Areas of expertise


Shaun started his research career as an applied mathematician and was awarded his PhD, concerned with morphological and hydrodynamic instabilities in binary alloy solidification, by Bristol University in 1990.

He then spent 5 years at BAE Systems' Sowerby Research Centre developing computational fluid dynamics algorithms and software for transonic and supersonic aerodynamics with an emphasis on methods for shocks and adaptive meshing.

In 1994 Shaun joined Cranfield University as a lecturer in Applied Mathematics at the Shrivenham campus. Since then, has built up an international research reputation in the field of Algorithmic (Automatic) Differentiation (AD). This has involved both developing new computer packages for differentiating programs written in Fortran and Matlab, and also applying AD techniques to computational fluid dynamics solvers for industrial partners such as: QinetiQ, BAE Systems Ltd, Airbus UK. He is actively involved in the international AD community and was co-editor of Recent Advances in Algorithmic Differentiation, the Proceedings of the 6th International Conference on Automatic Differentiation, July 2012.

Shaun, Emmanuel Tadjouddine, John Pryce and John Reid, were the first to implement the vertex elimination algorithm for automatic differentiation using compiler-like source transformation methods and demonstrated that such an approach could greatly improve the efficiency of automatically generated code for calculating Jacobians

Between 1998 and 2006 Shaun was involved in the development of the research codes Air3D and Air3Dfft, which are well known in the blast loading and hazard assessment community. This work was performed in conjunction with civil engineers and former colleagues Dr Peter Smith and Dr Tim Rose. This work has now been superseded by the development of the commercialised ProSAir package.

Current activities

Shaun has an international research reputation in the field of Algorithmic (or Automatic) Differentiation, the differentiation of functions specified by computer programs. He led, with colleagues Patrick Farrell (University of Oxford), Andreas Griewank (Yachaytech University), Paul Hovland (Argonne Labs.) and  Jens-Dominik Mueller (Queen Mary University of London) the organisation of the 7th International Conference on Algorithmic Differentiation held in Oxford in September 2016. He is guest editing a special edition of Optimization Methods and Software for papers arising from the conference.

Shaun has commercialised ProSAir, Cranfield University's Cartesian Mesh CFD solver specialised for assessing blast loads on structures. ProSAir is available for both Commercial and Academic use and is a successor to the Air3D research code.  For more details see the ProSAir homepage.

Using his expertise from the ProSAir project, Shaun is currently working with Stephen Johnson and other colleagues on development of terrorism risk models for insurers Pool Re and Guy Carpenter.

Since 2001 Shaun has developed the MAD package for algorithmic differentiation of computer programs written in the Matlab programming language. MAD has featured in many publications and is commercially licensed by the University for distribution by the TOMLAB company.


  • Pool Re and Guy Carpenter - developing models for terrorism risk assessment.
  • Airbus UK and BAE Systems - use of Algorithmic Differentiation to develop an adjoint unstructured finite volume CFD solver.
  • BAE Systems - CFD tools for blast prediction.
  • TOMLAB Optimization (USA) - commercialisation of the MAD package for Algorithmic Differentiation in MATLAB.


Articles In Journals

Conference Papers