Operational Research: Making Things Better - The mathematics of improvement (Colin Reeves)

Operational research (OR) developed during World War II as a means of improving the use of the country's military resources. Since then its fields of application have become ever more numerous, but the fundamental issue is still the same - how can we improve? how can we make things better?

Many processes in business, industry and government are highly complex systems where it is not at all clear how things can be improved. The distinctive feature of the OR approach is to construct a model of the system in order to search for better solutions. Often this reduces to a search among a finite but very large set of possibilities---what we call a combinatorial optimization problem.

Many methods have been developed over the last 30 years, and the concern of academic researchers has been how to make the methods better. This lecture will focus on one of these methods - evolutionary algorithms, and illustrate some ways in which mathematics can be used to understand how they solve optimization problems.

Lecture date: March 14 2006