Sustainability and Robust Decision-Support Strategy for Multi-Echelon Supply Chain System Against Disruptions

This paper considers the dynamic characteristics and resilience issues of multi-echelon supply chain system against disruptions, where the dynamical behaviour is modelled by a set of nonlinear difference equations. While the supply chain evolves continuously with complex dynamics in reality, resilience becomes more fundamental for sustainable development due to its sensitivity to external disturbance impacts. The complex nonlinear system is evaluated by several methods such as Lyapunov exponent, bifurcation diagram, permutation entropy and Poincaré map. The predictability of the time-series evolution is also explained by Hurst exponent computation, which is necessary to implement echo state network (ESN) in order to forecast dynamical data on the development of the supply chain and also to replicate the complex management schemes. Based on decision-support strategy with machine learning, resilience issue concerning various types of disturbing factors such as parametric variations, external disturbances and time-delay is discussed in order to ensure the sustainability of a supply chain. Since external disruptions are unavoidable, planners must have an insight into the overall system dynamics and make strategic and critical managerial decisions to overcome those challenges connected to disruptive incidents. Numerical simulations are given to validate the analysis of the dynamical behaviour and verify the efficiency of control theory with machine-learning, guaranteeing resilience and sustainability of a supply chain against market volatility. This work will be the smaller steps ensuring more visibility and more traceability for enterprise supply chain management under a highly uncertain environment