Research On Robust Optimization Of Ship Oil Inventory Routing Under Uncertain Demand

With the development of shipping industry,the consumption of ship oil accounts for an increasing proportion of the apparent consumption of fuel oil.By 2015,the apparent consumption of fuel oil in China is 28.16 million tons,of which the consumption of ship oil is 15 million tons,accounting for about 53%of the total apparent consumption of fuel oil.The demand for ship oil is affected by factors such as the price of crude oil and the number of ported vessels and port throughput,and the uncertainty is strong.In this paper,aiming at the demand uncertainty of ship oil,the inventory path of ship oil is researched.Taking Z Company as an example,the demand of each port is uncertain in each planning period,and the inventory capacity of demand port is certain,how to determine the optimal path to satisfy the demand and make the total cost minimum.In this paper,a mixed-integer programming model is established.First,the inventory routing model of ship oil under certain demand is established.A decision variable is introduced to represent the proportion of the ports that the demand is satisfied in each period.The optimal path and quantity are determined by establishing the path flow constraint,inventory constraint,berth constraint and capacity of the ship constraint,which minimizes the total cost composed of transportation,inventory holding,and shortage costs.Based on the deterministic model,considering the uncertain demand of each demand port in each period,the "budget of uncertainty"parameter is introduced to establish the inventory routing model of ship oil based on single-stage robust optimization.Then,under the different demand deviation percentage and budget of uncertainty,considering the change of the total cost and the model of robustness,also analysis the relationship of them.In this paper,we use Lingo 11.0 software to solve the certain demand and single-stage robust optimization model respectively.The result indicates that the maximum deviation percentage for each demand becomes larger,the value of the objective function also increases;when the demand for a certain deviation,budget of uncertainty becomes larger,the price of robustness also increases.On the contrary,when the value of budget of uncertainty is smaller,the price of robustness slightly increases,and value of the objective function that is the total cost is also smaller.In the single-stage robust optimization model,although the total cost larger than the deterministic models,the solutions of the model are feasible.