Patent Application
20150046194
Catastrophes caused by natural disasters such as earthquakes, floods, tsunamis, snowstorms, hurricanes and terrorist attacks result in billions of dollars of losses each year. Insurance can provide for protection for many of these catastrophes and insurance companies generally have many procedures for handling these tragic events. A large part of responding to such a catastrophe involves the insurance company evaluating and assessing damage, performing site visits for insurance adjusting and estimating and claims personnel staffing. Traditional methods for catastrophe mapping and claims response generally rely heavily on forecasted models. For the most part, the insurance company will not know the extent of the damage in an area until claims personnel can travel to the location and perform analysis. Many times access points to these affected areas are often impaired and obstructed, making it difficult for these assessments to occur. As a result, there is a significant time gap between the data in the insurance company's forecasted models and when ground-level data from claims teams is available to plan for damage assessment and claims response actions.
Speed of damage assessment and claims response is a critical factor and performance component of any insurance company, and is one of the leading indicators used by firms such as J.D. Powers & Associates in ranking insurance carrier performance. Generally, it can take many days and even weeks before claims personnel could reach damaged areas after a catastrophe event. Currently, use of satellites to evaluate afflicted areas may not provide the most current data as it may take days between satellite passes. Furthermore, weather patterns can linger over weather related catastrophes for days, which can reduce the usefulness of the satellite imagery.
Accordingly, it would be desirable to have a system that could provide insurers and consumers with real time, accurate and timely data during and after catastrophes to speed damage assessment, claims response, adjusting and eventual settlement with the insureds. Such a system would benefit both the insurers and the insureds greatly by expediting the claims process by both sides during a catastrophe.
The present invention in some embodiments relates to, a system for intelligently compiling and assessing pictorial based data for insurance claims operations, the system comprising at least one processor; a memory coupled to the at least one processor; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the at least one processor, the one or more programs including instructions for: segmenting a selected geographic location into a plurality of regions to form a virtual pictorial mapping of one or more insured properties; accessing crowd sourced based pictorial data for each of the plurality of regions; compiling the crowd sourced based pictorial data into the virtual mapping by correlating location based information associated with the pictorial data to the virtual mapping; determining if one or more gaps exist within the virtual mapping; accessing pictorial data for the one or more gaps; associating the accessed pictorial data to the one or more gaps; evaluating the virtual mapping to determine an insurance based action related action; and issuing a communication to an insurance claims system related to the evaluation of the virtual mapping.
In other embodiments, the invention relates to a computer program, comprising a computer usable medium having a computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method for assessing catastrophe damage in an afflicted geographic area, said method comprising: dividing an area into a plurality of photo regions to form a virtual area map; receiving crowd sourced digital images; assigning the digital images to one or more regions on the virtual area map; determining if one or more regions require data supplementation; acquiring additional image data for the regions requiring data supplementation; evaluating the virtual area map to form an insurance recommendation; and transmitting data to a insurance entity system related to the insurance recommendation.
In other embodiments, the invention relates to a computer-implemented method for intelligent automated catastrophe site evaluation comprising: segmenting a selected geographic location into a plurality of regions to form a virtual pictorial mapping; accessing crowd sourced based pictorial data for each of the plurality of regions; compiling the crowd sourced based pictorial data into the virtual mapping by correlating location based information associated with the pictorial data to the virtual mapping; determining if one or more gaps exist within the virtual mapping; requesting pictorial data for the one or more gaps; associating the requested pictorial data to the one or more gaps; evaluating the virtual mapping to issue an insurance based alert message; and transmitting the alert message to a policyholder device
A more detailed understanding may be had from the following description, given by way of example in conjunction with the accompanying drawings wherein:
Disclosed herein are processor-executable methods, computing systems, and related technologies for the processing and analysis of crowd sourced data for administration and management of catastrophe related insurance claims. The ability to quickly and efficiently manage enormous catastrophes, while simultaneously keeping conventional day-to-day claims serviced, is critical for a property and casualty insurance company. When properly handled, a catastrophe claim can demonstrate to the policyholders and the public the true value of insurance. During a catastrophe, the insurance company needs to quickly determine the scope and extent of the catastrophe and how to handle and service policyholders and how to evaluate insured properties within the catastrophe. Many issues will arise such as how a catastrophe team will be staffed especially since the physical environment of the catastrophe will present a logistic and practical challenge to a catastrophe team member such as insurance adjusters. The damages to property will be many and varied, requiring skill and ingenuity in many instances to estimate the scope of the loss and calculate and negotiate the various insurance settlements for the insureds' properties.
