In-transit perishable product inspection

We determine the value of monitoring perishable freight in-transit for a single vehicle traveling from an origin to a destination. We develop a computationally practical approach for determining the optimal expected cost function and an optimal policy, based on an infinite horizon partially observed Markov decision process model. Structural properties of the optimal expected cost function and optimal policy are determined. These results can lend insight when deciding whether to acquire the capacity to monitor freight status in transit and what actions to take, based on the data from the in-transit monitoring, that optimally increase expected supply chain productivity.

► We examine the value of monitoring perishable freight quality in transit. ► We model the problem as a partially-observed Markov decision process (POMDP). ► We develop an efficient procedure to solve the POMDP model. ► We present structural properties for both an optimal policy and an optimal expected cost function. ► We examine two benchmark cases of the original model, including the current practice in industry.