Reliable Inventory Management of Key Parts for Wind Turbine Production Under R&D Uncertainty

Due to the rapidly evolving technology within the renewable energy industry, especially in the wind power sector, companies are dedicating significant resources to research and development (R&D) efforts aimed at creating better and cheaper alternatives. As a result, it becomes imperative to account for the impact of R&D on the management of critical parts with lengthy lead times. This article underscores the necessity of considering R&D-related uncertainties in inventory management. The problem is formulated as a Markov Decision Process, wherein the reliability constraint poses challenges. To tackle this, the concept of the efficient frontier is introduced, which addresses reliability as one of the objectives alongside cost, resulting in a bi-objective optimization problem. In addition, computational challenges arise from the vast state space size, and the stochastic dual dynamic programming algorithm is employed to solve it efficiently. Finally, a compelling case study backed by industry data is presented, which demonstrates that our approach yields highly dependable inventory management strategies, achieving remarkable reliability while maintaining cost-effectiveness.