Research On Inventory And Pricing Decisions Considering Dual Replenishment Modes And Reference Price Effect

Inventory and pricing decisions are two very important aspects for operations man-agement of enterprises.With the rapid development of supply chain globalization and network information technology,traditional supply chain structure and people's con-sumption patterns have changed,which makes the traditional inventory and pricing strategies of enterprises unsuitable.In the context of the new era,the research on the optimal inventory and pricing strategies of enterprises has attracted wide attention from the industry and academia.Supply chain globalization makes unpredictable factors in-crease,which increases the risk of supply chain interruption,and dual replenishment modes is a very effective measure to reduce supply chain interruption.In addition,the rapid development of network technology makes consumers can easily obtain the his-torical price of products,and convert it into the reference price of products through some forms(e.g.,peak-end rule and exponential smoothing process).Clearly,it is nec-essary to consider the inventory strategy under the reference price effect in the context of information era.Although vast literature has studied the inventory and pricing problems under the dual replenishment modes and reference price effects,there are still many problems unsolved.On the basis of the existing research,this dissertation takes into account the factors that are more in line with the realistic background and derives the optimal or near-optimal inventory and pricing strategies through mathematical modeling,which provides guidance for the operations management of enterprises.Specifically,the core work of this dissertation is embodied in the following aspects:Firstly,based on the asymptotic optimality of tailored base-surge policy without considering the fixed cost,a modified tailored base-surge(M-TBS)policy with fixed ordering cost and ordering cycle is studied in this dissertation.The near-optimal param-eters of M-TBS are obtained by using policy optimization technique.The numerical experiments are conducted to compare M-TBS policy with the benchmark policy,i.e.,dual-index(S,R,nQ)policy.The results show that when the fixed cost of the regular replenishment channel is low,M-TBS policy is better than dual-index(S,R,nQ)policy only when the lead time difference between two replenishment modes is large.How-ever,when the fixed cost of the regular replenishment mode is high,M-TBS policy is obviously better than the(S,R,nQ)and the traditional constant ordering(TBS)policy,even though the lead time difference between the two delivery modes is small.Secondly,based on the existing literature on multi-echelon inventory system with dual sourcing is few,this dissertation considers a stochastic serial inventory system with dual replenishment/delivery modes for each echelon stage,and the lead time difference between two delivery modes is general,not a simple "0-1" combination.Furthermore,this dissertation considers the case with batch orders.A dual-index echelon-(R,nQ)policy is proposed,and a simulation-based optimization method is given to solve the optimal policy parameters.The theoretical result shows that the optimal expedited target level of the most downstream echelon can be computed by a newsvendor fractile.A heuristic is then provided to find the near-optimal policy parameters,which is shown to be very efficient by the numerical studies.We also numerically investigate the value of the dual replenishment modes and observe that as long as the regular delivery mode of the upstream echelon is only marginally less favorable than the regular delivery mode of the downstream echelon,the downstream echelon should be given priority by firms thinking of adding an expedited mode to the system.Thirdly,based on the phenomenon of reference price effect in the era of network information,this dissertation considers the joint inventory and pricing decisions with reference price effect and loss-averse consumers.The retailer sells a single product to loss-averse consumers within a finite planning horizon and discusses the influence of reference price on the retailer's optimal pricing and inventory strategies.Customers'demand is determined by their purchasing utility through a Multinomial Logit(MNL)model,and the purchasing utility depends on the reference price and the current sales price.We first give the optimal myopic policy for the retailer in a single-period.Next,we perform an equivalent transformation on the model such that the single-period rev-enue is jointly concave,which is critical to characterize the optimal policy in multi-period.Dynamic programming is then used to explore the optimal policy for the retailer.Interestingly,we find that a reference price-dependent base-stock list-price policy is proved to be an optimal policy in each period.The main innovations of this dissertation are summarized as follows:(1)The asymptotic optimality of TBS policy in a dual sourcing inventory system without fixed cost is modified and extended to a dual sourcing inventory system with fixed cost and ordering cycle;(2)A heuristic policy is proposed for the multi-echelon dual sourcing in-ventory system.This is the first that considers a stochastic serial inventory system with two delivery modes in each echelon,and the lead time difference between two delivery modes is general;(3)Simultaneously consider the impacts of the customers' reference price effect and loss-averse behaviors on the joint inventory and pricing decisions of the retailer,and the customers' demand is determined from their purchasing utility through an MNL model.