Spatiotemporal scheduling of the block assembly process in shipbuilding is to determine temporal information including a time period when each block is under assembly and spatial information including an assigned bay and the placement of the block within the assigned bay. Due to a large number of discrete variables to optimize, finding an optimal schedule in a reasonably short time is almost impossible. Instead, we divide the problem into two phases where the first phase determines the bay assignment and processing start date for each block and the second phase determines the coordinates and rotation of each block in its assigned bay. Our objective is to find a block assembly schedule that minimizes the number of blocks that fail to be processed on time and unfairness in workloads across bays and days. The proposed algorithm is tested on six datasets of block information provided by Korea Shipbuilding & Offshore Engineering Co., Ltd. (KSOE). Our algorithm speeds up the scheduling process and finds schedules of higher quality compared to the original schedules that are manually planned by KSOE.

造船中分段组装过程的时空调度是确定包括每个分段组装的时间段的时间信息和包括分配的舱位和分段在指定的舱位内的放置的空间信息。由于需要优化大量离散变量，在相当短的时间内找到最佳调度几乎是不可能的。相反，我们将问题分为两个阶段，第一阶段确定每个块的海湾分配和处理开始日期，第二阶段确定每个块在其分配的海湾中的坐标和旋转。我们的目标是找到一个区块组装时间表，最大限度地减少未能按时处理的区块数量以及跨湾和跨天工作负载的不公平性。该算法在韩国造船与海洋工程有限公司（KSOE）提供的六个区块信息数据集上进行了测试。与 KSOE 手动计划的原始计划相比，我们的算法加快了计划过程并找到了更高质量的计划。