Cloud Pak for Data

Reducing Severe Weather Risk 

Fri December 04, 2020 12:08 PM

About This Accelerator

Severe weather has a significant impact on insurers and their policyholders. This industry accelerator simulates a point in history where there were billions of dollars of damage caused by fire, wind and hail 2'' in size within the south western part of the United States. This accelerator will show how data engineers, data stewards, data quality analysts, data scientists, citizen data scientists, analysts, and developers can all focus on their role to reduce severe weather risk and work together at the same time using Cloud Pak For Data as the ModelOps Platform.
  • Collect, ingest and merge historical severe weather (hail, wind, and fire) data, coming from The Weather Company Data Packages, with historical policyholder claim data to understand what the weather was in the past when a claimed occurred. The same data will be used downstream to build, test and deploy claim damage prediction models and data built by data scientists and analysts.
  • Organize your analytic artifacts such as data, notebooks, models, data source connections and dashboards making it easy for others to find and collaborate through governed trusted catalogs.
  • Analyze as a data scientist or analyst, to build, test and deploy a model to predict claim damage or develop analytic dashboards to show many views of severe claims damage.
  • Infuse model results into downstream dashboards, online/batch scoring and other applications such as ESRI maps.
Cloud Pak For Data ModelOps Platform


Required services: To use the industry accelerators, you must install one or more of the following services on IBM® Cloud Pak for Data

Service Required for
Watson Knowledge Catalog Importing data governance artifacts, such as business terms and categories. See Installing Watson Knowledge Catalog.
Watson Studio Importing data science assets to an analytics project. See Installing Watson Studio.
Watson Machine Learning Deploying analytical models. See Installing Watson Machine Learning.
SPSS Modeler Customizing the predictive model. See Installing SPSS Modeler.
Analytics Dashboard Running the dashboard. See Installing Analytics Dashboards.

Data Assets Included With The Industry Accelerator

CSV files
  • CustomerClaims.csv The actual claim amount by customer.
  • CustomerLocations.csv The location of the claim.
  • CustomerReportedHail.csv The hail reported by customer.
  • ClaimsLocationHail.csv Merged file containing claims, location, and hail.
  • SPSSTreesModelPredicitons.csv Output from SPSS Modeler to be analyzed in Analytic Dashboard.
Analytics Dashboard
  • Reducing Severe Weather Risk Dashboard.json Create a new dashboard using this file.
SPSS Modeler Stream 18.2.1
  • Insurance Hail Risk SPSS Desktop Model.str
    • Use With IBM SPSS Modeler 18.2.1 and/or CP4D Modeler Flow
    • Download and try SPSS Modeler Desktop here or contact

Importing the accelerator

To use this accelerator on Cloud Pak for Data v3.5.0.0, complete the following steps:

  1. Download the reducing-severe-weather-risk-in-insurance-industry-accelerator.tar.gz file, which is available on the repository.
  2. Determine how you want to import the accelerator:
    • If your Cloud Pak for Data user account has Admin privileges, all components can be installed by following these steps:
      1. From a command prompt, run the following command to extract the contents of the package:

        tar -xvf {TARFILENAME}

      2. Run the following command to import the accelerator content into Cloud Pak for Data. When prompted, enter the required information.


        For a list of all available options, including how to pass arguments to the import script, enter the command:

        ./ --help

    • If you would like to pick which components you import, complete the following steps:
      1. Extract the contents of the package.
      2. Import the governance artifacts to Cloud Pak for Data:
        1. Log into the Cloud Pak for Data web client as a user that has the Manage categories permission and the Author governance artifacts permission.
        2. Import the CSV file that defines the categories. The file name should have the following format: accelerator-name-glossary-categories.csv.

          See Managing categories.

        3. Import the CSV file that defines the business terms. The file name should have the following format: accelerator-name-glossary-terms.csv.

          See Managing business terms for the business glossary.

      3. Import the ZIP file that defines the analytics project to Cloud Pak for Data. The file name should have the following format:

        See Importing a project.

Please view README in project data assets to see the full step by step to import and details regarding project artifacts.

Release Notes

This accelerator has been verified on:

  • IBM Cloud Pak for Data 3.5.0

About the developer:

Brian is a Financial Services Technical Seller for IBM Cloud Pak For Data @IBM helping customers modernize their data science platform to build, test, and deploy data science models, analysis, predictions dashboards, and other analytic artifacts to provide personalized ads to reduce churn, solutions to identify fraud, models to reduce financial risk, predict and forecast liquidity, reduce severe weather insurance risk and many other use cases in Insurance, Banking and Financial Services.

Brian develops and delivers technical solutions to win IBM sales opportunities. He gathers requirements, recommends product architecture, builds out recommended solutions while making sure to showcase value and return on investment.


This project contains Sample Materials, provided under license.
Licensed Materials - Property of IBM.
© Copyright IBM Corp. 2020. All Rights Reserved.
US Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.



0 Favorited
0 Files