Integrated methods for optimization. (Q855171)

The book deals primarily with the unification of mathematical programming and constraint programming. It brings the methods of both fields under one roof, so that they and their combinations are all available to solve a problem. The book is intended for those who wish to learn about optimization from an integrated point of view, including researchers, software developers, and practitioners. It is also for postgraduate students interested in a unified treatment of the field. Chapter 1 serves the function of introducing the main ideas of the book by example. One can get a very good idea of what integrated problem solving is all about by reading this chapter alone. Chapter 2, on search, begins with an overview of the algorithmic framework, and then successively takes up branching algorithms, constraint-directed search, and heuristic methods. Chapter 3, on inference, begins with some basic concepts from the constraints literature. It develops a theory of inference for inequality-constrained problems and propositional logic. Chapter 4, on relaxation, has a stronger flavor of mathematical programming due to its discussion of linear programming, mixed integer modeling, and heavy dose of cutting plane theory. Chapter 5 is something of an appendix.