Akademik Veri Yönetim Sistemi
OCEAN ENGINEERING, cilt.362, 2026 (SCI-Expanded, Scopus)
Digital transformation in shipbuilding involves high-stakes investment decisions under deep uncertainty, where deterministic planning approaches fail to capture risk propagation across interdependent technologies. Existing studies typically evaluate technology investments in isolation and treat uncertainty as a post-hoc assessment, limiting their applicability in complex industrial environments. This study introduces a simulation-embedded optimization framework that integrates stochastic risk evaluation into the evolutionary search process, enabling risk-aware, multi-period, dependency-constrained investment planning under uncertainty. The hybrid Genetic Algorithm-Forecasting-Monte Carlo (GA-F-MC) framework incorporates expected cash flows, technological dependencies, and budget constraints within a scenario-based stochastic environment. A real-world case study based on a medium-sized Turkish shipyard demonstrates that the approach achieves an 18.6% increase in expected Net Present Value (NPV) and a 23% reduction in variance compared to a deterministic baseline. Sensitivity analysis identifies an annual investment level of approximately USD 400,000 as an optimal risk-return balance, while Value-at-Risk (VaR) and Conditional Value-at-Risk (CVaR) confirm effective downside risk control. The results reveal a staged capability-building pathway, where foundational technologies enable the deployment of advanced automation systems. The proposed framework provides a transferable and scalable decision-support tool for risk-aware investment planning in complex engineering systems.