News

Using business intelligence to track hurricane aftermath

Tana Sabatino

This article originally appeared on the

    Requires Free Membership to View

BeyeNETWORK.

On the evening of October 25, 2005, just 36 hours after Hurricane Wilma ravaged South Florida, power was restored to my Fort Lauderdale home. At that moment only 6,500 out of 862,800 people in Broward County had power. And I was one of them. I felt as if I had won the lottery. Apparently almost all of my fellow residents’ substations and transformers were out of service after the hurricane. Priority was given to fire, police and hospitals. Fortunately, I share a grid with a critical care hospital. Once the emergency services were supported, priority was given. This priority was based on the number of people that would be affected when the problem was fixed. So if a large area can be brought back up by a repair, it would trump a small subdivision.

Over the next few days, from the comfort of my air-conditioned lit office, I used my high speed internet connection to track the progress of the power restoration for my fellow Browardians. At 1:00 p.m. on October 28, the Florida Power & Light (FPL) website reported that 231,300 people had their power restored. This left 631,500 in Broward County without power. All but two substations were back online. They are now painstakingly cutting debris off lines, rebuilding poles and finding and repairing broken lines across the entire county grid. According to Florida Power & Light, the people of Broward County will have their power restored by November 22. Shockingly, this is more than four weeks after the hurricane! Obviously, this is angering residents. With our American lifestyle, 30 minutes without power seems like a lifetime. Now imagine 30 days…  Fingers are already being pointed. "Didn't they learn anything from Andrew? Floyd? Katrina?" "Shouldn't they have inspected and replaced those wooden poles BEFORE they snapped??" "How could they have let this happen???"

Figure 1: Power outages caused by Hurricane Wilma.

But taking a step back, I looked at this story from another angle and was rather impressed. Briefly consider their knowledge and business intelligence capability. They have the information to make intelligent business decisions. They know not only the exact number of people affected at any moment, but can also identify the key repairs that must be made to support emergency services. They must also consider the most efficient repairs to bring back the largest number of people as soon as possible. 

One can compare this to business intelligence within the insurance industry. It is inconceivable to expect an insurance company to create a similar picture hours after a disaster, showing their customer base and risk exposure by line of business. 

Over the next few months, we will see the projected high costs of Wilma to the insurance industry. These projections are already here. Catastrophe modeler Risk Management Solutions initially projected that insured losses would be between $6-10 billion dollars. This number was subsequently increased to $8-12 billion. I expect these numbers to fluctuate by several billion over the next few months. In fact, Hurricane Katrina numbers have fluctuated anywhere from $2.5 billion to $18 billion over the past two months. It will really take a couple of years to completely add the final loss numbers. Why is this? Why do we have such little business intelligence with regard to the risks we face?

In 1992, Hurricane Andrew was the first natural disaster to truly show the insurance industry their inability to assess risk. Although the largest insurers in South Florida were severely overexposed, they had no idea that they were. They gladly sold the policies, and collected the premiums with little consideration of what this would mean if the area actually experienced a hurricane that was category two or greater. Many insurance companies had to file for bankruptcy. Others managed to absorb a one billion dollar loss, but realized they should have had better business intelligence and refused to sell insurance in Florida anymore. 

Now it's over 13 years later. While we may have become slightly smarter, we still know very little about our true risk as an industry. We know even less about these risks for a particular company. Moreover, understanding our total loss after the winds blow through is even more challenging.

Numerous factors come into play. The first consideration is the different types of insurance that factor into the equation. The most obvious of these factors is homeowners insurance. Does it include flood damage? How about wind damage, which is a separate policy for Floridians? Does it protect inside items or only structures? Is it for current value or replacement value? Does it include coverage for temporary housing? Similar questions are frequently asked about commercial property insurance. Another consideration is business continuation insurance. With the possibility of electricity being out for as long as a month, huge claims could be made. Personally, I am hoping that my auto insurance will cover the dents in my car, which were caused by flying debris.

Reinsurers also take a heavy hit in these situations. In September, Standard & Poorsrevised its outlook on the global reinsurance industry from stable to negative, citing the impact of Hurricanes Katrina and Rita. A revised outlook means expected rating downgrades. Clearly, Hurricane Wilma has not improved this outlook. Assessing a reinsurer's risk is even more daunting than that of a direct insurer. This is because direct insurers take on huge blocks of business from insurers. Although they are responsible for part of the risk, direct insurers can take business from other insurers as well. Furthermore, they may be several degrees from the individual filing the claim.

Assessing the damage is also much more difficult than the power company. We don’t have a grid where we can see what's on and what's off. Instead, we must send claims adjusters to everyone who filed a claim and look for ourselves. Before this can happen, though, we must wait for the claim to be made. This process often takes months.

Then there is the age old question—what is our responsibility? Is the flooded house a result of the roof being torn off or from the storm surge? In the case of Katrina, was the break in the levees considered part of the hurricane or was it a separate event? These types of questions are usually asked. They can take months or even years to determine.  For instance, it took over two years for the Supreme Court to rule that the attacks on the World Trade Center constituted two separate events, one for each tower. Another example of this was last year in Florida, when the west coast was hit by two hurricanes back-to-back. It was decided that simply because the hole in the roof had not been repaired yet, the resulting damage of the rain and wind during the second hurricane had nothing to do with that of the first. This means that insurers can collect two deductibles, which reduces their overall cost. So we cannot know our true costs until all of these parameters are determined.

Both knowing your potential risk exposure before an event occurs and understanding your losses after it require an in-depth understanding of your book of business, at a much deeper level than we can possibly do today. This is primarily due to the way companies set up their infrastructures, as well as their inability to share between the different layers within an organization.

Companies typically segregate themselves by line of business. For example, the P&C subsidiary rarely communicates with the life subsidiary. Personal lines are run from a different office and set of systems than commercial lines. Operations are regionalized. And the reinsurance division never talks to anyone else. 

After several hours of wind, Florida Power and Light knew exactly how many people were affected and where they were located. The same thing cannot be said about the insurance company. 

Just how long does it take to calculate your risk when the data is in disparate systems, across different company divisions and multiple product lines? When does the reinsurer become involved? How easy is it to track the progress of receiving and paying claims across this tangled web? When will a company know its full loss?

The insurance industry is one of the most highly criticized industries today. The individual homeowner cannot possibly understand why it takes so many months for a claim to be paid. I can truly understand this only after seeing the underpinnings of a typical insurance company. While we have so much "data," it's impossible to aggregate it into any meaningful form.

Insurers need better business intelligence information to both prepare for these events and understand their losses. An insurance company must break down the barriers between the different systems and lines of business. By doing this, the company can support its customers efficiently and manage its business intelligently. 

Over the next several months, I will explore how XML (eXtensible Markup Language) and the proliferation of data standards are finally breaking down these barriers, without having to replace existing technology. 

Rather than sitting in the dark (like many Floridians will do over the next month), insurers can run their businesses more intelligently. This can occur if they pursue strategies exposing the business intelligence information they seriously need.


There are Comments. Add yours.

 
TIP: Want to include a code block in your comment? Use <pre> or <code> tags around the desired text. Ex: <code>insert code</code>

REGISTER or login:

Forgot Password?
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
Sort by: OldestNewest

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to: