IBM Software Information Management

IBM Insurance Information Warehouse

Support for Catastrophe Modeling

Whitepaper
2 IBM IIW Support for Catastrophe Modeling

Introduction Business Context

Catastrophe modeling is the process of using computer-assisted Catastrophe models have been in widespread use for many
calculations to estimate the losses that might be sustained decades. They arose out of the severe financial impact on the
due to a catastrophic event such as a hurricane or earthquake. insurance industry from the aftermath of catastrophic events.
Catastrophe modeling is especially applicable to analyzing For example, Hurricane Andrew in 1992 led to the insolvencies
risk and is at the confluence of actuarial science, engineering, of insurers and reinsurers. The impact of these catastrophic
meteorology, and seismology. The process involves the modeling events has highlighted the need to better understand, manage
of catastrophic events in specified geographic locations in order and control the level of risk associated with them. Catastrophe
to analyze the exposure of insurers and reinsurers to these events. models have evolved over that period to become a part of how
Insurers and reinsurers use catastrophe models to help calculate insurance organizations view risk from underwriting, pricing,
the potential amount of capital needed to discharge claims reserving and reinsurance perspectives.
associated with catastrophic events.
The fundamental issue is that given the relatively recent
Catastrophe models calculate the exposure of (re)insurers over development of catastrophe models and the lack of enough
a prolonged period and not just over the course of a single year. historical catastrophe data, combined with changing climate
Catastrophe models help (re)insurers to identify, understand conditions (rising sea levels and changes in weather patterns) and
and manage the fluctuation in the number and severity of claims evolving building codes and regulations, it is a complex operation
associated with large catastrophic loss events. An assessment of to assess what the impact of these events are. A major catastrophe
exposure to risk should not only consider the incidence of claim event such as a hurricane can have an altogether different loss
occurrences and the level of losses, but also be informed by a outcome today than previously not only based on its trajectory
better understanding of all factors leading to claims including the of the hurricane but also due to changes in construction
location of insured risks and the characteristics of these risks also. standards and methods in addition to societal changes (level of
insured risks, quality and compliance with building codes). A
This understanding of risk exposure can inform and validate more holistic approach is required encompassing not only an
risk pricing, support the better risk management of exposures understanding of the hazard events themselves (where and how
in terms of risk concentration and also provide new insight in they occur) but also an understanding of the vulnerability of
support of reinsurance and risk transfer decisions. different types of exposures to these events.

When viewing catastrophe losses over time, it can be seen that

the level of losses in recent times is greater, possibly due to the
increasing value of assets, and the level of insurance penetration
and changing demographics. What is more significant is that
peak loss years are having a much higher financial impact and are
also happening more frequently.
 Information Management 3

Severe catastrophic events such as the earthquake and tsunami describing elements of risk so different policies, products and
in 2011 that devastated north-eastern Japan have heightened risks can be understood and compared in a similar way.
the interest in catastrophe modeling and raised concerns about
the accuracy/validity of the models. Historically, different IBM Insurance Information Warehouse (IIW) provides broad
catastrophe model providers have evolved different approaches coverage in support of catastrophe modeling and reinsurance.
and competencies. Organizations have depended on these IIW catastrophe modeling coverage aims to help insurers and
models to accurately project their level of risk. reinsurers understand how to capture the data inputs into,
and outputs (results) from, catastrophe loss models in a data
However, the variances in their own actual claims experience warehouse. This involves the identification of the parts of the
versus the projections provided by catastrophe model providers model that are relevant to this business requirement from
has led many to consider capturing a more unified view of reinsurance, policy and exposure data, claims data, hazard
risk. In conjunction with this changing industry viewpoint, footprints and event data, location information and also to
rating agencies and regulatory changes, such as Solvency II, are identify how the results can be captured.
mandating greater ownership of results from catastrophe models
by insurers and reinsurers. The comprehensive coverage in IIW can help catastrophe
modelling projects define products, pricing and risk information
Catastrophe modelling projects can suffer when ingesting data in a standardised way, with well defined and flexible data
from a multitude of different source systems, all defining and structures, which can capture data from the simplest of risk
describing risk exposures, products and risk pricing in variety of characteristics to complex reinsurance treaties.
different ways. There is a need for a standardized approach to

Figure 1. Natural catastrophe losses 1980-2010

4 IBM IIW Support for Catastrophe Modeling

Benefits of Catastrophe Modeling with IIW

Data Standardization Catastrophe Modeling

• Use IIW to apply a standardized approach to the capture • Model Integration - IIW supports one version of the data
and definition of products, coverages, policies, reinsurance and can provide support for the inputs to, and outputs
treaties, different types of policy limits and excesses etc. from, catastrophe loss models.
This can support a single version and understanding of the • Loss Model Validation - IIW data coverage for risk
inputs to the catastrophe models. assessments can be used to define a common method
for understanding the results for multiple catastrophe
Insight models and in so doing promote greater transparency and
• Risk Concentration/Portfolios – IIW supports improved confidence in these results.
risk analysis at the lowest granularity, and by geolocation.
The Analytical Requirements can help insurers Reporting
understand the risk related to where they are located (risk • Capital Requirements - Comprehensive support for
concentration and accumulation) and risks associated with Solvency Capital Requirements in particular to relation to
the type of risk exposures covered (portfolio analysis). Solvency II.
• Real-Time Analysis – New data sources and real-time data • Claims versus loss projections - IIW includes coverage for
such as flood sensor data can be integrated in IIW. This the capture of catastrophe model results (projected losses
allows for greater understanding of events as they occur for extreme event scenarios, and loss projections such
using analytical applications. as average annual losses) which, when considered with
claims, can support greater understanding of the variances
Decision Support between actual and predicted losses for multiple scenarios.
• Reinsurance/Risk Mitigation policy - Understand the • Visualizations – Data visualization tools, such as ESRI
exposure portfolio and match with the appropriate and IBM Cognos Business Intelligence, can be used with
reinsurance treaty using IIW improved support for IIW-based data warehouses/data marts to overlay data
Reinsurance. points (location of properties/claims), shapes (flood zones,
• Risk Pricing – an IIW-based risk exposure data warehouse elevation maps, flood scenarios) and lines (roads, hurricane
can help provide insight into how insurance products are trajectories) with other maps to give them context and
costed (coverages, terms and conditions). provide new insight.
 Information Management 5

Geospatial Support

IIW data models deployed on geospatial enabled database technology can provide a whole new understanding of the role that location
plays in risk exposures:

• Map data can be loaded into an IIW-based data warehouse for flood or hurricane zones. Geospatial functions can be used to
determine whether properties are located within specified zones.
• IIW supports geospatial data constructs such as area polygon, line segment, and point, enabling greater accuracy in capturing
location data and event data. It is aligned with geospatial libraries, such as IBM PureData Spatial ESRI Package (Netezza) and
DB2 Spatial Extender, enabling easy deployment and data manipulation.

IIW is designed to capture the data at the location resolution required.

Figure 2. This sample report is based on a sample data set for

Hurricane Katrina and includes the capture of the hurricane track
and also information about claims and policies exposures.

Figure 3. This sample report is based on a sample data set for a

simulated flood scenario and highlights the number of claims and
type of properties affected.
6 IBM IIW Support for Catastrophe Modeling

