Three models were developed and tested for the optimal capacity expansion planning in a hypothetical multiaquifer system. The response of the system was included in the models using response matrices. All models are 0/1 mixed-integer programming models and enable the determination of minimum capital investment and operation costs of well fields and associated pipeline facilities while satisfying a set of system constraints. The performances of the models are compared in terms of computational requirements and approximation to pumpage costs under three water-demand schedules. Multiobjective analyses were conducted to develop trade-off curves relating pumpage to drawdown. The sensitivity of the capacity expansion policies to variations in demand requirements, interest rates, and system parameters are analyzed. The first model produces minimum total drawdowns and requires lesser computation time while the remaining two models yield minimum total costs but require more computation time. The variations in demand requirements, interest rates and system parameters causes variations in selection of the potential well fields, their completion time, resulting total costs and minimum total drawdowns.