Stochastic search algorithms for optimal design of monitoring networks.

The article discusses a hybrid genetic algorithm (HGA) which combines the standard genetic algorithm (GA) with a local search (LS) operator for the design of monitoring networks. HGA was applied using data from ground level ozone in the U.S. environmental monitoring networks. Results show that HGA outperforms simulated annealing (SA) and GA in terms of quality of the solutions and reliability. However, computation time of HGA was noted to be longer than the SA and GA in a single-processor environment.