cost US insurance companies an average of $25 billion a year in claims. Data-driven predictive weather insights help insurers mitigate catastrophic scenarios in high-risk areas by using probabilistic weather forecasts as geocoded inputs to policyholder risk profile models. Moreover, multivariate models can help quantify the effects of severe weather on claims, predict concomitant effects on call center volumes and estimate claim activities both in real time and in the future. For automobile insurers, weather data can complement telematics data on driving behavior to provide understanding of individual risk profiles associated with rain, snow, sleet and other such conditions.
     
  • Operations
    On-site response is integral to how insurance companies address weather-related claims. By using analytics to predict the effects of impending storms, insurers can enable timely on-site response by rapidly adjusting staffing and supply chain strategies to service policyholders in affected locations.
     
  • Claims
    Predictive insights into impending storms could save up to millions of dollars each year in high-risk areas. Proactively alerting customers about severe weather events can help mitigate or avoid damage, commensurately reducing claims, with tailored messages sent through multiple channels to alert customers both before and after weather events. Moreover, insurers can use fraud analytics to flag suspicious claims that cite weather conditions that appear inconsistent with expected data. Suspicious claims can be flagged for further review, and certain claims can be automatically processed—allowing customers to receive payment quickly while saving claims reviewers time.
     
  • Experience
    Customers are vocal about hiccups in insurance company response to weather-related claims. By tapping into weather data, insurers can identify affected policyholders and monitor social media discussions about perceived lapses in the company’s customer service. Doing so will allow insurers to evaluate the effectiveness of their response, making adjustments as necessary to enhance the policyholder experience. Similarly, correlating weather data with driving behavioral data can help identify policyholders who should be targeted for safe driver discounts, proactive alerts and real-time assistance.

http://www.ibmbigdatahub.com/sites/default/files/insurancechallenge_embed.jpgStakeholders in the insurance industry who stand to benefit from data-driven weather analysis include chief risk officers, claims managers, field claims operations managers, customer service professionals and chief marketing officers. Such analysis should incorporate both an insurer’s own transactional data (policy underwriting, premiums, claims and the like) and weather data that enables comprehensive modeling of relevant scenarios. It might also include third-party information (historical, real-time, forecasted) about severe weather, seasonal patterns, air quality, pollutants, and pollen and other allergens, and it should include data derived from remote sensing earth imagery.

Learn how to do your own weather-driven analysis by attending IBM Insight 2015, scheduled for 25–29 October in Las Vegas. Together, IBM and The Weather Company will demonstrate how to use weather data packages and IBM data science tools—including Apache Spark as a Service in the Bluemix cloud—to build powerful smart data applications for meteorological forecasting to help you address urgent weather-related business challenges.

You, too, can use Spark as a Service to hack your weather-related analytic challenges. Find out how weather data and Watson Analytics are being used to predict insurance claims, then take an in-depth look at insurance weather datasets, discovering how insurers are affected by weather forecasts.