City University London
An empirical study of the performance benefits of spatial clustering analysis in the application layer of an insurance risk management platform.
Panos Bairaktaris
January 2012
Supervisor Dr Jo Wood
Abstract
Decision-making within the insurance sector is strongly influenced by data containing a spatial dimension. When understanding and managing the risks arising from natural catastrophes, spatial analysis often forms a core part of the overall analysis of both the insured asset (exposure) and the natural hazards. The design problem facing such spatial decision-making tools is how to perform very fast spatial analysis across very large and complex datasets over wide spatial extents. In the case of exposure data, a large global insurer might have a portfolio containing over a million policies spread across the globe. This project researches ways to visualise and interrogate these large datasets in a web mapping, corporate decision-making analytical platform, using modern GIS tools and the latest Microsoft .NET framework. The project will investigate the performance benefits of a spatial clustering process in the application layer rather than in the presentation layer. By identifying the underlying technical challenges, the proposed solution incorporates the latest technological advances and APIs in an orchestrated and well defined manner. Aside of increasing the analytical capabilities of the hosting insurance platform, the proposed solution also provides extensibility key points on which further development can expand and build upon, with a view to provide a spatial analysis clustering service layer that performs well under demanding usage scenarios, as well as being flexible for future extensions.
Decision-making within the insurance sector is strongly influenced by data containing a spatial dimension. When understanding and managing the risks arising from natural catastrophes, spatial analysis often forms a core part of the overall analysis of both the insured asset (exposure) and the natural hazards. The design problem facing such spatial decision-making tools is how to perform very fast spatial analysis across very large and complex datasets over wide spatial extents. In the case of exposure data, a large global insurer might have a portfolio containing over a million policies spread across the globe. This project researches ways to visualise and interrogate these large datasets in a web mapping, corporate decision-making analytical platform, using modern GIS tools and the latest Microsoft .NET framework. The project will investigate the performance benefits of a spatial clustering process in the application layer rather than in the presentation layer. By identifying the underlying technical challenges, the proposed solution incorporates the latest technological advances and APIs in an orchestrated and well defined manner. Aside of increasing the analytical capabilities of the hosting insurance platform, the proposed solution also provides extensibility key points on which further development can expand and build upon, with a view to provide a spatial analysis clustering service layer that performs well under demanding usage scenarios, as well as being flexible for future extensions.
The document:
Interesting blog post.This blog shows that you have a great future as a content writer.waiting for more updates...
ReplyDeleteDigital Marketing Company in India
It is really very excellent,I find all articles was amazing.Awesome way to get exert tips from everyone,not only i like that post all peoples like that post.Because of all given information was wonderful and it's very helpful for me.
ReplyDeleteSAP Training in Chennai
SAP ABAP Training in Chennai
SAP FICO Training in Chennai
SAP MM Training in Chennai