考虑减载移泊的散货港口船舶调度优化

为提高散货港口的服务水平, 充分利用现有泊位资源, 研究了采用减载移泊策略的散货港口船舶调度优化问题; 考虑大型船舶减载移泊对散货港口船舶调度的影响, 以船舶进出港次序、移泊次序和移泊位置为决策变量, 以进出港船舶总等待时间最小为目标函数, 构建了混合整数线性规划模型; 基于模型特点设计了混合算法, 给出了生成初始种群的启发式规则, 提出了新种群的邻域构造策略, 并在模拟退火算法中引入有效的改进措施; 为验证方案及其算法的有效性, 对比了基于实际调研资料设计的方案与采用模型和算法优化的方案, 并分析了船舶乘潮比和进出港时段长度对方案优化结果的影响。研究结果表明: 与采用先到先服务思想和贪婪策略的2种现行船舶调度方案相比, 所得方案的平均优化率分别为11.07%和9.84%;船队规模从20艘增加到50艘时, 混合算法的求解耗时均在2min以内, 且所得目标函数值与下界的平均相对偏差为6.92%;随着船舶乘潮比的增加, 方案优化率和目标函数值先呈指数趋势增长, 而后趋于平稳, 乘潮比为50%左右时出现拐点; 随着进出港时段长度的增加, 方案优化率和目标函数值呈“M”形趋势变化, 且在进出港时段长度为130min左右时方案优化效果最为显著, 表明船舶调度优化模型与混合算法可行。 

Ship scheduling optimization on bulk cargo port considering ship lightening and berth shifting

To improve the service level of bulk cargo port and make full use of its existing berth resource, the ship scheduling optimization on bulk cargo ports using ship lightening and berth shifting strategies was studied.The impact of large ship lightening and berth shifting on ship scheduling in a bulk cargo port was considered, ship inbound/outbound order, berth shifting order and berth shifting position were taken as decision variables, the minimum waiting time of inbound and outbound ships was taken as the objective function, a mixed integer linear programming model was constructed.Based on the characteristics of the model, a hybrid algorithm and heuristic rules for generating an initial population were given.A neighborhood strategy for constructing a new population was proposed, and the effective improvement measures in the simulated annealing algorithm were introduced.To verify the effectiveness of this scheme and the algorithm, the comparison tests for the scheme and algorithm based on actual research data were given, and the influences of ship tidal ratio and length of inbound/outbound period onthe optimization results of the scheme were analyzed.Research result shows that comparing with the two current ship scheduling schemes with the first-come first-served idea and greedy strategy, the average optimization rates of the two proposed schemes are 11.07% and 9.84%, respectively.When the fleet size increases from 20 to 50, the calculation time of the hybrid algorithm always be less than 2 min, and the average relative deviation between the objective function value and lower bound is 6.92%.With the increase in ship tidal ratio, both the optimization rate and objective function value of the scheme increase exponentially at first, and then tend to be stable.The inflection point appears when the tidal ratio is 50%.As the length of inbound/outbound period increases, the scheme optimization rate and target value exhibit an M-shaped trend.The optimization effect is most significant when the length of inbound/outbound period is approximately 130 min.Obviously, the ship scheduling optimization model and the hybrid algorithm are feasible.