Marketing-operations interface: a study of compound bullwhip effect and information sharing in supply chain

The inquiry of how the interface between marketing and operations would affect the supply chain performance in term of bullwhip effect is addressed in this dissertation through three separate but correlated studies. In the first study, the bullwhip effect is analytically measured in a simple two-stage supply chain with one supplier, and one retailer who experiences a compound demand of a first-order autoregressive process and a stochastic retail price. Our study demonstrated that the bullwhip effect can be, under certain conditions, stronger or weaker than the case where the price is not considered. In addition, quantitative analysis is also performed to study the behavior of the bullwhip effect. It is found that the magnitude of both lead time and autocorrelation coefficient impacts on bullwhip effect has been affected by the appearance of price and its interactions with demand. In the second study, the bullwhip effect is analytically measured for a two-stage supply chain, including a manufacturer, and a retailer who serves a market demand. The demand is assumed to be a combination of a first-order autoregressive process, a retail price (set by the retailer), and an asymmetric customer reference price (representing for customer valuation). Furthermore, a series of existence conditions of the bullwhip effect with positively, negatively, and no correlated demands is identified by examining the compound effect of the autocorrelation, retail and reference prices, as well as the demand-price interactions. In addition, it is also found that the bullwhip effect in the case of demand with retail and reference prices is either stronger or weaker than that of demand without these prices in several scenarios. The finding indicates that an under- or over-estimating issue may arise if the retail and reference prices are not properly considered. Moreover, the behaviors of the bullwhip effect with respect to different model parameters are both analytically and numerically explore