Volume-07 Issue 12, Decembe -2023 ISSN: 2456-9348
Impac Fac o : 6.736
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [624]
VEHICLE GREENHOUSE GAS EMISSIONS STANDARDS FOR HEAVY-DUTY
VEHICLES — PHASE 3 AND THE UNCERTAINTY AHEAD
Muhammad Waleed Saleem, Muhammad Ha is Saleem, Muhammad Adeel Sajjad
Independen Resea che
ABSTRACT
The U.S. En i onmen al P o ec ion Agency (EPA) has p oposed new Phase 3 g eenhouse gas (GHG) s anda ds o
hea y-du y ehicles (HDVs) beginning in model yea (MY) 2027, ollowing he exis ing Phase 1 and Phase 2
p og ams. The p oposal ou lines mo e s ingen CO₂ emission a ge s in ended o accele a e lee deca boniza ion
while p ese ing manu ac u e lexibili y. Howe e , he Phase 3 ule is subjec o deep unce ain y ac oss mul iple
dimensions: echnology ealiza ion, in as uc u e eadiness, ma ke adop ion o ze o-emission ehicles (ZEVs), and
legal/ egula o y du abili y. This pape (i) syn hesizes he s a us o he Phase 3 p oposal, (ii) iden i ies he p incipal
unce ain y domains, and (iii) de elops a h ee-scena io amewo k wi h illus a i e quan i a i e bounds o guide lee
planne s, egula o s, and esea che s. Scena io-based es ima es sugges ha , plausible median educ ions o 15 −
25 % in well- o-wheel 𝐶𝑂₂𝑒 in ensi y by 2032 a e easible unde an op imis ic adop ion pa h, bu downside cases (e.g.
delayed ZEV deploymen o ollback isk) could comp ess educ ions o <10 %. The pape a gues o embedding
unce ain y disclosu e, egula o y du abili y sa egua ds, and s aged ansi ion lexibili y in o any inal ule.
1. INTRODUCTION
Hea y-du y ucks and buses (Classes 2b o 8) ep esen a c i ical piece o he U.S. deca boniza ion challenge: hough
ela i ely ew in numbe , hey consume la ge olumes o diesel uel and hus p oduce disp opo iona e g eenhouse
gas (GHG) emissions in he eigh and ansi sec o s. The exis ing Phase 1 (MY 2014–2018) and Phase 2 (MY 2018–
2027) EPA/NHTSA p og ams ha e manda ed inc emen al imp o emen s in uel e iciency and CO₂ emissions o
HDVs and engines. In Ap il 2023, EPA published a p oposed ule o Phase 3 GHG s anda ds, a ge ing MY 2027
onwa d, and solici ed commen s h ough mid-2023.[1]
A his s age, key design decisions a e public, bu much emains unse led, pa icula ly on c edi ing, ZEV ajec o ies,
and compliance lexibili ies. The ime ho izon o MY 2027–2032 magni ies unce ain ies in echnology, in as uc u e,
and legal isk. Gi en he long li e and capi al in ensi y o hea y ucks, decisions made now will in luence compliance
pa hways and lee s a egies o yea s o come.
This pape aims o (a) cla i y he cu en s uc u e o he p oposed Phase 3 ule, (b) ca ego ize he majo unce ain y
domains ha will a ec ealized ou comes, and (c) p opose a scena io + bound me hodology ha lee s, OEMs, and
egula o s can use o s ess- es ou comes. Al hough de ailed nume ical simula ion equi es p op ie a y lee da a, he
amewo k he ein is designed o be implemen able using open sou ces (e.g. EPA’s HD TRUCS ool, indus y da a,
LCAs) and o encou age anspa en unce ain y epo ing in he inal ulemaking.
2. STATUS OF THE PHASE 3 PROPOSAL
2.1 O e iew o p oposal and iming
On Ap il 27, 2023, EPA published he P oposed Rule: G eenhouse Gas Emissions S anda ds o Hea y-Du y Vehicles
— Phase 3 (Docke EPA-HQ-OAR-2022-0985) co e ing oca ional ucks and ac o s beginning MY 2027.[1] The
p oposal includes a d a Regula o y Impac Analysis (RIA) and solici s s akeholde commen on echnology
easibili y, cos assump ions, and c edi lexibili ies.
