Winning the Oil End Game -- Technical Annex …

Winning the Oil End Game -- Technical Annex                          www.oilendgame.org

Chapter 8

INTELLIGENT HIGHWAY SYSTEMS (IHS)

1. Summary
Instantaneous adoption of incident management, signal coordination, and ramp meters on
all 75 U.S. metropolitan areas' major roads vs. actual 2001 deployment would have
reduced wasted time and 5.7 billion gallons of wasted fuel by an incremental 15%. This
savings is equivalent to five years' congestion growth, or 0.85 billion gallons of roadway
fuel in 2001 [3]. Additionally, an approximate estimate for full deployment to the same
metropolitan areas of electronic toll plazas gives another 0.1 billion gallons per year. This
estimate is based on an approximate extrapolation of the experience of New Jersey
Turnpike [8]. Implementation of a modest amount of advanced routing technologies and
of 25% of a set of other technologies would imply a 2001 savings of an additional ~0.5
billion gallons of fuel per year in the conventional technology case.

The total, 1.45 billion gal/y of fuel, agrees very well with both the CEF report [1] and the
10-year plan launched in 2002 by IHS America [6]. Fully implemented, these 1.45 billion
gal/y of fuel versus the 2000 EIA Annual Energy Outlook (AEO) baseline would give
1.68b gal/y in savings versus the EIA Annual Energy Outlook 2004 with Projections to
2025 (AEO 2025) baseline given conventional technology implementation in other
highway sectors, alternatively, some 0.98 billion gal/y given State of the Art
implementation in other non-IHS sectors. Respectively, this equates to some 0.11 Mbbl/d
of product (we use product here to mean the same as roadway fuel, i.e., before conversion
to crude oil) given conventional non-IHS technologies or 0.06 Mbbl/d of product given
State of the Art non-IHS savings. Using either baseline, approximately 0.9% of the AEO
2025 total crude oil consumption would be saved deploying the conventional
technologies.

Our Conventional Wisdom (CW) technology portfolio excludes at least six other major
technologies that collectively could save between 17+% and 45+% off a given baseline.
These include signal priority modeling for bus rapid transit, intelligent cruise control,
very close vehicle spacing, vehicle classifiers, routing algorithms, and agent-based
computing infrastructure, all described and referenced in the final section of this chapter,
see below. If and when deployed along with the Conventional Wisdom technology suite,
a subset of these technologies represents our State of the Art IHS technology portfolio.
Estimating the impact of the SOA portfolio in terms of fuel saved and cost is particularly
difficult, so rather than go with the summed savings from all of the additional
technologies, we have conservatively estimated the impact of any subset to be double the
total of the conventional technology portfolio. In actuality, the impact could be manyfold
better.

Any estimates, let alone reliable ones, for measure costs are very difficult to come by.
Experience indicates that IHS costs are quite low, and we outline a set of four findings


                                                                                           1
that indicate that it is likely that total costs are very low indeed. Because the value of non-
fuel related benefits outweigh fuel-related benefit by at least one order of magnitude,
saved time being the most important, we have assumed that all IHS investments would be
made irrespective of fuel savings. For this reason we assign a zero cost of saved fuel to
IHS-related investments across both technology portfolios.

2. Overview of methodology and results
Collectively, Intelligent Highway Systems technologies have the potential to integrate
vehicles, systems users, and infrastructure towards more seamless interaction and
therefore improve surface transportation safety, efficiency and convenience. IHS
technologies include a broad range of wireless, electronic, and automated technologies,
most of which facilitate rapid processing and analysis of real-time information for better
traffic management. The definition covers both in-vehicle and systems-based
technologies. Examples of the former are precision docking for buses, automated
guideways, and collision avoidance systems. When IHS technologies are applied to
systems-management, they can reduce fuel consumption. This is achieved by enabling
one or more features, for example traffic smoothing, route planning and timing, direct
congestion reduction, pricing and demand-management, enhancing attractiveness of
public transportation mode use, vehicle transmission adaptation to variable conditions
and terrain, and facilitation of small platoons of closely spaced vehicles [2].

We use the definitions and categories as classified in [2]. We then briefly review the
impact of IHS assumed in The Interlaboratory Working Group 2000 report "Scenarios for
Clean Energy Futures" [1] Advanced scenario. Next we review the impacts from
deploying several technologies looked at by the Texas Transport Institute, [4] and [5].
This review reveals that the number of estimates of impacts on fuel use from IHS
technology deployment is low in relation to the number of available technologies. For our
conventional technology suite we therefore select those technologies for which we have a
reasonably accurate impact estimate. Given the number of unanalyzed technologies and
their possible impact, and the relatively modest conventional technology impact, we have
simply assumed that the State of the Art technology portfolio would, if deployed, double
the savings relative to the conventional technology suite.

