Parallel Non-Linear Optimization: Towards the Design of a Decision Support System for Air Quality Management

Abstract

We address the optimization component of a decision support system for air quality management, and describes some benchmark studies assessing its effectiveness. Because of the computationally demanding nature of the objective function we are interested in parallelizing the quasi-Newton BFGS algorithm selected for initial study. This is achieved by concurrent evaluation of functions in finite difference approximations to the derivative and a method of interval subdivision in simple bound constrained line searching. In a realistic case study, use of the parallel optimization algorithm is shown to have significant performance gains over other methods of solution.