W. D. Potter, W. Bi, D. Twardus, H. Thistle, M. J. Twery, J. Ghent, and M. Teske
The AGDISP Aerial Spray Simulation Model is used to predict the deposition of spray material released from an aircraft. The prediction is based on a well-defined set of input parameter values (e.g., release height, and droplet size) as well as constant data (e.g., aircraft and nozzle type). But, for a given deposition, what are the optimal parameter values? This problem is considered to be a parametric design problem or more generally a configuration problem. Attempting to optimize a configuration based on some set of constraints is known to be extremely difficult (NP-Hard). We use the popular Genetic Algorithm to heuristically search for an optimal or near-optimal set of input parameters needed to achieve a certain aerial spray deposition. Having this knowledge can benefit forest managers substantially, especially regarding such issues as cost, environmental safety, and forest treatment accuracy.