Production planning of a perishable product with lead time and non-stationary demand

We study a production planning problem for a perishable product with a fixed lifetime, under a service-level constraint. The product has a non-stationary stochastic demand. Food supply chains of fresh products like cheese and several crop products, are characterised by long lead times due to maturation processes. Long lead times in combination with non-stationary demand, tend to induce high safety stock levels. For perishable products, this can lead to an increase in outdating. To minimise waste due to outdating, one issues the oldest product first, i.e. a FIFO issuing policy. In case of out-of-stock, demand is backlogged. The objective of our study is to develop a production planning method for a perishable product with non-stationary demand and a long deterministic production lead time, that meets fluctuating demand and provides low safety stocks and few outdating. For the case of a fixed lifetime perishable product confronted with non-stationary demand, zero lead time and a FIFO issuing policy, Pauls-Worm et al. (2012) formulate an MILP model that determines a stock-age-dependent replenishment cycle policy. Specifically, an a-service-level constraint is applied. In our study, we adapt the MILP model adding a lead time and we determine safety stocks based on the desired fill rate. Based on this adapted model, we develop a heuristic procedure to determine a production plan for a perishable product with a long lead time. As the input demand forecasts become more reliable with an approaching selling date, given the long lead time, it may be advantageous to correct the production plan for more accurate forecasts. The procedure is based on a rolling planning horizon in which demand forecasts are updated. The first results are promising: the procedure allows for relatively low safety stocks and is able to meet the desired fill rate.