In today's mobile technology environment, there is an increased level of photo activity as people document catastrophe damage to share with others, for example, pictures uploaded to social media sites/channels or news sites. Generally many people take digital pictures with their camera phones or similar devices and post them on electronic social networking or social media sites like Twitter, Facebook, Instagram, Google+, etc. These photos are generally taken with mobile devices such as smartphones which embed geographic information system (GIS) data or any type of methodology or system designed to capture, store, manipulate, analyze, manage, and present all types of geographical data, which may be referred to herein as “geocoded data” or “location based data,” related to each photo. By combining the crowd sourced photos having geocoded data and geographical data, such as GIS data for locations of insured properties, a new and powerful “virtual ground-level walk-thru” of catastrophe areas may be constructed in embodiments of the present invention by intelligently and selectively overlaying the crowd sourced photos over the geocoded data of the insureds' properties and over geocoded data relating to roads, utilities and other facilities.
With this overlaying or mapping the insurance company can more accurately deploy claims resources to those geographical areas that are hardest hit, and can even guide claims resources to those geographical areas, particular properties in geographical areas, locations and around road hazards. With embodiments of present invention, initial claims teams can perform initial intelligence gathering from a centralized location, allowing the insurance company's claims teams to focus on working with insureds on claims, and providing a good claims experience. Insurance companies can proactively reach out to insureds to assess their condition based on photos collected having geocoded data matching or near locations of insured properties, instead on waiting for insureds to contact the insurance company. Utilizing embodiments of the present invention, the insurance company can quickly assess total insurance and operational risk exposure from a realized catastrophe event, allowing enterprise risk management teams to re-allocate financial assets for claims in a more timely manner. Insurance companies can also provide claim advances to insureds within hours of the storm or other catastrophe, providing them with resources for shelter and emergency food based on an assessment of photos from their immediate area. The virtual maps of embodiments the present invention may also be shared with Federal and State emergency responders for use in disaster response, or with the public for use allowing the identification of family and friends in the impacted area. Additionally, timestamps in photos would allow for “time-lapse” maps of areas, which would be useful in future catastrophe modeling to determine how certain areas react to certain types of catastrophic events.
The term “crowd-sourced” as used herein means, for data, data that is collected in a process of obtaining the data from members of a group or the general public, either in response to a request or by obtaining data from one or more repositories of data. The crowd sourced data is generally obtained from a large number of individual providers of data. The providers are not necessarily requested in advance to provide the data. The persons requested to provide the data may exclude, or include primarily persons other than, those with an existing business arrangement with the insurance company, such as insurance company employees, contractors, agents, policy holders, adjusters and the like.
In operation, insurance subsystem 160 may implement spider/webcrawler technology to search via network 140 for data such as crowd sourced pictorial data in the form of digital photographs and associated location based information on third party web systems 150a-n that have been uploaded to third party web systems 150a-n by a plurality of third parties. Insurance subsystem 160 may also communicate with user devices 132a-n to obtain data such as digital photographs and associated geocoded data directly from one or more users.
Referring to
The insureds' information database 116 may store information, data and documents that relate to insureds' policies such as home, business and/or automobile related policy information as well as location information. Crowd sourced information database 118 may store information, data and documents from user devices 132a-n and third party systems 150a-n. Insureds' information database 116 and crowd sourced information database 118 may be spread across one or more computer-readable storage media, and may be or include one or more relational databases, hierarchical databases, object-oriented databases, one or more flat files, one or more spreadsheets, and/or one or more structured files. Insureds' information database 116 and crowd sourced information database 118 may be managed by one or more database management systems (not depicted), which may be based on a technology such as Microsoft SQL Server, MySQL, Oracle Relational Database Management System (RDBMS), PostgreSQL, a NoSQL database technology, and/or any other appropriate technology.
