The document summarizes the results of real-world emissions testing of over 1200 vehicles. It finds that nitrogen oxide (NOx) emissions are on average 4 times legal limits, and real-world fuel economy and carbon dioxide emissions are 29% and 41% higher than official figures, respectively. Performance differs significantly between models certified to the same standard. Upcoming regulations may improve but not solve the problem. The need for an independent, real-world emissions standard is greater than ever.

1. Measuring and monitoring emissions of
nitrogen oxides from passenger cars
- a 1200 vehicle case study
Jane Thomas
12 October 2016

2. 2
• Nitrogen oxide emissions are on average four times the legal limits in real-world driving
• Real-world MPG for model year 2016 vehicles is on average 29% below official figures
• Carbon dioxide emissions are 41% above official levels
• Performance differs significantly between models homologated to same standard
• The new regulations for MPG, CO2 and NOx will improve but not solve the problem
• After 8 years of work at UN and EU, that will disappoint the market
• The need for an independent, real-world standard is greater than ever
The issue

3. 3
• Founded in 2011
• Headquartered in UK, with operations in London, Los Angeles and Stuttgart
• Specialist in PEMS testing and data analysis
• 1200+ vehicles tested
• Largest commercially available database of real-world emissions data
• Works with OEMs, Tier 1/2 suppliers, fuel and chemical companies, regulators,
consultancies, consumer media
Emissions Analytics’ credentials

4. 4
• Rolling in-service surveillance programme of production vehicles
• Target to test 500+ vehicles per year
• To complement type approval
• Air quality, greenhouse gases, fuel economy
• Independent, making data available to all
• Top-level ratings for sales and marketing
• Benchmark rankings
• Deep-dive analysis
Objectives

5. 5
Equipment
• SEMTECH-DS and -LDV
• Portable Emissions Measurement System
connects to tailpipe
• Captures emissions for CO2, CO, NO, NO2,
total hydrocarbons
• At 1 Hertz
• Air temperature, pressure, humidity
• GPS for speed and altitude
• Engine data via CANBUS
• Fuel economy derived via carbon balance
• Weight addition 100 kg

6. 6
How PEMS works

7. 7
• Reproducibility must be weighed against authenticity
• Aim to drive same cycle identically each time
• Control what you can – driver, vehicle conditioning, driving style
• Careful QA of data
• Normalise out as much of remaining variability as possible
 Approximate to laboratory
 Use better economics of testing to test more cars
 Trading slight deterioration in precision for greater sample size
 Greater overall confidence
Emissions Analytics concept

8. REGULATORY TRENDS
8

9. 9
Test cycles
New European
Driving Cycle
(NEDC)
World Light Duty
Transient Cycle
(WLTC)
Real Driving
Emissions (RDE)
Emissions
Analytics
Measurements MPG, CO2, CO, NOx MPG, CO2, CO, NOx NOx
MPG, CO2, CO, NOx,
NO, NO2
Cycle Defined speed trace Defined speed trace
Undefined with
aggregate and
dynamic constraints
Defined route
Test location Laboratory Laboratory On road with PEMS On road with PEMS
Conduct By OEM By OEM By OEM Independent
Normalisation None None
CO2 windows or
power binning
Based on dynamic
characteristics
Average speed (mph) 21 29 Varies 28
Average acceleration
(mph/s)
1.1 1.0 Varies 1.5
Average gradient (m/s) 0.0 0.0 Varies 0.8

10. 10
Tightening regulatory levels
2016 2017 2019 2021
Gasoline
CO2 (g/km) – fleet
average
130 on NEDC 130 on NEDC 130 on NEDC 95 on NEDC
NOx (g/km) – vehicle
level
0.060 on NEDC
0.126 on RDE
for new types
0.126 on RDE
for all cars
0.090 on RDE
For all cars
CO (g/km) – vehicle
level
1.000 on NEDC 1.000 on WLTC 1.000 on WLTC 1.000 on WLTC
Diesel
CO2 (g/km) – fleet
average
130 on NEDC 130 on NEDC 130 on NEDC 95 on NEDC
NOx (g/km) – vehicle
level
0.080 on NEDC
0.168 on RDE
for new types
0.168 on RDE
for all cars
0.120 on RDE
For all cars
CO (g/km) – vehicle
level
0.500 on NEDC 0.500 on WLTC 0.500 on WLTC 0.500 on WLTC

11. VEHICLE PERFORMANCE TRENDS
11

12. 12
• Rolling 12-month
average of exceedance
factor in blue
• Step-change technology
launched Q2 to Q3 2014
• No further reduction in
Conformity Factor since
then
• Average CF dipped to 2.7
in 2014, but risen to 3.5
since then
NOx Conformity Factor
0.00
5.00
10.00
15.00
20.00
25.00
02/11/2011
02/01/2012
02/03/2012
02/05/2012
02/07/2012
02/09/2012
02/11/2012
02/01/2013
02/03/2013
02/05/2013
02/07/2013
02/09/2013
02/11/2013
02/01/2014
02/03/2014
02/05/2014
02/07/2014
02/09/2014
02/11/2014
02/01/2015
02/03/2015
02/05/2015
02/07/2015
02/09/2015
02/11/2015
02/01/2016
ConformityFactor
Testdate
Euro5/6 NOx diesel ConformityFactor over time

