Conference Agenda

The aim of berth allocation planning is to derive conflict-free vessel assignments to the quay of a container terminal. The berthing schedule resulting from solving the corresponding Berth Allocation Problem (BAP) consists of the berthing times and positions of all vessels that are expected to arrive within a certain timeframe; these vessels are scheduled according to their respective arrival and handling times. However, both these times are uncertain due to different influences, e.g., weather, technical breakdowns or maintenance.

Deviations from the planned arrival and handling time lead to delayed vessel departures, which cause waiting time for the succeeding vessels and can ultimately result in conflicts that may impede the schedules’ feasibility. Hence, updating or re-planning of berthing schedules can become necessary, but this is costly and may be impossible when a plan is already under execution. Therefore, the aim of this work is to derive robust berthing schedules which are resistant to uncertainties, especially of handling times, by applying time buffers.

Two main strategies can be distinguished regarding the development and use of time buffers for mitigating these uncertainties: Maximizing the slack (buffer) between each pair of succeeding vessels or considering a predetermined individual time buffer in the optimization of the berthing schedule. The results of both approaches are evaluated in this work from an ex-post perspective using real vessel AIS data, i.e., the actual arrival and handling times of vessels approaching an existing port.

Uncertainty in demands is a pervasive issue in manufacturing and logistics.

The implementation of just-in-time approaches and minimal inventory management has led to significant difficulties in the supply chain, particularly in the planning of transportation of goods.

This has resulted in avoidable transports, which have caused unnecessary costs and greenhouse gas emissions.

When some of the input data is uncertain, robust optimization is a method to model these uncertainties and find efficient solutions.

It is well-known that applying classic robust optimization techniques may result in overly conservative decisions, which may not always be the best course of action in real-world applications.

Adjustable robust optimization (ARO) relaxes this notion by introducing a second stage, where some of the decisions can be made after the uncertain data is known.

The use of ARO achieves better results compared to the application of classic robust optimization.

We develop an ARO model for a transportation problem where the demands for materials to be transported from a source to a sink are uncertain.

All necessary physical constraints are taken into account, including maximum vehicle load and 3D cargo bay packing.

For the demand uncertainty, different uncertainty sets are applied, and methods are investigated to estimate the parameters for these uncertainty sets.

In practise, this model and budgeted cumulative uncertainty sets are used to develop a heuristic solution approach for the problem. This resulting approach uses meta-heuristics like simulated annealing and genetic algorithms.

We evaluate its performance on both random generated data and real-world data.

In airport operations, optimally using dedicated personnel for baggage handling tasks plays a crucial role in the design of resource-efficient processes. Teams of workers with different qualifications must be formed and baggage tasks must be assigned to them. Each task has a time window within which it can be started and should be finished. Violating these temporal restrictions incurs severe financial penalties for the operator. In practice, various components of this process are subject to uncertainties. We consider the aforementioned problem under the assumption of stochastic travel times across the apron. We present two binary program formulations to model the problem at hand and solve it with a Branch-Price-Cut-and-Switch approach, in which we dynamically switch between two master problem formulations. Furthermore, we use an exact separation method to identify violated rank-1 Chvátal-Gomory cuts and introduce an efficient branching rule. We test the algorithm on instances generated based on real-world data from a major European hub airport. Our results indicate that our algorithm is able to significantly outperform existing solution approaches. Moreover, an explicit consideration of stochastic travel times allows for solutions that utilize the available workforce more efficiently, while simultaneously guaranteeing a stable service level for the baggage handling operator.