Impact of inventory storage and retrieval schemes on productivity

Abstract

The operational management of high volume, multi-line distribution warehouses is a monumental undertaking, which only a handful of companies in the world have chosen to tackle. Amazon.com is amongst the few, and has further differentiated itself because of its direct to customer method of distribution and complex order mixes. There is no other retailer that carries and directly delivers as many different products (over 4 million different unique items) in as wide range of product categories (from music to cosmetics to electronics to garden hoses) in as high of volume as Amazon.com. The nature of Amazon's retail model and its organic growth over the past decade has made its fulfillment centers a complex beast to decipher. Decisions on the fulfillment center floor are composed of intricate balances between demand constraints, equipment bottlenecks, storage limitations and labor costs, making the true cost associated with each variable dependent on every other variable. The goal of this thesis is to document a practical exploration of inventory storage and retrieval schemes and its relationships to productivity (and subsequently cost), as well as identify implementable changes that yields higher throughput, lower lead time for order fulfillment, and ultimately dollar savings. Of particular interest are operationally transparent process changes, which improve processes in a manner that minimize impact on the fulfillment center floor. This concept will be the central theme of all recommendations resulting from this thesis.