13. 13
• German manufacturers
leading introduction of Euro
6 vehicles
• And offer the cleanest
• US manufacturers slow to
bring Euro 6 vehicles to
market
• Japanese very consistent
around market average
conformity factor
NOx by country
0.00
5.00
10.00
15.00
20.00
25.00
18/10/2012 06/05/2013 22/11/2013 10/06/2014 27/12/2014 15/07/2015 31/01/2016
ConformityFactor
Test date
Euro 6 diesel NOx Conformity Factorby country
Germany Japan Other US

14. 14
• Earliest Euro 6 vehicles
were LNT and EGR-only, but
LNT quickly became
dominant technology
• More recent switch to SCR
• Nevertheless, good
performance achieved by
many LNT systems
• Similarly wide spread of
performance between LNT
and SCR
NOx by after-treatment type
0.00
5.00
10.00
15.00
20.00
25.00
18/10/2012 06/05/2013 22/11/2013 10/06/2014 27/12/2014 15/07/2015 31/01/2016
ConfirmityFactor
Test date
Euro 6 diesel NOx Conformity Factor by after-treatment type
Exhaust Gas Recirculation (EGR)-only Lean NOx trap (LNT) Selective Catalytic Reduction (SCR)

15. 15
• Conformity factor
initially 2.1 – 168 mg/km
• Further increased by
exclusions due to
boundary conditions
• PEMS test-to-test
variability of 30%, so
OEMs will need to target
~129 mg/km to avoid
getting caught by in-
service surveillance
• 1 in 5 vehicles already
meet this
Will Real Driving Emissions help?
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
14/09/2011 01/04/2012 18/10/2012 06/05/2013 22/11/2013 10/06/2014 27/12/2014 15/07/2015 31/01/2016
ConformityFactor
Testdate
Euro 5 Euro 6

16. 16
• Historically, CF ~4 and
rising real-world fuel
economy
• As many Euro 6
vehicles hit the market
in 2014, CF began to
fall
• In 2015, real-world fuel
economy fell
significantly for first
time
• Effect particularly of
LNT and EGR?
Knock-on effect on fuel economy
2.5
3.0
3.5
4.0
4.5
5.0
44.0
45.0
46.0
47.0
48.0
49.0
50.0
2012 2013 2014 2015
ConformityFactor
Fueleconomy(MPG)
Year
Real-world fuel economy NOxConformityFactor

17. 17
Diesel MPG declining
30.0
35.0
40.0
45.0
50.0
55.0
60.0
65.0
2011 2012 2013 2014 2015
Fueleconomy(MPG)
Model Year of vehicle tested
Diesel- EQUA Mpg Diesel- official Petrol- EQUA Mpg Petrol- official
• Real-world average
diesel MPG declining
since 2012
• Official MPG kept
rising until 2014, but
even that now
declining
• Petrol still showing
gently improving trend

18. 18
Perils of downsizing
-45.0
-40.0
-35.0
-30.0
-25.0
-20.0
-15.0
-10.0
-5.0
0.0
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
EQUAMpgvariancetoofficialMPG
Enginesizeclass(litres)
Diesel Petrol
• The smaller the
engine, the bigger the
gap between real-
world and official MPG
• Sub-1.5 litre petrol
engines have
particularly large
deficits
• NEDC rewards
downsizing, but 1.5-
2.0 litre optimal for
mixed driving in reality

19. INDEPENDENT RATINGS
19

20. 20
• Vehicle rating scheme based on their real-world NOx emissions, launched in April 2016
• Non-statutory complement to new Real Driving Emissions regulations
• But will also
• Discriminate between high and low emitters, rather than just pass/fail
• Be updated for each model year to keep up with new calibrations
• Ratings are published and into the public domain for free
• Manufacturers and consumer media can adopt as independent, voluntary standard
• Similar to New Car Assessment Programme (Euro NCAP, Global NCAP)
 Robust, independent standard needed to measure and incentivise actions to bring about air
quality improvements
EQUA Air Quality Index

21. 21
www.equaindex.com

22. 22
Rating bands
Rating Lower bound
(g/km, exclusive)
Upper bound
(g/km, exclusive)
External reference point
A 0.00 0.08 Meets Euro 6 limit for diesels, and meets Euro 4 limit
for gasoline
B 0.08 0.12 Meets 1.5 Conformity Factor under Euro 6 Real
Driving Emissions regulation
C 0.12 0.18 Meets Euro 5 limit for diesels (and similar to 2.1
Conformity Factor under Euro 6 Real Driving
Emissions regulation)
D 0.18 0.25 Meets Euro 4 limit for diesels
E 0.25 0.50 Meets Euro 3 limit for diesels
F 0.50 0.75 No comparable Euro standard: roughly equal
to 6-8 times Euro 6 limit
G 0.75 1.00 Roughly equal to 8-12 times Euro 6 limit
H 1.00 None Roughly equal to 12+ times Euro 6 limit

23. 23
Diesels with A ratings

24. 24
Diesels with A ratings

25. 25
www.equaindex.com

26. 26
www.equaindex.com

27. 27
www.equaindex.com
• MPG values for almost all vehicles on
sale in last five years
• Over 66,000 model variants
• Around 5,000 test values
• Remainder extrapolated using new
proprietary model of real-world MPG,
based on technical characteristics of
vehicles
 Comprehensive alternative to official
system

28. Jane Thomas, Business Manager
jane@emissionsanalytics.com
+44 (0) 20 7193 0489
+44 (0) 7958 405409
www.emissionsanalytics.com
www.equaindex.com
28