Amazon Same Day Delivery - Last Mile Economics and Challenges

Amazon Same Day Delivery - Last Mile Economics and Challenges

Supply Chain Digest Blog Entry: October 15, 2013 (here)

Last month I wrote about how close Amazon is towards achieving a same day delivery distribution model based on its current fulfillment network.  What a difference a month makes!  In the last 30 days, the company has announced two new mega fulfillment centers for 2014.  A new facility in Kenosha, WI will service Milwaukee and Chicago; and a new facility in Ruskin will service the Tampa, St. Petersburg and Lakeland areas in Florida.     

In this month's entry, I thought that it would be interesting to investigate the challenges associated with same day delivery and the concept of the last mile. For the sake of this discussion, let’s start with an example:

• Let us say that a customer in New York City orders a product from a fulfillment center near Louisville, Kentucky.  Furthermore, let us say that the parcel weighs 5 pounds and that it is shipped UPS Ground with regular 2-day delivery service.
• The list cost to ship the parcel is $8.81 and with negotiated discounts let us say that the discounted cost is closer to $7.00.  This covers the expense of putting the parcel onto a local truck in Louisville, processing the parcel through the UPS Worldport sortation center, transporting the parcel to the local UPS hub near New York City where it is then sorted by delivery route and delivered to the customer. 
• With a local Amazon fulfillment center positioned near New York City (e.g. Robbinsville, NJ opens in 2014), much of this handling and transport activity is eliminated therefore one would assume that the cost per delivery is significantly cheaper.  But this can be a dangerous assumption to make.  The cost of the last mile can be very volatile and it is highly dependent on what I refer to as “hit density” on the delivery van. Hit density refers to the average number of minutes between deliveries on the outbound route.

First off, in logistics vernacular, we refer to the last mile as the distance between the fulfillment center and the customer's doorstep.  With a same day fulfillment center positioned close to a major metropolitan market, customer orders received before a certain cut-off time can be picked, packed and shipped to the customer’s doorstep within the same day.  Amazon has designed their fulfillment centers to handle same day order fulfillment from a pick/pack ship perspective. However, the outbound delivery of the last mile has real challenges associated to it which we will discuss using some representative data and assumptions below to illustrate this point.

As a starting point, let us define some basic operating expense assumptions for the parcel shipper performing the last mile delivery.  Our firm pulled the figures in the tables below by studying actual financial and operating data for a major parcel carrier.

Next we state our assumptions regarding the operating data associated with an outbound route:

Finally we analyze the cost per parcel delivery based on 4 different scenarios as follows:

1. Baseline Scenario: Fulfillment center is positioned 10 miles away from the epicenter of the delivery network being serviced (i.e. the driver must drive an average of 10 miles to reach the zone where the customer deliveries will be performed).
2. Scenario 2 – Fulfillment center is 25 miles away from customer delivery zone.
3. Scenario 3 – Fulfillment center is 50 miles away from customer delivery zone.
4. Scenario 4 – Fulfillment center is 75 miles away from customer delivery zone.
5. Scenario 5 – Fulfillment center is 100 miles away from customer delivery zone.

From the analysis above we learn the following:

1. The further we locate the fulfillment center away from the epicenter of the delivery network, the more expensive the cost per delivery becomes.  For example, the package that costs $4.05 to deliver from a 10-mile distance becomes $13.09 from a 100-mile distance.
2. The average minutes between deliveries has a huge impact on the cost per delivery.  Notice that the parcel cost of $4.05 applies when there is 4 minutes between deliveries.  Change this to 7:24 minutes between deliveries and the same parcel cost is $8.53 – a cost increase of 110%.
3. Notice that the cost per delivery has less to do with the weight of the parcel since the cost relates more to the hit density on the route – i.e. the number of deliveries that you can get onto the 24’ delivery van.  However, parcels with larger cubes will reduce the number of packages that can be placed on the van. Therefore larger cube packages will reduce the deliveries per route which in turn drives up the cost per delivery.  Hence we can conclude that cube plays a more important role than weight in defining the cost of a same day delivery.

In the case of Amazon, clearly the challenge will be to increase the hit density on the delivery truck to keep the last mile delivery to a figure that is below the cost of shipping the same parcel from a fulfillment center that is far away.  Returning to our opening example of the $7.00 package that was shipped from Louisville to New York City, if that same parcel is shipped same day from a local fulfillment center that is positioned 75 miles away from the customer delivery network, then the cost to deliver the package jumps to $8.53 which is a very significant cost increase.  In other words, depending on the circumstances, it could actually be less expensive to ship the package from 745 miles away than from 75 miles away!

Of course the same day delivery concept is all about raising the bar for customer service to a level where Amazon can continue to dominate the e-commerce retail market into the future.  Our point is that the logistical and economic challenges to support the last mile delivery for same day delivery are in fact quite challenging.  The risks are high if (a) the facilities are not positioned close enough to market and (b) the deliveries per outbound route are not dense enough. 

Conclusions

Based on the discussion points raised in this article, we conclude the following:

1. We predict that Amazon will deploy major fulfillment center locations within a 25 – 50 mile radius of the top 20 major metropolitan markets in the US.   Once these markets are conquered by 2014-2015, smaller footprint fulfillment centers can be deployed close to secondary metropolitan markets with smaller populations.  This strategy is already underway in the UK.
2. Amazon.Fresh requires a completely new distribution infrastructure because of the cold temperature storage requirements for perishables merchandise.  In Seattle, Amazon.Fresh charges $7.99 for a grocery delivery with a minimum order size of $50.  This cost structure is reasonable in that it covers the cost of delivery to the customer.  As Amazon.Fresh rolls out to other markets, the consolidation of food and non-food orders will help to increase the total number of daily deliveries to be serviced, therefore the  hit density of outbound routes are improved.  In other words, as the average minutes between deliveries is reduced, the cost per delivery is also reduced which is why Amazon.Fresh is really a Trojan horse designed to improve the economics associated with the same delivery concept.
3. As Amazon transitions to a same day delivery model, the company will need to ensure that the last mile transportation economics make sense.   By leveraging existing couriers that already have delivery volume within metropolitan market areas, the hit density on outbound delivery vans is improved for both the courier and for Amazon due to the consolidation of volume.  As Amazon’s volume picks up in a given market area, it may eventually develop its own delivery fleet to service the market rather than rely on third party carriers.  It is difficult to predict where this is going because margins are relatively low in the parcel delivery business (e.g. UPS domestic operating margin in 2012 was 1.42%).  We believe that it is most probable that Amazon will steer towards a model where third party courier services are leveraged to service metropolitan markets rather than setting up dedicated private or third party fleets. Having said this, since is the most vulnerable part of the supply chain in terms of breakage, Amazon may ultimately choose to control the final mile to ensure deliveries arrive on time as promised.

Marc Wulfraat is the President of MWPVL International Inc.  He can be reached by clicking here.  MWPVL International provides supply chain / logistics network strategy consulting services. Our services include: supply chain network strategy; distribution center design; material handling and automation design;  supply chain technology consulting; product sourcing; 3PL Outsourcing; and purchasing; transportation consulting; and operational assessments.