New Approach to Catastrophe Modeling

Catastrophe modeling providers are responding to the changing This new approach should focus on consolidating and
needs of the insurance industry by opening up their modeling standardizing the data sources that feed catastrophe models.
platforms and allowing greater scrutiny. New standards, such as Having more accurate data can help calibrate loss model data
Oasis Loss Modeling Framework, are emerging which advocate inputs, and can better reflect the characteristics of the risks,
alternative open and transparent methods of calculating levels of provide for better predictions, and facilitate new insights,
risk and uncertainty. including real-time analysis, which might help mitigate against
losses before they occur. The critical component needed to
New data sources which can increase our understanding of operate successfully in this changing landscape is an exposure
risks are also available, for example detailed maps developed by data warehouse which provides a consolidated view of risk,
specialist model providers (including high resolution hazard foot considering not only the inputs into catastrophe models both
prints), real-time sensor data collected for floods, windstorms also relating the catastrophe model results with those inputs..
etc., and a growing number of publically available data sources
on the intensity of events for example, average wind speed IIW provides business content and comprehensive data
and trajectory of a hurricane from National Oceanic and structures to support catastrophe modeling initiatives and helps
Atmospheric Administration (NOAA). to accelerate the scope definition, data model design and physical
deployment of an exposure data warehouse.
This data can be used to harness the inherent insight available in
claims and underwriting experience to derive new understanding IIW can help catastrophe modeling projects deliver value and
of the vulnerability of risk exposures to these catastrophic identify new patterns in exposure data. IIW’s coverage can
events. This changing landscape, combined with continuing support the development of a risk exposure data repository
improvements in geospatial technologies for the definition and around the key data components in the lifecycle of catastrophe
visualization of geospatial data, highlights the need for a fresh modeling.
look at how catastrophe modeling is supported in insurance
businesses and to identify opportunities to transform and extract
new value and insight from catastrophe modeling.

Figure 4. Approach to catastrophe modeling

 Information Management 7

Inputs Hazards/Events - IIW supports different types of intensities

Policy Terms - Capture different types of policy terms (for (for example, wind speed, rainfall, richer scale) and scores for
example, limits/sub limits/deductibles/reinsurer share) regardless catastrophe event footprints from the trajectory of hurricanes to
of level of complexity of insurance or associated reinsurance the damage path of a flood. These events can be defined in time
contracts. This flexible coverage can support the appropriate and with different location designations such as GPS points,
allocation of policy financials to ground up losses which are shapes (lines for example, hurricane trajectories), or polygons
needed to determine the true financial loss for the exposures on for example, flood zones. It might not be necessary to capture
risk. IIW can help with moving from the difficulties of trying to all this data within the data warehouse but IIW is designed to
define types of products and policies from different systems to capture the data at the different levels of detail as required. This
a standard definition using one model and one set of business can include place assessments/scores, readings for catastrophe-
terminology. related events, and event categorisation, which are critical to
defining the level of intensity of these events and that can be
Risk exposures - IIWs rich support and detailed definitions for related to claims experience.
insured risks, including property and location characteristics,
allows for a more detailed definition of risk characteristics. Claims and Vulnerability - The data coverage for catastrophe
Capturing this data is an opportunity to harness additional events, such as floods or hurricanes, can be used in combination
policy/proposal information available in source systems or with IIW’s claims coverage to derive new statistical insight from
hidden within policy documentation (unstructured data). This an insurers’ own claims experience, bridging the relationship
can then be included in loss modeling scenarios (for example, between flood levels or wind speeds and the level of claims.
understanding how elevation of a property within a high risk External data sources such as damage data for different types
flood zone can impact claims.) of building construction or specialist flood models with high-
resolution maps can also be loaded into IIW statistical data
coverage.