Communication between the insurance data system 110 and the other elements in the example architecture 100 of
Referring still to
In operation, devices 132a-n and terminal 130 may be used to update insureds about the status of their claim, condition of their property, provide payments and settlements, and other claims related activities. The web site system 120 may include a web application module 122 and a HyperText Transfer Protocol (HTTP) server module 124. The web application module 122 may generate the web pages that make up the web site and that are communicated by the HTTP server module 124. Web application module 122 may be implemented in and/or based on a technology such as Active Server Pages (ASP), PHP: Hypertext Preprocessor (PHP), Python/Zope, Ruby, any server-side scripting language, and/or any other appropriate technology.
The HTTP server module 124 may implement the HTTP protocol, and may communicate HyperText Markup Language (HTML) pages and related data from the web site to/from client devices 132a-n and 130 using HTTP. The HTTP server module 124 may be, for example, a Sun-ONE Web Server, an Apache HTTP server, a Microsoft Internet Information Services (IIS) server, and/or may be based on any other appropriate HTTP server technology. The web site system 120 may also include one or more additional components or modules (not depicted), such as one or more switches, load balancers, firewall devices, routers, and devices that handle power backup and data redundancy.
Referring still to
The example architecture 100 of
Each or any combination of the components/modules 112, 114, 122, and 124 shown in
Referring to
Storage devices 220 may include suitable media, such as optical or magnetic disks, fixed disks with magnetic storage (hard drives), tapes accessed by tape drives, and other storage media. Processor 210 communicates, such as through bus 211 and/or other data channels, with communications interface unit 212, storage devices 220, system memory 230, and input/output controller 240. System memory 230 may further include non-transitory computer-readable media such as a random access memory 232 and a read only memory 234. Random access memory 232 may store instructions in the form of computer code provided by application 214 to implement the present invention. One or more computer programs may be stored in memory, or computer usable media, such as storage devices 220 and random access memory 232, in the form of computer readable program code adapted to be executed by at least one processor, such as a processor central processing unit 210. The one or more computer programs may include instructions for performing steps of methods of embodiments of the invention described herein. System 200 further includes an input/output controller 240 that may communicate with processor 210 to receive data from user inputs such as pointing devices, touch screens, and audio inputs, and may provide data to outputs, such as data to video drivers for formatting on displays, and data to audio devices.
Storage devices 220 are configured to exchange data with processor 210, and may store programs containing processor-executable instructions, and values of variables for use by such programs. Processor 210 is configured to access data from storage devices 220, which may include connecting to storage devices 220 and obtain data or read data from the storage devices, or place data into the storage devices. Storage devices 220 may include local and network accessible mass storage devices. Storage devices 220 may include media for storing operating system 222 and mass storage devices such as storage 224 for storing data related to catastrophe data, insured customer information and claims related data and information such as claim advance and settlement data.
Communications interface unit 212 may communicate via network 204 with other computer systems such as third party servers 208 and 209 as well as other internal and external servers, computer systems of remote sources of data, and with systems for implementing instructions output by processor 210. Insurance processing and data system 206 may also be configured in a distributed architecture, wherein databases, data storage devices and processors are housed in separate units or locations. Some such servers perform primary processing functions and contain at a minimum, a RAM, a ROM, and a general controller or processor. In such an embodiment, each of these servers is attached to a communications hub or port that serves as a primary communication link with other servers, client or user computers and other related devices. The communications hub or port may have minimal processing capability itself, serving primarily as a communications router. A variety of communications protocols may be part of the system, including but not limited to: Ethernet, SAP, SASTM, ATP, Bluetooth, GSM and TCP/IP. Network 206 may be or include wired or wireless local area networks and wide area networks, and over communications between networks, including over the Internet.
