Over the last couple of years there has been huge improvements in optimisation techniques but there has been a divide between academics and the real word with only big corporations being able to bridge these gaps.
On the one side we have scholars producing a vast number of papers on solving standard problems (NP complete problems) on the other hand we have real word problems that are not always well defined and difficult to formulate.
Our skills have been to bride this gap between real word problems and mathematical problems. Reformulating the constraints and objectives into standard or semi-standard mathematical models and then use latest techniques to optimise and solve these models. These techniques are mainly based on MILP and genetic algorithms.