A model for analyzing multi-channel distribution systems

Aydin Alptekinoǧlu, Christopher S. Tang

Research output: Contribution to journalArticlepeer-review

108 Scopus citations


To capture customers in different market segments, many firms distribute their products using multiple channels involving in-store, mail-order and electronic mediums. Nevertheless, the business literature is replete with examples where the management of multiple distribution channels becomes unwieldy. As a way to evaluate different distribution strategies arising from multiple channels, we develop a model of a general multi-channel distribution system subject to stochastic demand. Since the exact analysis of our model is intractable, we propose a decomposition scheme that enables us to develop a near-optimal distribution policy with minimum total expected distribution cost. We utilize our analysis of this multi-channel system in two specific applications arising from multi-channel distribution. The first. application deals with a store-based retailer expanding into a direct sales channel. The retailer can fill the orders generated from this additional sales channel by using the existing stores or the existing depots. For this application, we evaluate the trade-off between these two options. The second application intends to find a cost-effective way to distribute products by utilizing the combined distribution system of two retailers after a potential merger-and-acquisition (M&A). For this application, we quantify the benefit of an M&A in terms of distribution efficiency.

Original languageEnglish (US)
Pages (from-to)802-824
Number of pages23
JournalEuropean Journal of Operational Research
Issue number3
StatePublished - Jun 16 2005

All Science Journal Classification (ASJC) codes

  • General Computer Science
  • Modeling and Simulation
  • Management Science and Operations Research
  • Information Systems and Management


Dive into the research topics of 'A model for analyzing multi-channel distribution systems'. Together they form a unique fingerprint.

Cite this