(s,S) policy model for liner shipping refueling and sailing speed optimization problem

This work expounds on implementing an effective dynamic (s,S) policy to solve a liner shipping refueling and speed determination problem under both bunker prices and consumption uncertainties. While solving an optimization model which incorporates a continuous distribution is extremely challenging, we use sample average approximation method to solve it. However, the resulting problem is still a very large-scaled problem. Therefore, we propose two variations of the progressive hedging algorithm to tackle it. Numerical results show that our solution method is efficient and, in addition, our dynamic (s,S) policy model has significant cost reduction potential compared to stationary models.