Figure 5. Approach to catastrophe modeling - detail

8 IBM IIW Support for Catastrophe Modeling

Outputs - Risk Assessments - IIW data models provide for the Analytical Requirements
capture of risk assessment data requirements including both the IIW contains an extensive list of Analytical Requirements,
inputs and outputs. reflecting the most common queries and analyses for business
performance measurement and reporting, while supporting
Analysis - IIW provides coverage for catastrophe modeling other analytical functions such as ad hoc reporting, data mining
reporting requirements. This coverge connects the inputs to, and and decision support.
outputs from, the catastrophe model which can support greater
insights into the loss model results at the required granularity. Analytical Requirements enable rapid scoping and prototyping
of data marts, which provide a subject-specific analytical layer in
a data warehouse solution. Analysts and business users can use
IIW Support for Catastrophe Modeling Analytical Requirements to gather the reporting and analysis
requirements of their organization quickly.
IIW is a set of models that enables insurance organizations
to build and deliver a business-oriented, enterprise-class data Each Analytical Requirement can be divided into measures,
warehouse or data repository. IIW comprises the following which are numerical facts that convey quantitative information of
components: importance to the organization, and dimensions that categorize
measures. These measures and dimensions are mapped back
Project Views to the data warehouse, so that the scoping of the reporting
Project Views are business-subject-area views that span across all and analysis requirements automatically selects the most
IIW components with the following functions: appropriate data warehouse entities and attributes to support
those requirements. The analytics development team can use
• Provide predefined business scope of business terms related Analytical Requirements to create designs for specific data marts
to the various components of the catastrophe modeling or dimensional solutions that can serve as the source for a range
from a data perspective (catastrophe model inputs and of reports and charts. Data Definition Language (DDL), for the
results, hazards, exposure data, policy terms including required physical database implementation, can be generated to
reinsurance and so on) accelerate reporting development. IIW supports the automatic
• Provide pointers to the parts of the IIW models which generation of DDL into IBM PureData System for Analytics
would be relevant to a catastrophe model project data (Netezza.)
scope or to address a particular exposure analysis reporting
requirement. Analytical Requirements for catastrophe modeling include event
• Can be extended as required with an organizations specific analysis (for hurricanes and floods), loss modeling (leveraging
customizations. insight from catastrophe model outputs) and reinsurance
analysis.
 Information Management 9

Business Terms Business Data Model

The Business Term glossary enables non-technical business The Business Data Model is a logical model that represents the
experts to describe and define, in their own words, the concepts essential entities and relationships of the insurance industry. It
they use every day. Clearly defined business terms help provides a business view that excludes technical implementation
standardization and communication within an organization. considerations such as details related to any specific database.
Mappings to the data models make it possible to create a
common, enterprise-wide picture of the data requirements and The Business Data Model is the first point at which the various
to transform these requirements into IT data structures. business requirements are brought together and modeled
in an entity-relationship format. It enables organizations to
As part of IIW, the glossary is a comprehensive list of terms perform the initial modeling of their business requirements
pertaining to insurance, financial services and general business and helps the organization understand the various constraints,
that includes: relationships and structures that can be implied in their business
require¬ments. This is the essential model of the business,
• Definitions written in plain business language providing the overall business context and a common basis
• Detailed data elements that specify what each business for the downstream models that can be used in the actual
term means for the insurance organization. deployment of the physical data warehouse. The information
• Terms that might be related to one another through reflected in the data model is independent of organizational
relationships. structure and has been validated by multiple sources within the
industry.
Significant extensions have been made to the Business Terms
coverage not only for catastrophe modeling but also for Atomic Warehouse Model
reinsurance. The Atomic Warehouse Model is a logical, specialized model
derived from the Business Data Model. It is optimized as a data
Business Example Diagrams repository that can hold long-term history, usually across the
IIW also includes business view diagrams: entire enterprise. The Atomic Warehouse Model provides the
data design support needed to create a uniform model of the
• Overview diagrams provide an explanation of where in enterprise-level business requirements defined by the Business
the model the coverage for catastrophe model data can be Data Model into specific, flexible and efficient structures
found using IIW business terminology. dedicated to the long-term storage of historical facts.
• Entity/Relationship scenario diagrams provide an initial
scoped model diagram including entities and relationships
to provide an initial acceleration point for catastrophe
modeling, risk exposure or reinsurance data warehouse
project, which can be customized as needed and from
which a physical model can be generated and deployed.
10 IBM IIW Support for Catastrophe Modeling