Pe he p oposal, a ge s a e se ela i e o he Phase 2 MY 2027 baseline, wi h s eepe annual igh ening h ough MY
2032. The ule would main ain a echnology-neu al, pe o mance-based app oach, simila o ea lie phases, allowing
OEMs o mix in e nal combus ion engine (ICE) imp o emen s, hyb ids, o ZEV deploymen o comply.[2]
Volume-07 Issue 12, Decembe -2023 ISSN: 2456-9348
Impac Fac o : 6.736
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [625]
Some public sou ces no e ha he inaliza ion ime able may aim soon o he inal ule, hough commen a o s cau ion
delays a e likely gi en legal and s akeholde complexi y. Because o hose isks, many lee s and OEMs a e al eady
scena io-s ess es ing hei in es men pa hs.
2.2 Key s uc u al elemen s and cons ain s
• Baseline and glide pa h: The p oposal uses Phase 2 MY 2027 le els as he s a ing poin . The glide pa h igh ens
CO₂ pe uni dis ance o on-mile me ics in oca ional and ac o classes.[1][2]
• ZEV adop ion ceilings and in e pola ion: EPA’s d a modeling imposes maximum ZEV pene a ion caps (e.g.
20 % in MY 2027, amping o pe haps 70 % in 2032 in some segmen s) wi h linea o piecewise in e pola ion.[3]
• C edi , banking, and lexibili ies: The p oposal e ains a e aging, banking, ading (ABT) mechanisms, o -cycle
c edi s, and p o isions o ad anced echnology c edi s (e.g. o ea ly ZEV deploymen ), wi h s akeholde
commen solici ed on hei pa ame e s.[1][2]
• Subca ego y di e en ia ion: The ule delinea es mul iple ac o subca ego ies (cab s yle, oo heigh ) and
oca ional subca ego ies laye ed by du y cycle. Each has speci ic CO₂ (o g/ on-mile) a ge s.[3]
• Feasibili y cons ain s: EPA’s RIA e e ences he HD TRUCS (Hea y-Du y Technology Resou ce Use Case
Scena io) ool o e echnology eadiness, cos payback, and adop ion cu es in key ehicle classes.[4]
• Public commen s & deba es: S akeholde s (OEMs, lee ope a o s, u ili ies) ha e aised conce ns a ound
in as uc u e lead imes, ba e y cos ends, g id capaci y, u ili y in e connec ion, du y-cycle deg ada ion, and
wa an y/main enance isk. The docke indica es ha EPA will conside pe o mance c edi limi s, ZEV amp
con ols, and compliance lexibili y in esponse.[1][2][5]
In sho , he Phase 3 a chi ec u e is well a icula ed in p oposal o m, bu he inal ule’s pa ame e iza ion will c ucially
shape eal-wo ld deca boniza ion ou comes.
3. MAJOR SOURCES OF UNCERTAINTY
To an icipa e how he e en ual Phase 3 s anda d will play ou ac oss lee s and emissions ajec o ies, ou majo
classes o unce ain y should be explici ly conside ed in analysis and planning.
3.1 Technology Realiza ion Risk
E en i he p oposed a ge s a e echnically wi hin each on pape , a ia ion in eal-wo ld du y cycles, auxilia y load
penal ies, aging, deg ada ion, and misma ch be ween es and ope a ional condi ions can e ode gains. ZEVs in oduce
addi ional unce ain ies: ba e y du abili y, pack cos declines, ene gy densi y p og ession, cold-wea he pe o mance,
and ehicle o al cos o owne ship (TCO) eliabili y in hea y-du y du y cycles.
The RIA’s modeling assump ions o ICE imp o emen s o hyb id in eg a ion may unde es ima e nega i e
in e ac ions (e.g., weigh penal ies, packaging cons ain s). Thus, ac ual lee -le el gap o a ge may be la ge han
he p oposal’s “cen al case” sugges s.
