Pattern classification using tabu search to identify the spatial distribution of groundwater pumping

Pattern classification and Tabu Search are integrated to optimize the zonation and associated average groundwater pumping rates. A simulated problem analogous to the Choushui Creek Alluvial in Janghauh county of Taiwan is used to examine the proposed method. Three problems are tested including (1) optimizing zonation patterns with known pumping rates and number of zones, (2) optimizing pumping rates and zonation patterns with a predetermined number of zones, and (3) optimizing all pumping rates, number of zones, and zonation. The results indicate that the proposed method successfully identifies spatial patterns of groundwater pumping for all three cases. However, some initial zonations result only in local optimums. Thus, using more than one initial zonation to increase search diversification is recommended. Application to different problems such as the spatial distribution of hydraulic conductivity values and area weighting of rainfall stations, is encouraged.