Towards Digital Twinning of Fabrication Lines in Shipyards

Abstract

The digital twinning concept stands as a remarkable opportunity to integrate sophisticated mathematical models within the context of existing manufacturing systems. Such models may provide shipyard management with predictive analytics, improving the final results at the strategic, tactical, and operational levels. Therefore, the possibility of integrating the Markovian-framework-based finite-state method into the context of ship production is presented in this study, including its outline, digital thread, and factory-floor data reliance. First, the predictive analytics problem is addressed by the finite-state method in the case of the shipyard’s fabrication line, and the obtained results are validated afterward using a numerical model through discrete-event theory. The predictive analytics indicate an almost ideal balancing of the fabrication line, except for the buffers storing stiffeners before the coat-dying and marking operations. In addition, the improvability analysis of the shipyard’s fabrication lines extended the scope of the predictive analytics using bottleneck identification and affecting the key performance indicators through a digital thread, as well as by improved maintenance strategies.