3.2 In as uc u e, Ma ke , and Adop ion Risk
ZEV adop ion is cons ained by in as uc u e eadiness (cha ging/hyd ogen s a ion deploymen , g id capaci y, u ili y
in e connec imelines, demand cha ges). Delays o cos o e uns in in as uc u e could slow ehicle up ake, pushing
he lee mix owa d ICE-dominan compliance pa hs and educing agg ega e GHG aba emen .
Also, lee decision beha io , inancing, esidual alue unce ain y, and leasing dynamics in luence adop ion cu es.
Ma ke ine ia in ehicle u no e a es u he compounds delays.
3.3 Ups eam Ene gy and Li e-cycle Risk
While Phase 3 egula es ailpipe CO₂ (i.e. ank- o-wheel), ac ual well- o-wheel (WTW) emissions depend on ups eam
e ining ( o diesel o al e na i e uels) and elec ici y sys em ca bon in ensi y ( o BEVs). Regional a ia ion in g id
mix o unexpec ed delays in elec ici y deca boniza ion will in luence ne GHG bene i .[6] I elec ici y emains
ca bon-in ensi e in key egions, he ela i e bene i o elec i ica ion is educed.
Hence, in co po a e GHG accoun ing o clima e planning, he unce ain y in ups eam emissions is non i ial and mus
be included in any aba emen es ima e.
Volume-07 Issue 12, Decembe -2023 ISSN: 2456-9348
Impac Fac o : 6.736
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [626]
3.4 Policy Du abili y and Legal Risk
Pe haps he mos c i ical—and leas quan i iable—unce ain y is whe he he inal Phase 3 ule will emain s able o e
i s li e ime (2027–2032) in he ace o poli ical, judicial, o adminis a i e e e sal isks.
• Legal challenges: Opponen s may challenge he ule on Clean Ai Ac in e p e a ions, cos jus i ica ion,
un easonable bu dens, o c edi ing scheme design.
• Adminis a i e e e sal: A u u e adminis a ion migh a emp o e ise o oll back po ions o Phase 3 ia
no ice-and-commen o ia ein e p e a ion o s a u o y au ho i y.
• Cong essional dynamics: Legisla i e changes o ide p o isions could cons ain EPA’s au ho i y o equi e
changes o es p ocedu es o c edi ing.
Because hea y-du y ucks a e long-li ed asse s, unce ain y abou egula o y du abili y may induce in es men limbo
o g adual adop ion a he han on -loaded compliance e o .
4. SCENARIO FRAMEWORK AND METHODS
Th ee-scena io amewo k:
• Scena io A (“Base/Mode a e”): Phase 3 is adop ed la gely as p oposed, ZEV adop ion p oceeds mode a ely pe
EPA’s cen al modeling, and echnology ealiza ion is a e age.
• Scena io B (“High Ambi ion”): S onge ZEV up ake ( as e in as uc u e, cos declines), obus ICE/hyb id
pe o mance, and ull compliance.
• Scena io C (“Conse a i e/Delayed”): Slowe ZEV up ake (in as uc u e delays), echnology sho alls, o
pa ial ollback p essu es, pushing compliance o ely mo e on ICE imp o emen s.
Each scena io yields a ange o 𝐶𝑂₂𝑒 in ensi y educ ions by 2032, in bo h ailpipe (𝑔𝐶𝑂₂/𝑚𝑖) and WTW
(𝑔𝐶𝑂₂𝑒/𝑚𝑖 o pe on-mile) e ms.
𝐷𝑖𝑒𝑠𝑒𝑙: 𝐸𝑊𝑇𝑊 ≈ (𝐸𝐹𝑇𝑇𝑊 +𝐸𝐹𝑢𝑝𝑠𝑡𝑟𝑒𝑎𝑚) 𝑚𝑝𝑔
⁄
𝐵𝐸𝑉: 𝐸𝑊𝑇𝑊 ≈𝑘𝑊ℎ𝑚𝑖
⁄ × (𝑔𝑟𝑖𝑑 𝐸𝐹)
Mon e Ca lo (≥ 10,000 𝑑𝑟𝑎𝑤𝑠) samples 𝑚𝑝𝑔 𝑘𝑊ℎ − 𝑚𝑖
⁄, u iliza ion, ups eam ac o s, ZEV sha es, and policy
iming o deli e medians and 90% in e als; a iance decomposi ion a ibu es sha es o inpu s.
