||Competition between supply chains has become a major battleground in the 21st century. There are four types of supply chain designs: make-to-stock, make-to-order, engineer-to-order and assemble-to-order. Assemble-to-order (ATO) consid-ers the trade-off among factors such as the lead time, price, and personalization and becomes increasingly the preferred supply chain design. This thesis focuses on two important issues in managing an ATO system: the dynamic quotation of price and lead time and the evaluation of the average order-based backorders. First, we formulate a semi-Markov decision process (SMDP) model to address the joint and dynamically quote a price and lead time to an arriving customer so as to maximize the firm's expected profit. The customer's acceptance prob-ability is a function of the quotation. Under some general assumptions on the acceptance probability function, we show that this difficult SMDP problem can be transformed into a single-variable problem so that the structure of the opti-mal quotation is identified and an efficient and exact algorithm is developed for this quotation. Second, the average order-based backorder is another important measure in an assemble-to-order system. A key difficulty of evaluating it is to know the joint inventory positions distribution. This thesis presents a general and efficient sufficient condition for the distribution to be uniform, and also gives out a necessary and sufficient condition. Finally, the thesis introduces a con-cise and explicit method to develop a closed form expression for exactly average order-based backorders.