BSR logo

Collaborative Initiatives | Future of Fuels | Case Study Library

Amazon.com: EV Case Study

Amazon.com: EV Case Study Overview

This case study, provided by Amazon, highlights how electric performed in the context of a six month test carried out in France, Italy, and Spain with multiple electric vehicles (EV) delivery vans.

Part 1: Baseline Fleet and Technology

This section describes the characteristics of the existing technology in the fleet (i.e. prior to the test). This provides information on the fleet characteristics, duty cycle details, and region of operation for the technology being tested.

Company Name Amazon.com, Inc.

Vehicle Type Mixed fleet, class 1, class 2 urban delivery vans

Fuel Type Mixed, diesel and gasoline

Feedstock Public fueling stations

Refueling Over the road

Average Vehicle Miles Traveled (VMT) 28,000 kilometers/year

Hours of Operation 2,800 hours

Average Load 100 kilograms/220 pounds

Maximum Load 300 kilograms/660 pounds

Length of Haul 75 kilometers/46 miles

Return to Base? Yes

Countries France, Italy, and Spain

City or Regions Paris, Milan, and Madrid

Lifespan 5 years

Description Local delivery routes are comprised of stem distance from the delivery station to the delivery zone, low speeds and high-frequency stopping while on zone, then return to base.


Part 2: Technology and Test Purpose

This section describes the type of technology tested and primary reasons behind the test.

Technology Tested Electric delivery vans.

Test Purpose Amazon has a long-term goal to power our global infrastructure using 100 percent renewable energy, and we are making solid progress. With improvements in electric vehicles, aviation bio fuels, reusable packaging, and renewable energy, for the first time we can now see a path to net zero carbon delivery of shipments to customers, and we are setting an ambitious goal for ourselves to reach 50 percent of all Amazon shipments with net zero carbon by 2030. We are calling this project "Shipment Zero”. The purpose of electric delivery van testing is to enable the transition of a growing share of Amazon’s delivery fleet from conventional internal combustion vehicles to cost-effective electric delivery vehicles. We are testing a variety of vehicles, charging technologies and operational models to develop solutions that meet our customer delivery requirements.


Part 3: Test Parameters

This section describes the actual parameters used to test the new sustainable technology, including vehicles tested, testing timeline, additional training and infrastructure requirements, etc.

Test Standard Class 1 delivery EVs were deployed in active commercial use in three countries and over a wide range of operating temperatures. All vehicles were equipped with telematics, providing a common monitoring platform across the four different electric vehicle models evaluated. The wide range of real-world variability (route characteristics, ambient temperature, geographies, driver behavior, and traffic patterns) enabled a thorough evaluation of the real-world performance of different electric vehicles. Particular focus was applied to the real-world range in diverse operating conditions. Ongoing testing is quantifying the impact of each of those operating conditions on range.

Test Dates The test results were gathered between June and December 2018.

Total Miles Tested Over 20,000 miles.

Total Hours Tested Over 6,000 hours of delivery operations with over 2,000 hours of active vehicle operation.

Average Load Tested Vehicles were operated in active commercial use, including standard payloads associated with last-mile delivery. Exact payload mass was not measured as part of the testing.

Maximum Load Tested Given the real-world application, a deliberate maximum load test was not completed. Vehicles were operated under normal last-mile delivery loading, which cubes out before it weighs out in these small delivery vehicles.

Time to Fuel Charging time varies heavily as a function of operating conditions and charging infrastructure. While actual values varied heavily, the average usage with standard level 2 depot charging infrastructure for vehicles with this size battery (<45 kWh) would result in an average charging time of 3-4 hours.

Testing Barriers Data parameters frequently have different definitions between manufacturers and can vary between models from a given manufacturer. For example, some electric vehicle models report absolute State-of-Charge while others report usable State-of-Charge. Another common variation between models are the reported energy consumption values, which are measured at different points in the electric powertrain for different models. Given the importance of having comparable units of measure across all vehicle models, these various units must be normalized into a common unit. In this study, a third party telematics system was utilized to provide consistent units of measure across all makes and models.


Part 4: Supporting Services

This section describes the additional supporting services needed for the sustainable technology tested, including details on fuel type, infrastructure requirements, and personnel training.

Fuel Type Electric (EVs)

Feedstock Not applicable.

Level of Readily Available Infrastructure Medium

New/Special Infrastructure Requirements For the number of vehicles involved in this testing, infrastructure was readily available. However widespread electric vehicle deployment will result in expanded infrastructure requirements.

Special Training Requirements Beyond basic training of plugging in the vehicle and initiating a charge session, no special training was provided.


Part 5: Operational Performance

This section describes the key metrics used to measure operational performance of the alternative fuel or technology, benchmarked against the current technology used in the fleet.

GHG Emissions Reduction % 100% reduction tailpipe emissions.

Air Quality Emissions % 100% reduction of NOx emissions.

Fuel Economy (MPDGE) While the vehicles had sufficient real-world range to complete the daily operations on the routes selected, the real-world range was lower in all cases than the values reported on standardized test cycles. Real-world range, when averaged over a month of operation, was at least 19% lower than the standardized test cycle results. Given the operating conditions and payload mass, it was anticipated that real-world range would be less than standardized test cycle results. Of note is the significant variability in real-world range in different operating conditions. Real-world fuel economy, when averaged across a month of operation, ranged from 26.2 to 85.3 kWh/100km (0.42 to 1.37 kWh per mile). Those values correspond to 44 to 145 MPDGE.

Driver Satisfaction Driver feedback is mixed. Some enjoy the new technology, driving cleaner vehicles and having easier access to parking and roads in restricted areas and congestion zones. Others take a few trips to get used to regenerative braking, and some have experienced range anxiety, afraid to use the heating in winter for fear of draining the battery. But for the most part, vehicle operation is very similar to conventional delivery vans.

Special Training Requirements Basic vehicle operating instructions were provided; however, there was no training or coaching for the drivers to modify their operating habits in ways that would increase efficiency or range.

Special Maintenance Requirements While a suggested benefit of electric vehicles is a reduced level of maintenance (i.e., oil changes), maintenance benefits were not a primary objective of this testing and was not actively tracked. The telematics system provides the necessary data to track battery health, however the test period was not long enough to measure battery degradation. These items will be tracked as part of future, longer-term testing as the telematics system provides the necessary data to track battery health.

Additional Benefits for Fleet A number of jurisdictions have deployed, or are deploying, low emission zones. In these zones, the use of diesel or gasoline vehicles may result in a fee or may not be permitted at all. Accordingly, the use of electric vehicles in these zones can reduce access costs and/or increase access to customer delivery locations.

Additional Challenges for Fleet First, there is a need to manage EV routing differently than we have in the past for conventional vehicles that can fuel anywhere mid-route in only a few minutes. Electric delivery vans will need to be assigned to routes we are confident they can complete without causing range anxiety for our delivery partners. Second, as the number of electric vehicles increases at a specific depot, peak power demand can become costly and active charge management will be required. Any charge management solution must ensure the vehicles are adequately charged by the start of daily operation.


Part 6: Financial Performance

This section describes company’s expectations for financial and economic performance of the technology, benchmarked against the incumbent fuel/technology. Where noted, minus (-) is savings and plus (+) is additional costs for the fleet.

Fuel Premiums/Savings Percentage We continue to experience more than 50% savings in per route energy costs in the EU. The energy savings are highly specific to local fuel and electricity pricing. And given the impact of time-of-use electricity rates in many jurisdictions, the time of day that vehicles charge will also have a significant impact.

Maintenance Premiums/Savings Percentage Maintenance data is not available. While maintenance savings are anticipated from electric vehicles, this was not a specific focus of this testing.

Capital Premiums/Savings Percentage Vehicle procurement is currently estimated at 20 to 50 percent premium in Europe, before factoring in any jurisdictional incentives. Today, upfront premiums are even higher in the U.S. The longer we use electric vehicles in terms of distance and years, the better our ability to recuperate higher upfront costs through operational savings. Longer term testing is required to validate total maintenance costs, and the number of years required to achieve a return on investment.

Estimated Residual Value in US$ Currently estimated at 0-20 percent of capital costs after five years. Residual value remains an area of uncertainty for electric vehicles, especially commercial electric vehicles. Analyzing battery health is possible through the telematics system used and will be monitored as an input to residual value determination.

Did You Use Subsidies? No


Part 7: Conclusions

Will You Include this Technology in Your Fleet? Yes

Additional Comments Yes, Amazon intends to collaborate with our delivery service providers to include an increasing number of electric delivery vehicles in our last mile fleet on the road to Shipment Zero. Last mile delivery is an application well suited to fleet electrification, and we expect the higher upfront costs to come down in the coming years.

Would You Recommend this Technology for Other Fleets or Applications? Yes

Summary of Findings Multiple electric vehicle models were operated and monitored in real-world delivery fleet applications. Spanning multiple countries and wide temperature variations, the actual vehicle range was tested in real-world conditions. The vehicles were able to complete the daily routes. However, real-world vehicle ranges in our duty cycle were a minimum of 19 percent lower than the published range of these vehicles in standardized test cycles. We also saw significant additional range loss in colder months. Real-world range being lower than range reported from standardized testing was expected due to the payload mass and route characteristics of delivery applications. These results highlight the need to understand and quantify the impact of specific route characteristics on real-world range, and how multiple factors combine. For instance, it was found that routes that have frequent stops and lower distance between stops result in a higher temperature sensitivity to real-world range. Routes with less stops and longer travel distances between stops are less impacted by ambient temperature. Work continues on quantifying these impacts and benchmarking the performance of new electric vehicles being brought to the market.



The information contained on BSR’s Future of Fuels Case Study Library website is for general information purposes only. We recognize that companies may choose to use the information provided to make critical business decisions, therefore BSR and its partners endeavor to provide the best available data.

However, BSR and its partners make no representations or warranties of any kind, express or implied, about the completeness, accuracy, reliability, suitability or availability with respect to the information or related graphics contained on the website for any purpose. Any reliance you place on such information is therefore strictly at your own risk.

Additional Solutions

Sustainable Fuel Buyers’ Principles

To meet joint objectives of economic growth and emissions reductions, operators will need to leverage a diverse, often confusing, set of existing and emerging solutions. These Principles provide a robust framework to catalyze lasting change to the road freight fuel system.

Learn More


Fuel Sustainability Tool

Today, more people are expecting vehicles and fleets to run on cleaner, more sustainable fuels—but which fuels are clean? And what makes them sustainable? Our Fuel Sustainability Tool simplifies sustainability for fleet owners and provides credible, comparable data for fleet planning and fuel purchasing.

Register below to download the Fuel Tool (Excel format) and Reference documentation.

Download


Fuel Sustainability Briefs

Developed in 2015, the Fuel Sustainability Briefing series gives practitioners information to develop practical approaches that accelerate the transition to low-carbon fuels by summarizing our research on the market outlook, key issues, and sustainability potential for five fuel categories.

Learn More