A simulation–optimization framework for enhancing robustness in bulk berth scheduling. (August 2021)

• Record Type: Journal Article
• Authors: de León, Alan Dávila; Lalla-Ruiz, Eduardo; Melián-Batista, Belén; Moreno-Vega, J. Marcos
• Is Part Of: Engineering applications of artificial intelligence. Volume 103(2021)
• Publication Date: 2021-08
• Subjects: Maritime logistics -- Berth Allocation Problem -- Metaheuristics -- Simulation–optimization -- Buffer management -- Simheuristics; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2021.104276
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: The service time of the vessels is one of the main indicators of ports' competitiveness. This, together with the increasing volume of bulk transportation, make the efficient management of scarce resources such as berths a crucial option for enhancing the productivity of the overall terminal. In real scenarios, the information available to port operators may vary once the planning has been elaborated. Unforeseen events, errors, or modifications in the available information can lead to inefficient terminal management and the initial scheduling might become unfeasible. This implies that the use of deterministic approaches may not be enough to maximize productivity. Therefore, in this work, proactive simulation–optimization approaches that utilize the information collected during the simulation for guiding the optimization search to provide robust solutions are proposed. Moreover, a multi-objective approach based on the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) for jointly tackling the problem objective as well as the deviations because of stochastic changes is developed. Finally, we also investigate the contribution of time management strategies such as buffers to absorb stochastic modifications and hence increase solutions' robustness. The computational results indicate, on the one hand, the benefit of integrating both types of objectives (i.e., deterministic and stochastic) to guide the simulation–optimization process, and on the other hand, the benefit of using the multi-objective approaches like NSGA-II. Finally, the incorporation of buffers leads to better performance in terms of reducing penalty costs due to disruptions, shortening the planning risks related to only considering deterministic planning. Highlights: Proactive approach for the Bulk Berth Allocation Problem. Novel simheuristic dealing simultaneously with deterministic and stochastic information. Integration of simulation during the search process allows reaching robust solutions. Buffer times increase the robustness of the solution to disruptive events.