A biased random-key genetic algorithm for the time-invariant berth allocation and quay crane assignment problem. (15th December 2017)

• Record Type: Journal Article
• Authors: Correcher, Juan Francisco; Alvarez-Valdes, Ramon
• Is Part Of: Expert systems with applications. Volume 89(2017)
• Page Start: 112
• Page End: 128
• Publication Date: 2017-12-15
• Subjects: Container terminal -- Berth allocation -- Quay crane assignment -- Metaheuristic -- Genetic algorithm -- Local search; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2017.07.028
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: We address Berth Allocation and Quay Crane Assignment Problems in a heuristic way We propose a Biased Random-Key Genetic Algorithm for BACAP and its extension BACASP Solutions of the Genetic Algorithm are improved by a Local Search The complete procedure obtains high-quality solutions for large instances Abstract: Maritime transportation plays a crucial role in the international economy. Port container terminals around the world compete to attract more traffic and are forced to offer better quality of service. This entails reducing operating costs and vessel service times. In doing so, one of the most important problems they face is the Berth Allocation and quay Crane Assignment Problem (BACAP). This problem consists of assigning a number of cranes and a berthing time and position to each calling vessel, aiming to minimize the total cost. An extension of this problem, known as the BACAP Specific (BACASP), also involves determining which specific cranes are to serve each vessel. In this paper, we address the variant of both BACAP and BACASP consisting of a continuous quay, with dynamic arrivals and time-invariant crane-to-vessel assignments. We propose a metaheuristic approach based on a Biased Random-key Genetic Algorithm with memetic characteristics and several Local Search procedures. The performance of this method, in terms of both time and quality of the solutions obtained, was tested in several computational experiments. The results show that our approach is able to find optimal solutions for some instances of up to 40 vessels and good solutions for instances of up to 100 vessels.