An integrated approach to robust multi-echelon inventory policy decision

As collaboration between different supply chain echelons gains increasing attention, it is imperative to consider inventory policies from a network perspective rather than supposing each stage to be a single isolated player. Yet, optimized inventory policies obtained through traditional approaches are mostly based on deterministic and stable network conditions. They are not capable of delivering the desired results under current market dynamics or even greatly deteriorate the performance of the entire supply chain, leading to high stock levels or short sales. Thus, to cope with current turbulent market demands, robust multi-echelon inventory policies are needed for distribution networks in order to lower inventory costs as well as to maintain high responsiveness. An extensive academic literature analysis revealed that multi-echelon inventory models have been analyzed extensively in recent years. However, computational scale, integrity and non-convexity make the corresponding optimization problem intractable to exact analysis. In consequence, various approximation methods and heuristic algorithms have been proposed by researchers in the last few decades, mostly for two-echelon networks. Nonetheless, even the most delicate models force abstraction of reality and involve some kinds of simplification or approximation.These deficiencies can be compensated to a large extent by simulation models as they allow to reproduce and to test different decision-making alternatives (e.g. inventory policies) upon several anticipated supply chain scenarios (e.g. forecasted demand series). This allows ascertaining the level of optimality and robustness of a given strategy in advance. Nevertheless, simulation itself can provide only what-if analysis. Even for a small-sized problem, there exist a large number of possible alternatives, making exhaustive simulation impossible. In response to such difficulties, a new approach to robust multi-echelon inventory policy decision has been developed, which is composed of three interrelated components: an analytical inventory policy optimization, a supply chain simulation module and a metaheuristic-based inventory policy optimizer. Based on the existing approximation algorithms designed primarily for two-echelon inventory policy optimization, an analytical multiechelon inventory model in combination with an efficient optimization algorithm has been designed. Through systematic parameter adjustment, an initial generation of optimized multi-echelon inventory policies is calculated. To evaluate optimality and robustness of these inventory policies under market dynamics, the suggested inventory policies are automatically handed over to a simulation module, which is capable of modeling arbitrary complexity and uncertainties within and outside of a supply chain and simulating it under respective scenarios. Based on simulation results, i.e. the robustness of the proposed strategies, a metaheuristic-based inventory policy optimizer regenerates improved (more robust) multi-echelon inventory policies, which are once again automatically dynamically evaluated through simulation. This closed feedback loop forms a simulation optimization process that enables the autonomous evolution of robust inventory policies. The proposed approach has been validated by an industrial case study, in which favorable outcomes have been obtained.