Managing multiproduct supply chains with two supply paths

Joo Yol Maeng, Purdue University

Abstract

In this dissertation, we study multiproduct supply chain management problems with two supply paths in the context of a leading wine bottle manufacturer. The company seeks to provide a high level of customization and service for its customers (vineyards or bottlers). It operates under the make-to-sales-order system, based on sales orders from customers. However, the highly volatile nature of demand in the wine industry often creates a huge discrepancy between sales orders and actual demand, which results in frequent order cancellations. The company wanted to reduce inventory while maintaining the current variety and high customer service levels – a necessity for customer satisfaction in the industry. The company has over 370 distinct wine bottles which go into more than 3,800 customized cases called UPIDs. At the end of production lines, glass bottles can be packaged either in UPIDs (customized bottles) or in bulk (bottles in standard containers) on pallets. There are three parts in the thesis, and each of them is motivated by the company's supply chain management problems. In the first part, we develop an optimization model to make the optimal decision to store a selection of products having poor demand forecast in bulk to seek the benefit of a pooled replenishment system. We present comparative statics results, and a heuristic that helps identify characteristics of products potentially move to bulk. In the second part, we bring the problem context and intuition of the first part and study under a stochastic inventory model. We present the close-to-optimal cutoff value to make the optimal bulk-to-case decision, and reconfirm the comparative statics results from the previous part. In the last part, we use empirical analysis to study the company's current repack process that minimizes backorders and improves its customer service level. We also generate various repack heuristics in order to find the policy that is the closest match to the company's actual repack practice.

Degree

Ph.D.

Advisors

Iyer, Purdue University.

Subject Area

Management

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS