OPTIMAL ORDERING POLICIES WITH 2 SUPPLIERS WHEN LEAD TIMES AND DEMANDS ARE ALL STOCHASTIC

Splitting orders among several suppliers is a common practice. Earlier theoretical studies considered only the strategy of splitting an orde r equally among the suppliers; they concentrated on the effects of ord er splitting on reducing the 'effective' lead time demand and safety s tocks, but the effects on the overall ordering policy have not been co nsidered. This paper presents a procedure for determining the optimal order policy; i.e., the total order quantity, reorder point and propor tion of split between two suppliers. The optimization objective is: mi nimize the sum of annual holding and order costs, subject to a maximum stockout risk. Our procedure handles any stochastic form of daily dem and and lead times. Some interesting observation from our solutions ar e: (i) The major advantage of order splitting is in the reduction of c ycle inventory cost, whereas the safety-stock reduction is usually uni mportant (contrary to earlier findings); (ii) although intuitively one might use suppliers with the shortest (mean) lead times, lower cost i s actually achieved if the second supplier's mean lead time is 'suitab ly' larger than the first's; this often means excluding suppliers with the lowest mean lead times; (iii) the optimal proportion of order spl it varies with, among other factors, the difference in the suppliers' mean lead times.