The Atomic Warehouse Model features a flexible atomic This repository holds sufficient and complete data to meet the
data area (primary data storage area) as well as the typical needs of business user analysis. Dimensional models are more
summaries needed by most insurers to roll up the detail data for easily understood by business users. They are optimized for
analysis purposes. A portion of the Atomic Warehouse Model data querying rather than for transactional speed, and their
is generated in the initial project phase. Other areas can be structure makes it is easier to extend them to support new data
generated as the insurer covers more business areas over time. requirements. New queries can be created without having to
The Atomic Warehouse Models contains data structures redesign the data structures, while old queries can still operate
needed by an insurance organization to support a wide variety without change.
of business requirements such as regulatory requirements
(Solvency II reporting requirements), customer insight, claims, The Dimensional Warehouse Model contains star schema style
fraud, life insurance, investment management, general insurance, dimensional data structures organized around fact entities that
reinsurance and catastrophe modeling. support the Analytical Re¬quirements.

Dimensional Warehouse Model

The Dimensional Warehouse Model is a logical model derived
from the Business Data Model and the Analytical Requirements
and provides an optimized data repository for supporting
analytical queries. The Dimensional Warehouse Model provides
the data design support needed to transform the enterprise-
level business requirements in the Business Data Model into
busi¬ness-specific and efficient structures dedicated to the design
of a dimensional data repository.
 Information Management 11

Features & Benefits

Business Terms Logical Model Design

• Significant expansion of coverage of catastrophe and • Atomic and Dimensional Warehouse improvements to
reinsurance-related business terminology support the capture of catastrophe events including hazard
• Terms now available in IBM InfoSphere Information footprints, geospatial location definitions, policy and
Governance Catalog reinsurance terms, claims and catastrophe risk assessments.
• Business Example diagrams for the catastrophe model
Analytical Requirements scenario which can help accelerate project scoping and
• Illustrated Analytical Requirements mapped through requirements definition.
to logical design to accelerate and provide inspiration • Models available in InfoSphere Data Architect can be
for the definition and deployment of catastrophe model readily deployed on numerous database options including
projects. Now available in IBM InfoSphere Information IBMPureData for Analytics (Netezza), DB2 and Hadoop
Governance Catalog. etc.

1 3 5 7
New insight on Results stored for
vulnerability to analysis to determine
John wants to analyze a Flood and hurricane catastrophes can be derived if too exposed to floods
portfolio of risk exposures model data including by relating historical in one region and to
to project losses for geospatial location policy/claims to the buildings of timber
hurricanes and floods. are saved. events that construction for
impact them. hurricanes.

2 4 6
Risk exposure and
Policy, exposure, hazard locations are Inputs collected in
premium and claims related to one another, previous steps can be
data stored for risk e.g. property in a flood used to feed catastrophe
exposures in portfolio zone, proximity to models and to better
are identified hurricance path. understand the
results.

Figure 6. Approach to catastrophe modeling - scenario

12 IBM IIW Support for Catastrophe Modeling

Hurricane Event Analysis Use Case

The amount and severity of claim incidents resulting from severe IIW business content and logical model coverage supports this
windstorm and hurricane events are increasing and is of growing need. IIW helps combine policy data, exposure data (such as
concern to insurers and reinsurers. Changing weather patterns, structure, claims, premium) together with new data sources for
changes in demographics (where people live) and changes in hurricane events (historical hurricane tracks, hurricane models)
the type of construction mean that insurers cannot rely on past and the loading of this into the data warehouse to improve
historical claims as a predictor of future losses. Insurers and consolidate insight and analytics.
reinsurers need to better understand which types of structures
(construction/material) have greater propensity to wind damage IIW includes data coverage to feed all aspects of hurricane-
and the structure profile of their exposures. New approaches related catastrophe loss modeling both in terms of inputs to
need to be considered and new data sources used to improve the and outputs from these models. Using IIW as the basis of a
understanding of hurricane risk. risk exposure data warehouse allows for this consolidation of
insight into one data store to support the different views of risk
exposures needed which might inform the types of buildings to
provide hurricane cover for.

