OASIS conference

20 Sep 2017

While the recovery after Hurricanes Harvey and Irma was still underway, approximately 80 natural perils modelling experts met in Zürich to discuss and enhance the models that will help us to better estimate and manage natural catastrophe risk in the future.

Hurricane Andrew was the start

In 1992, Hurricane Andrew left a path of destruction from the Bahamas to the Gulf Coast of the US. The damages triggered changes in building codes and hurricane risk management. It forever changed the insurance industry as well. Facing multi-billion dollar losses, the industry began to question if the outcome of hurricanes and other perils like earthquakes and floods could not be better modelled to mitigate losses. Martin Bertogg, Head of Catastrophic Perils at Swiss Re, pointed out, "It was the start of probabilistic natcat modelling". Insurers and reinsurers realised that, in order to remain in business, they needed to estimate and manage their natural hazard risk more precisely.

From proprietary to "open-source"

Since Hurricane Andrew, the industry has come a long way. Several insurers started to develop their own models and a few modelling companies were founded. As a result, catastrophe models increased in number, availability, and capability. Today's models cover hurricanes as well as windstorms, earthquakes, volcanoes, floods and storm surges.

For a long time, this natural catastrophe modelling market was dominated by a few large modelling companies. Increased license cost pressure on the risk taker side and the regulatory requirement that insurance companies build their own view of risk were the main motivation to found Oasis.

"Oasis opens the floor for smaller model vendors to enter the cat modelling market place. This will lead to a diversity of risk models available. This diversity will in turn foster the building of the own view of risk. The collaboration aspect ("bundling of forces") of Oasis will reduce inefficiencies to build a risk view and reduce the cost of natcat model licenses going forward," said Beat Aeberhardt, Head NatCat Tools at Swiss Re and member of the Oasis board, at the start of the meeting.

Still, in a world of rapidly developing cities/markets in regions at risk for natural hazards, insurers began to understand that with their resources alone, they could not keep up with all the opportunities that new technological developments offer. "We noticed that to be innovative, we need to collaborate and share with all stakeholders", said Paul Nunn, Head of Catastrophe Risk Modelling at SCOR. It was time to open up. This was the birth of Oasis.

Crowdsourcing new ideas by going "Open Source"

Moving from purely proprietary to "open source" – especially for the data – created a new ecosystem to improve risk assessments. "Any stakeholder can use the platform for free, share his applications and models with others, and thereby help to improve the resilience for many in all regions of the globe", noted Tracy Irvine, Managing Director of the Oasis Hub. If you need high resolution flood footprints after an event like Hurricane Irma, or a volcano dataset, it's all only one click away in a format that can be downloaded, used, modified and shared.

Marc Wüest, Expert for Atmospheric Perils at Swiss Re, demonstrated how seamless it is to install Oasis and how the OASIS platform can be integrated into Swiss Re's in-house natcat risk assessment platform "CatServer". "It's plug and play for model developers", he said.

Martin Bertogg, Head Catastrophic Perils, Swiss Re

The open source world of OASIS

Because it is accessible to all, the range of innovative solutions has been increasing rapidly since the founding of OASIS. A recent example of combining new tech with cat models is the integration of data collected by drones. User numbers vastly expanded too. Tracy Irvine referred to institutions such as the Potsdam Institute for Climate as well as development aid agencies as clients. And it does not stop there.

Collaborate, develop, exchange, repeat

One example is the OASIS Platform for ASIA Risk Assessment, which was presented by Claire Souch, Director at AWHA Consulting. In one part of this project, the governments of Germany and the Philippines join forces with academia and insurance companies to develop a first flood model for the country. It will be the foundation for the implementation of the national disaster risk financing and risk reduction strategies, which is important in a country where eight of the ten costliest typhoons in history have occurred in the last ten years. The OASIS framework is used as it defines a common language for all relevant model components (Hazard, Exposure or Vulnerability) to build models that can be used in the area of risk management and mitigation.

Dickie Whitaker, Chief Executive, Oasis LMF

Taking OASIS to the next level

The example of the Philippines shows how big the need and the potential of OASIS is. Making our world more resilient against the threats of natural perils will require collaboration, exchange and openness so that innovation can easily happen.

That the necessity is there can be seen in the number of modelling providers committed to the Oasis Platform and the steep increase in the number of models that are available in the Oasis format as of today.

"Many want to help make our world fit for the inevitable events to come. OASIS gives everyone the tools to simulate the future, develop their own solution or look for an existing one to a given problem out there. Developing further a broad ecosystem around our open source platform is the key to unlock this potential in the years ahead," said Dickie Whitaker, Chief Executive of OASIS.

Summary of the Swiss Re Institute's OASIS conference in September 2017.
See also the blog "Cat modelling uncertainty and transparency - needs you!", written by Claire Souch, on Open Minds.

We use cookies to gather information that will help us provide the best possible service. By using this site, you are accepting our cookie policy.

In our ongoing efforts to improve the quality and relevance of our publications, we would like to know more about you.

* required fields

Interested in subscribing to our content? Visit our subscription page.