Drones are receiving popularity with time due to their advanced mobility. Although they were initially deployed for military purposes, they now have a wide array of applications in various public and private sectors. Further deployment of drones can promote the global economic recovery from the COVID-19 pandemic. Even though drones offer a number of advantages, they have limited flying time and weight carrying capacity. Effective drone schedules may assist with overcoming such limitations. Drone scheduling is associated with optimization of drone flight paths and may include other features, such as determination of arrival time at each node, utilization of drones, battery capacity considerations, and battery recharging considerations. A number of studies on drone scheduling have been published over the past years. However, there is a lack of a systematic literature survey that provides a holistic overview of the drone scheduling problem, existing tendencies, main research limitations, and future research needs. Therefore, this study conducts an extensive survey of the scientific literature that assessed drone scheduling. The collected studies are grouped into different categories, including general drone scheduling, drone scheduling for delivery of goods, drone scheduling for monitoring, and drone scheduling with recharge considerations. A detailed review of the collected studies is presented for each of the categories. Representative mathematical models are provided for each category of studies, accompanied by a summary of findings, existing gaps in the state-of-the-art, and future research needs. The outcomes of this research are expected to assist the relevant stakeholders with an effective drone schedule design.

Disruption occurrences in liner shipping operations affect schedule reliability and may increase the total cost of delivering cargoes at ports. If a vessel experiences disruption occurrences either at ports of call or in sea, the liner shipping company is required to decide on the schedule recovery action to execute in order to recover the resulting delays. This study formulates a novel mathematical model for the vessel schedule recovery problem (VSRP) in liner shipping. The objective aims to minimize the total profit loss, suffered by the liner shipping company due to disruption occurrences at a given liner shipping route. A total of four recovery strategies are considered in the model, which include: (1) vessel sailing speed adjustment; (2) vessel handling rate adjustment; (3) port skipping without container diversion; and (4) port skipping with container diversion. The proposed mathematical formulation for the nonlinear VSRP model is solved to the global optimality using BARON. A set of computational experiments are further performed for the Middle East/Pakistan/India-West Mediterranean (WM3) route, which is served by the OOCL liner shipping company, for various scenarios of disruption occurrences in sea and at ports. The results from the performed analyses demonstrate potential benefits for liner shipping companies from using the proposed methodology for various realistic scenarios of disruptions.

Increasing size of vessels and formation of alliances are the common strategies, which have been widely used by liner shipping companies over the last years, aiming to serve the growing demand for the international seaborne trade efficiently. However, without a proper design of vessel schedules liner shipping companies may incur substantial monetary losses. This paper proposes a novel collaborative agreement, according to which multiple vessel arrival time windows, start and end times for each time window, and multiple handling rates during each time window are offered by the marine container terminal operator to the liner shipping company at each port of the given liner shipping route. The vessel scheduling problem is formulated as a mixed integer nonlinear programming model, where the total liner shipping route service cost is minimized. A set of linearization techniques are applied to the original model, and the linearized model is solved using CPLEX. A number of computational experiments are performed for the Pacific Atlantic 1 liner shipping route, served by the NYK liner shipping company. Results showcase effectiveness of the adopted solution methodology. Moreover, the proposed collaborative agreement outperforms the existing collaborative agreements, where either vessel arrival time windows or handling rates are offered to the liner shipping company at ports, in terms of the total route service cost on average by 10.9% and 6.6%, respectively. Therefore, the proposed collaborative agreement can be considered as an efficient alternative for improving the liner shipping operations.

Considering a substantial increase in the international seaborne containerized trade volumes, marine container terminal operators have to improve efficiency of the processes inside their terminals in order to meet the growing demand. An efficient berth scheduling is of a high importance for the terminal's performance, as it significantly influences the turnaround time of vessels. This paper proposes a novel Evolutionary Algorithm to assist with berth scheduling at marine container terminals that, unlike published to date studies on berth scheduling, applies a parameter control strategy. Specifically, an adaptive mechanism is developed for the mutation operator, in which the mutation rate is altered based on feedback from the search. The objective of the proposed mixed integer model aims to minimize the total weighted vessel service cost. A set of numerical experiments are conducted to assess performance of the developed algorithm based on a comparison against a typical Evolutionary Algorithm that applies a constant mutation rate value, determined from the parameter tuning analysis. Results indicate that the optimality gap does not exceed 0.80% for both algorithms. Furthermore, deployment of the adaptive mechanism for the mutation operator yields an average of 5.4% and 8.5% savings in terms of the total weighted vessel service cost for medium and large size problem instances, respectively, without a significant increase in the computational time.