Figure 7. Hurricane event analysis use case

 Information Management 13

IIW supports geospatial coverage, including defining the IIW provides support for the capture of industry damage
location of exposures relative to the trajectory of hurricanes, both databases, which provide a profile of wind damage intensity.
in terms of previous actual hurricanes such as Hurricane Katrina Separately, IIW provides the ability to collate historical hurricane
and simulated hurricane events. Using geospatial tools and tracks and wind damage claims allowing them to be related to
technology, risk exposures can be defined within the boundaries each other.
of hurricane zones or alternative zones created for example,
hurricane zones, which consider not only the risk propensity of With this consolidated view, a new insight and picture emerges
a hurricane but also the type of building construction prevalent for example, an understanding of how the intensity of the
in that zone. This multiple view of location can be captured in hurricane at particular locations relates to claims might indicate
IIW. IIW hazard coverage supports the capture of footprints that particular types of housing construction are more resilient to
for previous historical catastrophes, such as Hurricane Katrina, hurricanes.
within the data warehouse.
Using this new insight, catastrophe loss projections can be
Using geospatial libraries provided, such as those available with calibrated with risk exposure information and assumptions
IBM PureData for Analytics (Netezza, for example Spatial ESRI more suited of the true risk profiles of the underlying risks. The
Package), geospatial queries can be performed to determine the outcome from this lower-level analysis might lead to changes
intersections between the damage paths of hurricanes and the in policy conditions for example, exclusions; endorsements or
location of risk exposures. This common IIW-based definition pricing, which more accurately reflect the risk characteristics
of location across exposures, claims and hazards can shed new leading to claims. It might also lead to more accurate and
light on the contributing factors for hurricane claims and their discrete catastrophe risk assessments and inform a more
relationship to location. The key elements of wind catastrophic measured policy to mitigate against different risk profiles.
events, including wind speed, atmospheric pressure, the declared
categorizations of these events (for example, Tropical Storm IIW provides a number of supports for hurricane event analysis
category 5), and their various locations at a point in time, can be from Business Term and Analytical Requirements coverage
stored with IIW. which aid requirements definition; to logical data models, which
can capture the data needed and dimensional model coverage,
Claims for different types of buildings can be related to hurricane which provides support across risk exposures, wind events, claims
events to understand how different levels of wind speed or gusts and catastrophe model results.
impact wind damage claims for different types of buildings. IIW
can then be used to feed a catastrophe loss model to project
average annual losses over time, which can then be related to
actual premium collected. This comparison might identify that
certain types of buildings should only be covered under certain
conditions and might lead to a different view of risk pricing
associated with buildings in hurricane zones but which have a
rigid construction which has withstood previous hurricanes in
the past.
14 IBM IIW Support for Catastrophe Modeling

Flood Event Analysis Use Case

Flood damage is a growing problem for insurers and reinsurers. of data for risk exposures, flood events, claims and catastrophe
Climate changes appear to indicate an increase in the severity model results.
and incidence of extreme weather events. This level of
unpredictability requires insurers and reinsurers to reassess flood By using IIW as the basis of a risk exposure data warehouse,
risk areas to take into account these changing weather patterns all key information can be consolidated into one data store to
and to better reflect the exact location of a property within a support the different views of flood risk exposures needed for
flood zone. flood risk concentration analysis.

IIW business content and logical model coverage supports IIW facilitates the combination of policy data, exposure data
this need. IIW provides a number of supports for flood event such as structure data, claims data, and premium data with new
analysis such as Business Terms and Analytical Requirements data sources for flood events. These new flood data sources
coverage which aid requirements definition for flood risk (for example, flood sensor data, flood models, which may be
concentration analysis. IIW data models can capture the data in the form of maps) can be combined with reports to provide
needed for the analysis of floods while dimensional model visualization using tools such as Cognos and/or can be loaded
coverage provides assistance for flood event analysis in the form into a data warehouse to consolidate insight.

