Simulation-based analysis of the impact of information,cooperation and forecasting techniques on bullwhip effect in multi-stage supply chains

This research investigates the bullwhip effect in multi-stage supply chains with multiple customers and retailers. The impact of customer information sharing, cooperation, correlation and smoothing parameters are considered using simulation modeling and statistical analysis. In addition, this research evaluates the impact of the aforementioned factors on supply chain costs. The supply chain in this study is a distribution network consisting of a supplier, a manufacturer, and a distributor center that serves three retailers with three customers each. The customer demand is assumed to be stationary and is modeled using a first order autoregressive process. Two forecasting techniques, minimum-mean square error and exponential weighted moving average, are used to estimate the customer demand. Traditional periodic review order-up-to inventory policy is employed for order placing decisions. The lead time for order replenishment is assumed to be constant. A design of experiments is used to develop the twenty-four scenarios of the simulation study. An analysis of variance approach is used to evaluate the main effects and the interaction effects on the bullwhip and supply chain costs. Findings of this research reveal that the correlation and smoothing parameters of the two forecasting techniques have a larger impact on bullwhip effect than information sharing and cooperation.