Decision-making Models Of Ordering-Pricing For An Assembler Under Random Supply And Demand

In the fiercely competitive industry of manufacturing consumer electronics, products' life cycles are becoming shorter, industry profit is dropping, demand is oftenly random, and supply risks are increasing. Under such circumstances, an assembler's business success depends on his/her ability to launch new products to meet the market and to match supply with demand. Ordering and pricing decisions play key roles in surviving and growing for an assembler.Consumer electronics products usually consist of hundreds and thousands of components, and such somewhat fashionable products must contain some custom-made components that may become useless and then idle inventory if demand forecast is inaccurate. Moreover, such custom-made components are so special that they are hard to make, hence they often face random supply. Unfortunately, the assembler needs every component to assemble the final product. Therefore, the consumer electronics assemblers are often confronted with not only random demand, but also uncertain supply. In this case, it is important to study optimization models of ordering and pricing decisions of assemblers facing random demand and supply, so as to match supply with demand and improve their competitiveness.We consider a consumer electronic assembler who faces random supply and demand, with the objective of maximizing the assembler's expected profit, this thesis builds non-linear stochastic programming models of multi-dimensional decision variables, analyzes and solves optimal components ordering policy and optimal pricing of the final product. This thesis includes the following aspects.(1) The model of ordering for the assembler under random supply and demand is built. A simple system of two components is first studied. The optimal ordering policyis characterized and the effect of uncertain supply is investigated. The model is then extended to consider n components and additional assembling quantity decision and the extended model is decomposed, taking advantage of the problem characteristics. Finally, the optimal ordering policyand planned assembling amount are given in a systematic way.(2) The model of joint component ordering and final product pricing for the assembler under random supply and demand is built. Based on the ordering models, we first consider the case of deterministic price-dependent demand and uncertain supply, then the case of uncertain price-dependent demand and uncertain supply. After the n+1-dimensional model is reduced to be of two dimensions (i.e., an unified ordering decision and a pricing decision), we fix one decision and derive the corresponding optimal path of another decision, and then we solve the global optimal solution along the path.(3) A serial of numerical experiments are designed to inspect the effect of random supply and demand on the optimal ordering-pricing decisions. For the random supply, the increase of mean of supplier's capacity is related to investments in new production line or promises of overtime work, the decrease of variance of supplier's capacity is related to enhancements of preventive maintenance for manufacturing equipments or enhancements of training and education for workers to avoid mis-operations, etc., the simultaneous increase of mean and variance of supplier's capacity is related to supplier outsourcing its work to multiple second-tier suppliers. Based on these relations with practice, we come up with logical explanations and offer managers useful suggestions and insights. Eventually, for the case of a phone producer, we abstract the realistic problem to compare our method with present method in terms of solution quality and applicability.This thesis is a beneficial exploration for the conundrum of ordering and pricing under random demand and supply. The models built and the methods used both enrich the theoretical research in the field of Operation Management under uncertain environment. The conclusions of the optimal decisions and the relative analysis bear certain significance of managerial insights for producers producing consumer electronics or products of short life-cycle.