A proposal for quality monitoring in the shipyard panel line

Abstract The construction of cruise-ship hulls is a complex process, the initial stage of which focuses on panel fabrication and includes marking, cutting, shaping metal plates of considerable length (several tens of meters), and welding rigid structural parts onto them. The panel line, where this stage takes place, is characterized by a relatively rapid throughput of parts (within a few tens of hours), compared to subsequent stages involving (i) the fabrication of units, i.e., parts with dimensions of (20-40m)×(20-40m)×(3-5m) obtained by joining several panels, (ii) the fabrication of modules, i.e., parts with dimensions of (20-40m)×(20-40m)×(9-12m) obtained by stacking 3-4 units, and (iii) the final assembly, characterized by the erection of the hull in a dry dock or slipway [1]. The hull-construction process generates minimal waste, as defects found during fabrication are gradually corrected through in-process repairs; however, postponing these repairs until the final assembly stage can lead to production delays and increase costs, due to a sort of “error propagation”. To minimize such issues, it is crucial to identify and repair defects as early as possible. Therefore, implementing a conformity assessment of manufactured parts that focuses on critical quality characteristics (such as reference positions and distances) becomes essential. Given the large dimensions of the parts and the significant deformations they can undergo, dimensional tolerances are typically a few millimeters around their nominal values [2]. This paper proposes a novel statistical-quality-control (SQC) methodology for monitoring the panel line in cruise-ship shipyards. The methodology – which utilizes a standardized p control chart with variable-size samples – offers two unique features: (i) accommodating the high customization and specific quality characteristics of individual panels, and (ii) rigorously considering the measurement uncertainty of the large-volume metrology (LVM) instrument used for conformity assessment, in line with the ISO 14253-1:2017 standard [3]. Through a real-world case study conducted at a Fincantieri S.p.A. shipyard, the potential of the proposed methodology is demonstrated at two complementary levels: product conformity assessment, to identify and repair defects in manufactured parts promptly, and process stability monitoring, to "keep the pulse" of the process, detect and address potential issues before they escalate. The proposed approach offers flexibility, enabling quality managers to choose the most suitable “decision rule”; e.g., they can prioritize reducing the risk of false conformities while accepting a slight increase in the risk of false nonconformities, and vice versa). In the case study, a state-of-the-art Trimble SX10 total station – with a standard uncertainty (u) of a few tenths of a millimeter under typical shipyard conditions – was employed for conformity assessment [4]. However, the methodology can be adapted to any other LVM instrument, as long as the corresponding value of u is known. References 1. Mandal, N.R. (2017). Ship construction and welding, Vol. 329, Springer Series on Naval Architecture, Marine Engineering, Shipbuilding and Shipping 2. Singapore: Springer. 2. Montgomery, D.C. (2019). Introduction to statistical quality control (8th Edition), John Wiley & Sons. 3. ISO 14253-1:2017 (2017), Geometrical product specifications (GPS) – Inspection by measurement of workpieces and measuring equipment – Part 1: Decision rules for verifying conformity or nonconformity with specifications, October 2017, https://www.iso.org/standard/70137.html. 4. Maisano, D.A., et al. (2023). Dimensional measurements in the shipbuilding industry: on-site comparison of a state-of-the-art laser tracker, total station and laser scanner. Production Engineering, 17, 625-642.