可控制前置时间需求分布未知的部分缺货待补混合库存策略

Mixed Inventory Policy with Backlogging for Unknown Demand Distribution During Controllable Lead Time

同时考虑可控制前置时间与货币时间价值两项因素,假设前置时间内需求量的概率分布未知,建立了部分缺货欠待补混合库存模型,并对混合库存模型进行了数值验证.研究证明,混合库存模型的总变动成本函数为凸函数,然而由于前置时间内的随机性需求量会造成无法求取期望缺货量的精确值,因此采用最小最大准则找出具有最大期望总成本的现值函数,进而求取使得期望总成本现值为最小的最优值,以确定最优订购策略.

Taking into account two factors of the con- trollable lead time and time value of money, the thesis establishes a mixed inventory policy model with backlogging by assuming that the demand distribution is unknown. The re search has proved that the total variable cost function of mixed inventory model is a convex function, however, as the stochastic demand in the lead time will not lead to solve the exact value of expected shortage, the thesis uses Minimax Distribution Free Procedure to establish the total cost function with the maximum expected present value, and then solves the optimal value that can minimize the present value of the expected total cost to determine optimal ordering policy. Finally, the thesis verifies the mixed inventory model by numerical examples.