One or more public cloud, private cloud, hybrid cloud and cloud-like networks may also be implemented, for example, to handle and conduct processing of one or more transactions or processing of the present invention. Cloud based computing may be used herein to handle any one or more of the application, storage and connectivity requirements of the present invention. For example one or more private clouds may be implemented to handle catastrophe data and crowd sourcing data of the present invention. Furthermore, any suitable data and communication protocols may be employed to accomplish the teachings of the present invention.
Photos utilized in the present invention generally may be geocoded or geotagged. Geotagging results in the photo having accessible geographical identification metadata that usually consists of latitude and longitude coordinates, as well as altitude, bearing, distance, accuracy data, and place names. Geotagging can assist in the present invention by have the insurance subsystem search for images taken near a given catastrophe location by entering latitude and longitude coordinates into a suitable image search engine. Location identification may also include geocoding or using non-coordinate based geographical identifiers, such as a street address, name of a business, non-profit, facility, individual or landmark associated with the location and finding associated geographic coordinates for the photos or pictorial data in the present invention.
Generally, pictorial data or photos may be accessed and stored in a variety of formats including the JPEG file format where the geotag information will be typically embedded in the metadata stored in Exchangeable image file format (EXIF) or Extensible Metadata Platform (XMP) format. Location information such as latitude and longitude may be stored in units of degrees with decimals, such as in the form of global positioning coordinates, such as Global Positioning System (GPS) Latitude: 68 deg 48′ 66.73″ S; GPS Longitude: 12 deg 35′ 26.74″ W; GPS Position: 44 deg 28′ 61.34″ S, 11 deg 34′ 36.70″ E or alternatively location information could also be presented in formats such as: GPS Latitude 52.34512; GPS Longitude: 20.41736 and GPS Position: 47.65611 11.20233.
It is contemplated that each pictorial data or photo may be content analyzed by one or more algorithms for both content and/or quality. Certain features in the photo may be detected by such analysis to supplement the location based data so that the picture best encompasses one or more insured properties attributable to one or more insurers. The photos may also be ranked based on content and/or image quality especially where multiple photos may be accessible for the same general location. In such a ranking, the photos that best encompass the insured property and have the best image quality would be selected for the virtual mapping.
Where GIS is unavailable for that device, the IP Address of the originating source, such as the mobile device will allow for mapping of the photo to the areas 422 and 424, or if image is from a mobile device such as in 440, location could be established by cell tower triangulation, based on triangulating the cellular towers used to submit the photo.
In the present invention, photos forming the virtual mapping may also be periodically collected and time stamped to form a real time virtual mapping that can be compared to determine if any damage or change in damage, has occurred to one or more insured properties at certain instances in time. For example, on a periodic basis, systems may access social media sites and other sources of crowd based pictorial data, send requests for photos, and generate updated and time stamped real time mapping. The time stamped mappings may be stored and evaluated for changes in damage to insured properties.
Gaps may occur where the available photo data does not completely cover the respective segment for the insured property or property. Gaps may be identified based on comparing insured property location data with pictorial data location data and determining a gap based on a threshold separation between nearest pictorial data location data and insured property location data. Other factors such as number of distinct photographs in a region, quality of photographs (e.g., lower quality value to lower resolution photographs or quality value depending on source of photographs), quality of geographic data (lower quality value to landmark or neighborhood based than to longitude/latitude based). Algorithms to identify gaps may include as factors a duration subsequent to a time of a most recent photograph showing an insured property; a gap may be identified if a duration subsequent to a time of a most recent photograph is above a threshold. The threshold may be a fixed threshold value stored in a memory device, or may be a variable threshold value determined based on received data or determined data values of catastrophe type (e.g., a shorter threshold time value for a tornado or derecho than for a tropical storm), geographic distance between an insured property and locations associated with weather reports, media reports and/or social media reports of damage or severe weather, which data may be accessed or received on an ongoing basis by one or more systems according to embodiments of the invention and analyzed for geographic data and text and image data indicating severe weather and property damage and/or conditions likely to result in property damage, such as flooding) and other factors. The time associated with a photograph may be obtained from metadata associated with the photograph and/or time data associated with an upload to or publication by a social media site or other site. A gap determination algorithm may also employ as factors particular sources of image data and types of image data sources. For example, a gap may be identified based on such factors as: (1) an absence of image data received directly from mobile devices; (2) image data received directly from image devices below a threshold, exemplary thresholds being a threshold number of images, or a threshold quality factor including number of images and a quality factor based on number of images depicting insured properties, quality of depiction of insured properties (e.g., percentage of image area depicting insured properties), geographic distance between images and insured properties, and other factors; (3) an absence of image data from one or more media sites, an absence of image data from one or more classes of media sites, or a quality value, determined based on factors as described above, associated with one or more media sites or classes of media sites, being below a threshold. The identification of gaps may be an example of a determination that one or more regions require data supplementation. Other determinations of data supplementation may be made based on assessments of quality of data for a region and comparing a quality score or scores to threshold scores, a quality score or scores being below a threshold indicating a requirement for data supplementation.
