right_division Green SCM Distribution
Bookmark us
SCDigest Logo

SCDigest Expert Insight: Keep It Moving

About the Author

Marc Wulfraat


MWPVL International, Inc.

Marc Wulfraat is the president and founder of MWPVL International, a supply chain and logistics consulting firm.  Marc has 27 years of supply chain consulting experience across a variety of industry sectors and countries. His expertise is in supply chain strategy, facility design, material handling systems, automation, and supply chain execution technologies. He has managed many complex consulting mandates to help a diverse range of companies with their supply chain challenges. For more information, please visit

By Marc Wulfraat

September 3, 2013

How Close to Reality is Amazon Same Day Delivery?

A Look at How Many Fulfillment Centers are Needed for Amazon to Achieve Same Day Delivery of General Merchandise to the Top 20 Cities

For the past few years, Amazon has been on a relentless pursuit to reduce the time that it takes to deliver customer sales orders.  The company has invested hundreds of millions of dollars into distribution centers positioned close to U.S. major metropolitan markets to enable shipments to be delivered in hours rather than days.  Ultimately, time will tell if this strategy is a game changer that will revolutionize the e-commerce market as we know it.  For the moment, we thought it would be of interest to understand how close Amazon is towards achieving their same day delivery goal within the U.S. market.

Wulfraat Says:

Ultimately, time will tell if this DC strategy is a game changer that will revolutionize the e-commerce market as we know it.
What Do You Say?

Click Here to Send Us Your Comments
Click Here to See Reader Feedback

So the question is how close to reality is Amazon Same Day Delivery?  To answer this question with a reasonable degree of accuracy, we need two importance sources of data:  (1) Census information on U.S. population statistics by major metropolitan city and (2) Information on Amazon's existing fulfillment network.  We also need to make an assumption about the maximum radius of distance that is appropriate for a same day fulfillment center.  We have census data from 2010 and a detailed listing of all Amazon fulfillment centers which is available here.  For the purpose of this article we will assume a 100 mile radius as being the maximum distance around the fulfillment center to realistically achieve same day local delivery service.  Why?



If a driver can travel down a highway at an average speed of 50mph then 4 hours of time is needed to travel the return distance to/from the trading area perimeter which leaves say 4 hours to perform actual customer deliveries and 2 hours for break time, yard time, etc before a 10-hour work-day is reached.

  2. The "cost of the last mile" discussion is crucial to the concept of same day delivery. There needs to be a minimum "hit density" on a driver's route to achieve a low cost per delivery. We'll talk about the last mile in a future blog but suffice to say that if the radius of distance gets larger than 100 miles then we spend more time driving and less time delivering which increases the cost per shipment and penalizes the viability of the same day service proposition. In fact, I am certain that many experts would challenge our 100-mile assumption and suggest that a 50-75 mile radius is a better assumption to use so we are probably being conservative.

Let's take a look at the 2010 census data for the top 20 cities in the U.S. sorted in descending population sequence to get the ball rolling.   In the table below the population statistics reflect the number of people living in the greater metropolitan region of each city including the surrounding areas.  At the bottom of the table we can see that the top 20 cities total nearly 117 Million people  which accounts for 37% of the 2010 U.S. population of 316 Million people.

Columns by Marc Wulfraat

Locus Robotics - An Independent Consultant's Review Part 2

Supply Chain Comment: Locus Robotics - An Independent's Consultant Review Part 1

Keep It Moving: What Economic Impact Will the Amazon Sortation Network Have on UPS and FedEx?

Keep It Moving: The AmazonSupply Threat to the B2B Distribution Market

Supply Chain Comment: If You Thought that Amazon was a Game Changer in 2015, Hold onto your Hat

Logistics Comment: Amazon is Building a New Distribution Network - Quickly and Quietly

Keep It Moving: A Tale of Two Retailers - Amazon and Target

Keep It Moving: How the "Amazon Effect" is Changing the American Manufacturing Industry Forever

Keep It Moving: Should US Companies Think More Like the Europeans When it Comes to Automation?

Keep It Moving: Amazon Same Day Delivery - Last Mile Economics and Challenges

Keep It Moving: How Close to Reality is Amazon Same Day Delivery?

Keep It Moving: E-Fulfillment Wars - Will eBay Over-Promise and Under-Deliver?

Keep It Moving: Looking to Increase Warehouse Storage Capacity?

