Low-carbon berth allocation: An analysis of the effectiveness of an enhanced multi-objective artificial bee colony algorithm based on a case study

Abstract

Marine terminals are essential components of international trade networks and global markets. To guarantee the rapid and consistent growth in maritime trade, managers must employ suitable techniques to handle operational challenges and meet market needs. One of the critical decisions in operational planning is the allocation of berths. A well-designed berth allocation plan can greatly boost the productivity and competitiveness of seaports. Despite the extensive research on berth allocation, there remains a notable gap in studies focusing on low-carbon berth allocation. As energy shortages and global warming intensify, low-carbon has increasingly become key terms across various sectors. Under the circumstances, this work addresses a multi-objective stochastic berth allocation problem for minimizing the average carbon emission and total service time. Firstly, a stochastic programming method is employed to formulate the uncertain arrival time and operation time of vessels, then a multi-objective chance-constrained programming model is constructed to formulate the studied problem. Secondly, an enhanced multi-objective artificial bee colony algorithm incorporating stochastic simulation (EMOABC) is specially designed. Finally, a large number of comparison experiments between EMOABC and nondominated sorting genetic algorithm II (NSGA-II) are performed. Through observing and analyzing the experimental results, two conclusions are acquired as follows: (i) EMOABC obtains the smaller IGD values and larger HV values than NSGA-II on all the test instances, indicating that it has better performance than NSGA-II for solving the considered problem; and (ii) EMOABC uses less running time in dealing with test problems of different scales compared to NSGA-II, suggesting that it has lower computational complexity than NSGA-II.