The method continues with accessing and/or acquiring pictorial data for the one or more gaps, or for regions and/or time periods determined to require data supplementation, step 660. The accessing and/or acquiring pictorial data for the one or more gaps or regions and time periods determined to require data supplementation may include requesting pictorial data, including requesting crowd-sourced pictorial data. Pictorial data may be requested from the insured, an insurance company representative or agent or any other third party entity that may have access to the area associated with the gap or the region determined to require data supplementation. The requesting may include sending an alert to an insured related to an insured property, issuing instructions to one or more users related to a geographical location corresponding to at least one region to one or more users, or otherwise. The alert or instructions may be communicated in any manner, including text message, e-mail, notification via one or more apps, such as one or more mobile phone based apps, notification via a social network or other resource used to collect data, or otherwise. Pictorial data or image data is then received in response to the request, such as by direct communication from recipients to the insurance company systems, and/or scraping data from social network sites associated with recipients of the request or from other resources. By way of example, a recipient of a request may pass the request to a third party who then uploads pictorial data to a different social network site from that used by the recipient of the request. In embodiments, data may be accessed or acquired without a request. For example, a search may be conducted of one or more data sources containing pictorial data, the search being conducted using search strategies directed to obtain pictorial data having associated location and/or time data corresponding to one or more identified gaps and/or one or more regions or time periods determined to require data supplementation.
The method continues with associating the received, accessed and/or acquired pictorial data, which may include requested pictorial data received and/or acquired in response to one or more requests, to the one or more gaps or regions requiring data supplementation, step 670. The associating the requested pictorial data may include overlaying the requested pictorial data on one or more of the regions, such as described above. In embodiments, the virtual map may be again evaluated for gaps or for regions requiring data supplementation, and additional requests generated for additional pictorial data. The received pictorial data may include supplemental data, such as text or voice commentary, and may be provided by an application program on a mobile device. The method continues with evaluating the virtual pictorial mapping, including pictorial data, if any, received in response to one or more requests, to issue an insurance based action instruction, step 680, or to form an insurance recommendation. Evaluation may include evaluating the catastrophe scene for damage to assess risk to insurance company personnel as well as reviewing the insured properties to determine claim advances to insured, etc. By way of example, the evaluating to assess risk to insurance company personnel may include applying photo analysis algorithms to detect fires, such as by hot spots in infrared data included in images, patterns characteristic of smoke against sky or other backgrounds, patterns characteristic of downed utility poles, downed electrical wires, trees, utility poles and other objects blocking streets, by way of example. The insurance based action instruction may include instructions to insurance company personnel to perform on site evaluation of an insured property, to exercise caution or use particular devices or equipment in a region or at or near an insured property.
