We consider two Just-In-Time (JIT) purchasing models, one utilizing a few sources, and the other using the more conventional single source. We address the issue of splitting a large order quantity into multiple deliveries, taking account of the increase in the aggregate ordering, transportation, and inspection costs. For multiple sourcing, we formulate and solve a mathematical programming problem to obtain the optimal selection of suppliers and the size of the split orders. For single sourcing, we provide a procedure that yields the optimal number of deliveries.