Traditional methods of assessing fire risk are based on probabilistic treatment of fire incident data. Recent advances in the ability to make deterministic predictions of the consequences of specific fire scenarios present an opportunity to reduce this dependency on incident data, and greatly improve the ability to assess the risk associated with new products for which such data does not exist. This paper outlines a risk assessment method developed for such a purpose. A detailed report describing the methodology is available. This paper also reviews the results of four case studies performed as part of the development of this risk assessment method. The four reports, providing full details of these trial applications, are also available.