4.1 Assumed Baseline and Inpu s
• Baseline in ensi y: Diesel HD ac o s in eigh du y cycles a e age ~1,100 − 1,300 𝑔 𝐶𝑂₂𝑒/𝑚𝑖 ( ailpipe +
ups eam adde s).
• E icien ICE po en ial: Inc emen al gains o 15 −25 % o e Phase 2 baseline by 2032 unde pe ec execu ion.
• ZEV con ibu ion: In cen al pa hs, assume ZEVs con ibu e 15 −40 % o new sales by 2032, weigh ed by
segmen , wi h ull-li e subs i u ion e ec .
• Ups eam sp ead: Elec ici y ca bon in ensi y and diesel e ining emissions add ±10 −20 % ma gin o ne WTW
bene i s.
• Du abili y discoun : Unde unce ain y, cas a 5 − 15 % ma gin o egula o y d ag, ma ke lag, o ollback
slippage.
4.2 Illus a i e Reduc ion Es ima es
Scena io
Mid-Case Tailpipe
Reduc ion (2032 s 2023)
Mid-Case WTW
Reduc ion
Lowe Bound
Reduc ion
Uppe Bound
Reduc ion
A (Mode a e)
~18 %
~15 %
~10 %
~25 %
B (High Ambi ion)
~25 %–30 %
~22 %–28 %
~18 %
~35 %
C
(Conse a i e/Delayed)
~8 %–12 %
~6 %–10 %
~2 %
~15 %
Volume-07 Issue 12, Decembe -2023 ISSN: 2456-9348
Impac Fac o : 6.736
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [627]
These es ima es e lec agg ega e lee ou comes; indi idual segmen s ( oca ional, egional ucks, ac o s) may a y
signi ican ly. The gap be ween ailpipe and WTW unde sco es he impo ance o ups eam assump ions.
4.3 Va iance Decomposi ion (Quali a i e)
In a Mon e Ca lo o Sobol’ sensi i i y analysis, con ibu ions o ealized a iance migh b eak down app oxima ely:
• Technology ealiza ion: ~30–40 %
• ZEV adop ion & in as uc u e cons ain s: ~25–35 %
• Ups eam emissions a ia ion: ~10–20 %
• Policy d ag / du abili y isk: ~10–20 %
This decomposi ion highligh s ha no single dimension domina es en i ely; po olios o isk accumula e.
5. IMPLICATIONS AND POLICY PRESCRIPTIONS
5.1 Fo Flee s, OEMs, and In es o s
• Adop no- eg e echnologies now. E en i ZEV ollou lags, ad anced ae odynamics, low- olling- esis ance i es,
was e-hea eco e y, and hyb idiza ion yield uel sa ings unde any scena io.
• Modula s a egy hedging. Main ain lexibili y among ICE, hyb id, and elec i ied pla o ms so ha capi al
in es men s can pi o based on he ealized pa h.
• Ea ly pilo ZEV deploymen in a o able segmen s. U ban, egional, o sho -haul use cases wi h p edic able du y
cycles and depo access a e likely lowe - isk pa hs.
• Scena io modeling and lexibili y op ions. Build in e nal models wi h Scena io A/B/C bounds and ack key
leading indica o s (ba e y cos , in as uc u e pe mi lag, c edi ma ke design).
• Baseline WTW in ensi y o long-haul diesel ac o s: ~1,100 − 1,300 𝑔𝐶𝑂₂𝑒/𝑚𝑖 depending on mpg and
ups eam adde s.
Scena io A: ~15 −20% median WTW educ ion by 2032 (90% 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 ≈ −10% 𝑡𝑜 − 25%).
Scena io B: ~22–30% median (90% 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 ≈ −18% 𝑡𝑜 − 35% .
Scena io C: ~6–10% median (90% 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 ≈ 0% 𝑡𝑜 − 15%).
