|| Checking for direct PDF access through Ovid
The paper deals with the problem of minimizing the cost of pumping groundwater from a number of wells. A genetic algorithm approach is presented with the following characteristics: (a) Well locations are treated as optimization parameters and the well configurations are represented by a single integer combinatorial variable. (b) The objective function chosen is flat and appropriate scaling is performed. (c) No penalty functions are involved in constraint handling. Instead, the feasible region is transformed into an n-dimensional cube and at the same time a suitable boundary extension is introduced. Finally, the optimal well configuration is rapidly approached.