Keep It Moving: Goods to Person Automation - A Multishuttle Primer

Keep It Moving: Global Age Dependency Ratios (ADR) Will Drive Increasing Automation Investments

For each of the top 20 cities, we identify the nearest Amazon fulfillment center and its approximate distance to the epi-center of the metropolitan area.  As an example, the largest U.S. metropolitan population is New York City at  19 Million people which includes surrounding cities in NJ, CT, NY and PA.  The nearest Amazon fulfillment center to this megalopolis is in Breinigsville, PA where two Amazon facilities are located (ABE2 & ABE3).  The distance from these facilities to the center of Manhattan is 104 miles which exceeds the 100 mile maximum radius assumption that we defined earlier on, hence we state that Amazon is not currently in a position to provide same day service to New York City.  This will of course change if the company starts up distribution operations in Woodbridge, NJ but at the moment, based on our 100 mile assumption, we categorize New York City as a city that is not same-day ready.

Within the list of top 20 cities, we have identified 8 urban centers where Amazon is already within a 100-mile radius as of August 2013 and 12 cities where they are not positioned to provide same day service. The one exception to our 100-mile rule is Washington DC which is 95 miles from the Amazon fulfillment center in Middletown, DE but the reality is that the southbound and northbound traffic along the I95 is horrific and it would be futile to service Washington DC from this location on a same day basis.

Thus from the table above, we might conclude that Amazon, in its present state, has a distribution infrastructure to reach 44.6 Million people or 14.1% of the US population with same day delivery service.  In fact the percentage is slightly higher because Amazon also has existing general merchandise fulfillment centers positioned nearby: Indianapolis, IN; Reno, NV; San Antonio, TX; Chattanooga, TN; Nashville, TN; Louisville, KY; Cincinnati, OH; Columbia, SC; Richmond, VA; Harrisburg, PA; and Allentown/Bethlehem, PA.  These cities total 4.3 Million in population hence Amazon's same day reach is currently closer to between 49 - 50 Million people or 15.5% of the population.

Thus to be in a position to service the top 20 cities that account for 37% of the US population, Amazon needs to add a minimum of 12 fulfillment centers to its existing in-progress distribution network of 53 fulfillment centers in the U.S. for a total of 65 buildings nationwide.  The table above illustrates the cities where we can most likely expect these additions to take place.  Assuming that the average new fulfillment center size will be in keeping with the facilities that have been added over the past couple of years (i.e. 1,000,000 sq ft at $110 per sq ft)  Amazon will need to add a minimum of 12 Million square feet at a price tag of $1.3 Billion.  In reality, the figures will likely be substantially higher as some cities will need more than one facility and the buildings are getting bigger and more expensive over time.  Most importantly, this forthcoming spending spree does not even begin to address the distribution infrastructure requirements of Amazon.Fresh since this will require a completely parallel set of distribution facilities that are capable of supporting perishables distribution.

Final Thoughts

In conclusion, a minimum of 12 more fulfillment centers is needed for Amazon to achieve same day delivery of general merchandise to the top 20 cities that account for 37% of the US population. In a forthcoming blog we will discuss the economic challenges associated with last mile delivery and why being closer to market does not necessarily translate into lower outbound transportation costs in the world of e-commerce.

Recent Feedback

Your entire model is based on the assumption that a work day of 10 hours consists of a single dispatch, 4 hours of driving, 4 hours of delivering and two hours of break. These are valid assumptions when operating a TL or LTL dispatch system, but not valid when tendering shipments for residential delivery when ORDERS CAN BE PLACED UNTIL NOON...or perhaps later.

Once ordered, the product has to be picked, routed, loaded and dispatched...which may place the actual origin dispatch at 1 or 2pm. Most retailers consider (no later than) 9pm as same day - which gives the delivery operator only 7 hours for all of their driving and delivery functions. As you can see, that will dramatically shorten the operating window for true SAME DAY delivery..and likewise effect every number associated to population or distance in your study.

Secondly, you need to consider a few more assumptions. The average metro driving speed is actually closer to 35mph than to 50mph. Although the linehaul segment may run closer to 50 (assuming a remote DC location...which is not what Amazon is now building) the rest of the stem miles will be driven at a much slower speed giving the delivery driver a much lower MPH average. Also, using your 4 hour delivery window assumption, you need to calculate the required stops per work hour to make the required number of $5.99 deliveries (Amazon Prime cost) to pay for the entire 10 hour day's operating expense. This includes a driver wage of $12-15/hr, equipment cost, fuel, tolls..delay...etc.

I think you will find that even if you can reach the centroid of a population within four hours, completing enough stops in a 4 hour period to cover the entire day's operations cost will be difficult, and perhaps impossible at such a low revenue rate. Most courier models depend on only paying for the variable cost portion of the delivery rate (unused capacity within courier networks). The problem is, unused capacity is a fixed commodity when only paying the variable rate. At some point when the offering expands and has a higher demand, someone has to buy another truck/delivery vehicle. At that point the entire "unused capacity" model breaks down until revenue is increased to meet the total costs of the delivery operation!

Keep an eye out for the Shipt brand in coming months. We believe that we have a better solution....and certainly one that is better than what Amazon is trying to build.

Jim Kitz
Shipt LLC
Oct, 03 2014