Figure 8. Flood event analysis use case

 Information Management 15

IIW includes data coverage to feed all aspects of flood related Claims by location can be related to flood events to understand
catastrophe loss modeling both in terms of inputs to and outputs how flood levels, rain fall and water flow impact on the level of
from these models. Using IIW as the basis of a risk exposure data claims. IIW can further be used to feed a catastrophe loss model
warehouse allows for this consolidation of insight into one data to project losses in the event of a catastrophic flood scenario.
store supporting the different views of risk exposures needed for Understanding this impact might change an insurer’s perspective
risk concentration. on the level of risk retained and might indicate that no more
business should be written within a given rated/defined flood
By using geospatial libraries and database functionality, it is zone.
possible to determine where the locations of properties are
relative to different location definitions for example, are they
within a high risk flood zone. It is also possible to derive new
definitions of zones, which may be used to feed catastrophe
models for example by determining a particular neighborhood
might be zoned together given an insurer’s high risk exposure
concentration, a common type of building structure and the
zone’s propensity for flooding.
16 IBM IIW Support for Catastrophe Modeling

Deployment

IIW can be used to develop a risk exposure data warehouse IIW helps capture data inputs for catastrophe models such
and acquisition layer for proprietary catastrophe models or as reinsurance, policy, exposure, and claims data. It provides
alternatively for open source and in-house models. support in the following areas:
IIW can be used to define the data input and output
requirements for catastrophe models. Event analysis
• Flood zone risk concentration analysis
This can be readily deployed on IBM PureData for Analytics • Flood risk portfolio analysis
(Netezza) using InfoSphere Data Architect. Visualisation tools • Hurricane location analysis
such as ESRI ARCGIS in conjunction with reporting tools such • Hurricane building structure analysis
as Cognos can be leveraged to provide new geospatial insight on
risk exposures and catastrophe events. Loss modeling
• Annual exceedance probability analysis
Conclusion • Loss distribution and vulnerability analysis
• Occurrence exceedance probability analysis
IIW can help insurers and reinsurers address the technical
challenges of catastrophe modeling by supporting the Reinsurance analysis
deployment of a logical data warehouse architecture that • Finite reinsurance decision analysis
encompasses both new and traditional data structures and
deploys across multiple technology platforms.

Proprietary Catastrophe
Models

Data Sources

Data Acqusition Analytical Data Mart

Historical
IBM Cognos
Policy
Exposure

Reinsurance Open Source / In House IBM IIW

IBM IIW
Catastrophe Models
Disaster

IBM PureData / Netezza IBM PureData / Netezza

with in-Database Libraries with in-Database Libraries
or or
IBM DB2 with IBM DB2 with
Spatial Extender Spatial Extender

IBM PureData / Netezza IBM Cognos with

ESRI Maps

Figure 9. Sample deployment environment using IBM Software

© Copyright IBM Corporation 2014

IBM Corporation
Software Group
Route 100
Somers, NY 10589

Produced in the United States of America

November 2014

IBM, InfoSphere, Streams, DataStage, BigInsights, DB2, PureData, Cognos,

SPSS and the IBM logo are trademarks of International Business Machines
Corporation in the United States, other countries, or both.

Other company, product or service names may be trademarks or service marks

of others. A current list of IBM trademarks is available on the web at “Copy-
right and trademark information” at. ibm.com/legal/copytrade/shtml.

This document is current as of the initial date of publication and may be

changed by IBM at any time. Not all offerings are available in every country in
which IBM operates.

THE INFORMATION IN THIS DOCUMENT IS PROVDED “AS-IS”

WITHOUT ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING
WITHOUT ANY WARRANTIES OF MERCHANTABILITY, FITNESS
FOR A PARTICULAR PURPOSE AND ANY WARRANTY OR CONDI-
TION OF NON-INFRINGEMENT. IBM products are warranted according
to the terms and conditions of the agreements under which they are provided.

The client is responsible for ensuring compliance with laws and regulations
applicable to it. IBM does not provide legal advice or represent or warrant its
services or products will ensure that the client is in compliance with any law or
regulation.

Please Recycle

IMW14803-USEN-00