Va iance sha es (quali a i e): 30 −40% echnology, 25 −35% in as uc u e/adop ion, 10 −20% ups eam,
10 −20% policy du abili y. Segmen ou comes a y widely.
5.2 Fo EPA and Regula o s
• Embed unce ain y disclosu e. The inal ule should equi e ha compliance analyses and lee -le el plans include
dis ibu ional anges (e.g. 5 h–95 h pe cen iles), no jus poin es ima es.
• P o ide du abili y assu ances. Mechanisms such as mid-cou se e iew pe iods, s a u o y sa e ha bo s, o mul i-
yea c edi banking s eng hen egula o y con idence.
• S age in as uc u e signals. Coo dina e wi h DOE, DOT, and s a es o accele a e cha ging/ ueling in as uc u e
deploymen and minimize boom–bus cycles.
• Re ine c edi lexibili ies. Design ad anced c edi mul iplie s o ansi ion c edi s o cushion ea ly adop ion isk,
wi hou unde mining s ingency.
5. ILLUSTRATIVE BOUNDS (NON-BINDING, FOR PLANNING)
Baseline WTW in ensi y o long-haul diesel ac o s: ~1,100 − 1,300 𝑔𝐶𝑂₂𝑒/𝑚𝑖 depending on mpg and ups eam
adde s.
Scena io A: ~15 −20% median WTW educ ion by 2032 (90% in e al ≈ −10% 𝑡𝑜 − 25%).
Scena io B: ~22–30% median (90% in e al ≈ −18% o −35%).
Scena io C: ~6–10% median (90% in e al ≈ 0% o −15%).
Va iance sha es (quali a i e): 30–40% echnology, 25–35% in as uc u e/adop ion, 10–20% ups eam, 10–20%
policy du abili y. Segmen ou comes a y widely.
Volume-07 Issue 12, Decembe -2023 ISSN: 2456-9348
Impac Fac o : 6.736
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [628]
Figu e 1: Illus a i e dis ibu ions o 2032 WTW CO₂e educ ion ela i e o a 2023 baseline ac oss h ee
scena ios
Figu e 1 is In ended o planning use only.
6. IMPLICATIONS
Flee s/OEMs — P io i ize no‑ eg e e iciency; pilo ZEVs in eady segmen s; hedge wi h modula po olios; disclose
unce ain y bands.
Regula o s — Embed unce ain y disclosu e and mid-cou se e iew; s uc u e c edi s ha ewa d du able ea ly ZEV
deploymen s; coo dina e in as uc u e p og ams.
Resea che s — Expand du y‑cycle da ase s; open Mon e Ca lo oolchains linked o EPA models; e ine egional g id
p ojec ions o HDV WTW accoun ing.
7. LIMITATIONS
Quan i a i e en elopes a e illus a i e and should be e ined wi h lee -speci ic inpu s and inalized pa ame e s.
8. CONCLUSIONS
Phase 3 can deli e double‑digi GHG in ensi y educ ions by 2032 i echnology, in as uc u e, ups eam
deca boniza ion, and policy du abili y con e ge. Scena io planning wi h explici unce ain y bands and s aged
in es men op ions is essen ial.
REFERENCES
1) EPA (Ap . 27, 2023). P oposed Rule: G eenhouse Gas Emissions S anda ds o Hea y‑Du y Vehicles—
Phase 3 (Docke EPA‑HQ‑OAR‑2022‑0985), d a RIA, and HD TRUCS documen a ion.
2) Fede al Regis e (Ap . 27, 2023). G eenhouse Gas Emissions S anda ds o Hea y‑Du y Vehicles—Phase
3 (P oposed Rule).
3) ICCT (2023). U.S. Phase 3 HDV GHG s anda ds—policy b ie s and bene i s analyses.
4) EPA/NHTSA (2016). Phase 2 Hea y‑Du y Fuel E iciency & GHG s anda ds— inal ule and RIA.
5) Public commen s (mid‑2023) on in as uc u e eadiness, ZEV adop ion caps, and c edi ing design.
6) Open WTW/LCA e iews (2020–2023) on HDV ups eam emissions and g id ca bon in ensi y.
