Development of inventory policy for spare parts subject to age replacement has previously been studied using (s, S) type ordering policies. When age replacement causes a large percentage of spare part demand to result from scheduled replacement rather than random failure, this policy fails to take advantage of the presumably known schedule of planned replacement times. This paper uses simulation analysis to characterise the conditions under which a dynamic ordering policy such as Wagner-Whitin produces significant cost reductions relative to (s, S), by virtue of taking the planned replacement times into account. A modification to the basic Wagner-Whitin method is introduced to address the issue of overly frequent, small order quantities when the demand pattern is dominated by random failures. Simulation results demonstrate the Wagner-Whitin algorithm significantly outperforms the (s, S) policy when a sufficient percentage of demand is due to scheduled replacement. The modified Wagner-Whitin policy outperforms the (s, S) policy under any mixture of planned replacement and random failure spare parts demand.