Probabilistic Modelling


Simplifying exposure assessment

JBA Risk Management was delighted to be chosen as Catastrophe Risk Modeller of the Year at the 2014 London Market Awards.

Our innovative probabilistic models make it easy for re/insurers and governments to assess the potential risk they face on a return period basis.

We develop probabilistic models for meteorological perils, including sea surge, river and surface water flooding, tropical cyclone and extra-tropical windstorm.


JBA’s Great Britain flood model is the leading probabilistic model for UK flood. Based on JBA’s 5m-resolution river hazard maps, which are used by most of the UK insurance industry, the model offers the unique ability to carry out probabilistic analyses based on data that is compatible with market underwriting practice.


JBA’s bespoke development of a multi-peril tropical cyclone, sea surge and surface water model for La Réunion was the first to assess three perils.


JBA’s Thailand flood model was released within 3 months of the 2011 events, enabling insurers to immediately assess their level of exposure to this previously overlooked peril.  With flood hazard maps available globally and a global flood event set, we have the capability to quickly produce robust bespoke models in areas where concerns arise.


We provide clients with access to all our model data. Data is not encrypted so that it can be adjusted by users if desired.

Our probabilistic models are available to use in many of the mainstream cat modelling platforms, within your own cat modelling software or in our own cat model, JCalf. We license models for Great Britain, Thailand and Malaysia and also develop models for clients on a bespoke basis. If you prefer, you can license individual model components to help develop your own probabilistic model. We are also happy to provide bespoke probabilistic modelling and portfolio analysis services, either case by case or under an annual or multi-year retainer agreement.


Our proprietary event generation tool, JSheep, automates the generation of event sets for single or multi-peril probabilistic models. This allows us to quickly and efficiently develop stochastic event sets for locations all over the world, covering many natural perils including river flood, extreme rainfall, wind and hurricane.


JCalf is JBA's own catastrophe loss modelling platform.  It facilitates probabilistic loss assessment to a range of natural hazard perils, with models covering 20% of global GDP. Evolved around a flexible and intuitive design, the platform caters for re/insurance conditions empowering exposure managers, underwriters, catastrophe modellers, and risk analysts to manage their risk. JCalf handles all primary insurance terms and conditions, as well as facultative and treaty reinsurance. Users are able to report modelled results by site, policy, portfolio, coverage, line of business or any user-defined level of geography.

For more details on our probabilistic modelling and probabilistic model components, get in touch.