The Effect of Uncertainties on Multi-Echelon Serial Supply Chains

Abstract

Uncertainties are the major concerns in supply chain because existence of uncertainties degrades the performance of supply chain. Hence, business executives need to seriously focus towards controlling the effect of uncertainty on supply chain performance. In this study, a four echelon serial supply chain employed with reorder-point order-up-to level inventory replenishment (s, S) policy is modeled using system dynamics approach. Manufacturing systems adopting make-to-stock (MTS) and assemble-to-stock (ATS) manufacturing policy and operating under uncertain environment are modelled through system dynamics approach. A serial two-stage MTS manufacturing system is modelled through system dynamics approach and the behaviour is studied under the influence of uncertainty in demand, lead time, supplier’s acquisition rate, processing time and delay due to machine failure. Two different improved demand forecasting models are proposed to enhance the forecasting accuracy and reduce the bullwhip effect (BWE) and net-stock amplification (NSAmp). The first proposed model is the integrated approach of autoregressive integrated moving average (ARIMA) and generalized autoregressive conditional heteroskedasticity (GARCH) model denoted as ARIMA-GARCH to overcome the problem related to heteroskedastic nature of demand series. Second proposed model is the integrated approach of discrete wavelet transformation (DWT) and intelligence technique such as artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), least square support vector machine (LSSVM) and multi-gene genetic programming (MGGP) to deal with non-linear, non-stationary demand series.
Simulation study of multi-echelon supply chain indicates that target inventory significantly influence the BWE and it can be reduced through keeping target inventory at low level when there is low uncertainty in demand and lead time. From the analysis of manufacturing supply chain, it is observed that backlog at manufacturer’s end is significantly influenced by uncertainty in processing time and delay due to machine failure. The backup strategy adopted in manufacturing supply chain reveals that performance of manufacturing system is highly affected when uncertainty in supplier’s acquisition rate increases. The study proves that maintaining high service level at the bottom echelon is required to achieve high service level at the upper echelon of a supply chain. From the forecasting study, it is found that performance of the ARIMA-GARCH model outperforms the ARIMA model. Further, it is proved through case-study examples