Evaluation to review insured properties to determine claim advances may include comparisons of image data of insured properties after a catastrophe to earlier data to determine extent of changes indicating damage, comparisons of image data to one or more characteristic elements indicative of damage to walls, roofs and other features, by way of example. In embodiments, the system may be configured to provide insurance company personnel with displays of photos, policy information and response options such as notification of alerts as to dangers or determinations of claim advances. For example, identified images of standing water in photographs, or text data including wording such as “flooding,” may cause notification of flood danger or recommendations for use of sport utility vehicles or other high clearance vehicles. The method continues with transmitting the insurance based action instruction or insurance recommendation to an insurance entity server, step 690. The insurance based action instruction or recommendation may include an electronic communication to a claims staffing center to provide a claim advance or an assignment of claims personnel to the geographic location for adjustment or other assessment and/or a communication to an insured with insured property status data such as an alert as to the damage to their property, and/or instructions as to submission of a claim. Data to an insurance entity system may include a recommendation as to a number of claims personnel to be assigned to a region or sub-region; the recommendation may be based on algorithms or tables associating numbers of insured properties, numbers of damaged properties detected, estimated numbers of damaged insured properties in a region based on a fraction or percentage of insured properties for which pictorial data is available having damage, and numbers of claims personnel appropriate for a given number of damaged insured properties.
The peripheral device interface 712 may be an interface configured to communicate with one or more peripheral devices. The peripheral device interface 712 may operate using a technology such as Universal Serial Bus (USB), PS/2, Bluetooth, infrared, serial port, parallel port, and/or other appropriate technology. The peripheral device interface 712 may, for example, receive input data from an input device such as a keyboard, a mouse, a trackball, a touch screen, a touch pad, a stylus pad, and/or other device. Alternatively or additionally, the peripheral device interface 712 may communicate output data to a printer that is attached to the computing device 710 via the peripheral device interface 712.
The display device interface 714 may be an interface configured to communicate data to display device 724. The display device 724 may be, for example, a monitor or television display, a plasma display, a liquid crystal display (LCD), and/or a display based on a technology such as front or rear projection, light emitting diodes (LEDs), organic light-emitting diodes (OLEDs), or Digital Light Processing (DLP). The display device interface 714 may operate using technology such as Video Graphics Array (VGA), Super VGA (S-VGA), Digital Visual Interface (DVI), High-Definition Multimedia Interface (HDMI), or other appropriate technology. The display device interface 714 may communicate display data from the processor 718 to the display device 724 for display by the display device 724. As shown in
The memory device 720 of
The communication interface 722 may be, for example, a communications port, a wired transceiver, a wireless transceiver, and/or a network card. The communication interface 722 may be capable of communicating using technologies such as Ethernet, fiber optics, microwave, xDSL (Digital Subscriber Line), Wireless Local Area Network (WLAN) technology, wireless cellular technology, and/or any other appropriate technology.
An instance of the computing device 710 of
Alternatively or additionally, an instance of the computing device 710 may be configured to perform any feature or any combination of features described above as performed by the insurance data system 110. In such an instance, the memory device 720 and/or the storage device 716 may store instructions which, when executed by the processor 718, cause the processor 718 to perform any feature or any combination of features described above as performed by the interface module 112 and/or the business rules module 114. In such an instance, the processor 718 may perform the feature or combination of features in conjunction with the memory device 720, communication interface 722, peripheral device interface 712, display device interface 714, and/or storage device 716.
Alternatively or additionally, an instance of the computing device 710 may be configured to perform any feature or any combination of features described above as performed by the web site system 120. In such an instance, the memory device 720 and/or the storage device 716 may store instructions which, when executed by the processor 718, cause the processor 718 to perform any feature or any combination of features described above as performed by the web application module 122 and/or the HTTP server module 124. In such an instance, the processor 718 may perform the feature or combination of features in conjunction with the memory device 720, communication interface 722, peripheral device interface 712, display device interface 714, and/or storage device 716.
Although
Referring to
The present invention has a number of benefits including allowing for the gathering of more data more quickly from multiple sources during a time where speed of response is the most important factor. Utilizing the present invention a number of process and substantive areas are improved including: customer service and claims response; resource and capital allocation; proactive customer alerts; and information on exposure of the realized risk event to insurance company stakeholders. The present invention can automate the intelligence gathering, such that skilled/trained claims insurance personnel can focus on executing to provide customers with the best service possible during a very difficult time.
Although the methods and features described above with reference to
