Abstract:Port operations have been suffering from hybrid uncertainty, leading to various disruptions in efficiency and tenacity. However, these essential uncertain factors are often considered separately in literature during berth and quay crane assignments, leading to defective, even infeasible schedules. This paper addressed the integrated berth allocation and quay crane assignment problem (BACAP) with stochastic vessel delays under different conditions. A novel approach that combines both proactive and reactive strategies is proposed. First, a mixed-integer programming model is formulated for BACAP with quay crane maintenance activities under the ideal state of no delay. Then, for minor delays, buffer time is added to absorb the uncertainty of the arrival time of vessels. Thus, a robust optimization model for minimizing the total service time of vessels and maximizing the buffer time is developed. Considering that the schedule is infeasible when a vessel is seriously delayed, a reactive model is built to minimize adjustment costs. According to the characteristics of the problem, this article combined local search with the genetic algorithm and proposed an improved genetic algorithm (IGA). Numerical experiments validate the efficiency of the proposed algorithm with CPLEX and Squeaky Wheel Optimization (SWO) in different delay conditions and problem scales. An in-depth analysis presents some management insights on the coefficient setting, uncertainty, and buffer time.
What problem does this paper attempt to address?
This paper attempts to solve the problem of mixed uncertainty in port operations, especially the Berth Allocation (BA) and Quay Crane Assignment (QCA) problems. These problems are often affected by various uncertain factors in actual operations, such as ship delays (external and natural factors) and quay crane maintenance (internal and human factors). These uncertain factors lead to a decline in port operation efficiency and reliability. However, the existing literature usually considers these uncertain factors separately, resulting in the generated plans may be defective or even infeasible.
To meet this challenge, this paper proposes a new method combining Proactive and Reactive strategies, aiming to improve the robustness and agility of port operations. Specifically, the main contributions of the paper are as follows:
1. **Establish a new Proactive - Reactive BACAP mathematical model**: Different indicators are adopted for different uncertain factors to balance robustness and agility.
2. **Propose an Improved Genetic Algorithm (IGA)**: This algorithm can find a solution that is 22.86% better than the Squeaky Wheel Optimization (SWO) algorithm within 225 seconds.
3. **Conduct sensitivity analysis**: The analysis shows that port operators should set coefficients and buffer times according to the uncertainty of ship arrival times; the larger the buffer time, the stronger the risk - resistance ability, but the service efficiency will be reduced.
### Problem Description
This paper studies a container terminal with a continuous coastline. Berths are divided into 10 - meter segments, and the planning time is discretized into 1 - hour intervals. All berths and quay cranes are available at the start of the planning. The number of quay cranes assigned to each ship varies with time, and the specific quay crane assignment also changes over time. Therefore, quay crane assignment includes the assignment of quantity and specific quay cranes.
#### Maintenance and Virtual Ships
To improve reliability, quay cranes need to undergo Preventive Maintenance (PM). The start time of PM needs to be combined with the berth plan. Ships are not allowed to enter the berth area where the quay crane under maintenance is located. Virtual ships are introduced to accurately describe the relationship between PM and berth and quay crane assignments. For example, when quay crane k is under maintenance, assume that there is a virtual ship (Vessel i′) served by quay crane k. The length of the virtual ship is consistent with the working coverage width of quay crane k, and the berth time of the virtual ship (denoted as σk) is the same as the actual PM time of quay crane k. In this way, the PM decision - making problem of quay cranes is integrated into BACAP.
#### Uncertainty and Buffer Time
This paper focuses on the problem that the arrival time of ships at the terminal exceeds the expected time due to uncertain factors. To increase the robustness of berth and quay crane assignment plans, buffer time is added. As shown in Figure 1, ai is the expected arrival time of ship i, and ei is the processing end time of ship i. The buffer times of ships i and j are denoted as θi and θj respectively. When the delay is within the buffer time, the original plan can achieve the expected effect. It should be noted that the larger the buffer time, the stronger the ability of the corresponding assignment plan to absorb uncertainty, but the service efficiency of the terminal will be reduced. Therefore, it is necessary to balance the service efficiency and robustness indicators and set an appropriate buffer time.
### Mathematical Model
This paper constructs three mathematical models - a deterministic model, a baseline planning model, and a reactive model - to obtain robust plans and effectively evaluate the impact of uncertainty on berth and quay crane assignments and PM plans. The relationships among these three models are shown in Figure 2.
#### Deterministic Model
Objective function:
\[ \min f_1 = \sum_{i \in I} [(e_i - a_i) + \beta_i \cdot \Delta b_i] \]
Constraints:
1. \( b_i + l_i \leq b_j + L \cdot (1 - x_{ij}), \forall i, j \in I, i \neq j \)
2. \( e_i \leq s_j + H \cdot (1 - y_{ij}), \forall i, j \in I, i \neq j \)
3. \( x_{ij} + x_{ji} + y_{ij} + y_{ji} \geq \)