Research On Optimization Of Container Ship Speed Based On Lasso Regression To Estimate Ship Fuel Consumption

With coronavirus disease 2019(COVID-19)reshaping the global shipping market,in particular,complex changes have taken place in fuel prices,transportation demand and market environment.As a result,the fuel cost of ships in the course of sailing has also been greatly reduced,and the proportion of fuel cost in the total sailing cost will also be significantly reduced.The sailing speed of the ship will directly affect the fuel consumption,so the sailing speed of the ship is particularly critical.Controlling the sailing speed in a reasonable range can reduce the total sailing cost.Therefore,it is particularly important to optimize sailing speed and reduce sailing cost.Sailing speed will affect sailing time,so the tradeoff between sailing cost and sailing time is the research focus of this thesis.Many ships in the EuropeAsia trades that need to sail through the Suez Canal begun to detour via the much longer route,the Cape of Good Hope.In order to explain the behavior of ships detouring longer routes during the pandemic and to predict their future route choices with optimizing ship speed,a ship fuel consumption model based on LASSO regression estimation is constructed based on AIS data and Python Copernican meteorological data set to dynamically estimate the fuel consumption of ships during navigation in this thesis.Then,a multi-objective mathematical model based on the fuel consumption model obtained by LASSO regression was established to minimize the total voyage cost and total voyage time.Our thesis used a multi-objective particle swarm optimization algorithm(MOPSO)to find the Pareto optimal solution.After that,the weighted sum(WS)method is introduced to deal with the route selection problem in some specific situations.In this thesis,a case study was conducted on the real data from CMA CGM,a leading worldwide shipping company.This thesis analyzed the actual sailing process of the ship choosing two different routes,and the ship speed is optimized,as the same time,the feasibility of the model method framework proposed in this thesis is verified.Then four scenarios of fuel prices and charter rates were built and analyzed based on actual fuel price and charter rate.The results of the scenario analysis show that,with low fuel prices and charter rates,the detour around the Cape of Good Hope is preferred only in the scenario of low fuel price and low charter.In view of low fuel prices and charter rates,an analysis is made of how much discount the Suez Canal tolls need to be given in order for ships to re-pass through the Suez Canal.In addition,this thesis shows that adjusting the discount of canal tolls according to the actual fuel prices and charter rates can make ships re-pass through the Suez Canal,and the Suez Canal Authority can reduce the canal tolls according to the results of this thesis,so as to get ships re-pass through the Suez Canal.