We then review costs. While the costs are difficult to estimate, it is clear that on average
IHS-related technologies would not be particularly expensive and that the value of non-
fuel related benefits outweigh fuel-related benefits by at least one order of magnitude. If
implemented to the degree stipulated herein, they would be relatively very cost effective.
It is also clear that due to the high ratio of non-fuel related benefits to fuel-related
benefits IHS technologies would be deployed largely without regard to fuel-related
benefits. Based on this, we assume that the two portfolios of IHS related technologies
considered here would have an average CSE of zero. For comparison, we also show the
all-in cost on a CSE-basis. This cost simply assumes the cost of IHS technologies would
be equal to the average CSE of all highway vehicle-based measures. Our results are
summarized in Table 7-1.




                                                                                             2
3. Technologies and impacts
IHS architecture: The U.S. National Intelligent Transportation Architecture (of which
IHS would be a subset) has 32 IHS user services bundled into 8 main categories. These
eight categories are (1) travel and transportation management, (2) public transportation
options, (3) electronic payment, (4) commercial vehicle operations (CVO), (5)
emergency management, (6) advance vehicle control and safety systems, (7) information
management, and (8) maintenance and construction [2].

The Clean Energy Futures report: This report simulated the effect of increased usage of
IHS systems in their Advanced scenario by reducing by one percentage point the
degradation factor in NEMS. This is the factor that translates EPA values of fuel
economy into "on-the-road" values, and it accounts for congestion and other factors that
increase fuel usage over the value that would be computed using the EPA values. No
figures for the costs of this change were given, nor did the report outline a justification
for the magnitude. That said, for degradation factors ranging from 0.75 to 0.85, the net
effect of this adjustment is to reduce fuel consumption per mile driven by between 1.19%
and 1.35% from any given baseline mpg and VMT (Vehicle Miles Traveled). This is
quite similar to our findings for savings that would result from deployment of our
Conventional Wisdom technology suite. The CEF report does not estimate costs, and we
have taken this as another indicator that these costs are quite low.

Impacts are mostly independent of capital cycles: Impacts from systems-based IHS
technologies are relatively less dependent on long capital turnover cycles. The
importance of this factor is often overlooked in a sector where efficiency improvement is
generally heavily dependent on capital turnover. Many of these technologies may
therefore relatively very rapidly affect the efficiency of entire substocks of vehicles, e.g.,
roadway and airway vehicles. Put another way, these technologies may have an adoption-
advantage over vehicle-based technologies. Vehicle-based technologies are often not
easily suited for retrofit, so their actual impacts tend to correlate closely with the time
constants associated with vehicle turnover periods. However, deployment of systems-
based IHS technologies will tend to have a nearly immediate impact on the entire existing
stock of local vehicles (e.g. London's ring road fee, electronic toll payments, traffic
signal synchronization, etc.), thereby avoiding the capital turnover constraint.

Net benefits from and deployment strategies for system-based IHS technologies: In terms
of net benefits, systems-based IHS technologies can be very cost-effective as they often
involve little hardware, generally low costs, and usually entail multidimensional benefit
impacts. A single technology will often contribute to all of congestion reduction, lowered
traffic risks, improved safety, and increased vehicle stock average fuel economy.
For rigorous benefit quantification of some mainly congestion-oriented systems-based
IHS technologies we have mainly relied on work done by the Texas Transport Institute
(TTI). TTI finds that traffic congestion in the 75 largest (of 400) U.S. urban areas in 2001
cost $69.5b by wasting 3.5 billion hours (lost productivity) and 5.7 billion fuel gallons
[2]. This is therefore the size of the reduction potential. The potential is up from $8b in
1982 [3, Exhibit 12, p. 24]. This estimate relies in part on abatement measures analysis in




                                                                                            3
[4] and method in [5]. Regarding lowering congestion, a key TTI conclusion is that the
best solution entails a portfolio of options:
       It is clear that adding roadway at about the same rate as traffic grows will slow the
       growth of congestion. It is equally clear, however, that only five of the 75 areas studied
       were able to accomplish that rate. There must be a broader set of solutions applied to the
       problem, as well as more of each solution than has been implemented in the past, if more
       areas are to move into the "maintaining conditions or making progress on mobility"
       category... This analysis shows that it would be almost impossible to attempt to maintain
       a constant congestion level with road construction only. Over the past 2 decades, only
       about 50 percent of the needed mileage was actually added. This means that it would
       require at least twice the level of current-day road expansion funding to attempt this road
       construction strategy. An even larger problem would be to find suitable roads that can be
       widened, or areas where roads can be added, year after year. Most urban areas are
       pursuing a range of congestion management strategies, with road widening or
       construction being one of them.
        [3, p. 31 and p. 34. See also Exhibit 17, p. 33]
See [6] for updated information on metropolitan area progress in IHS deployment.
Current understanding of the potential impact of IHS on fuel consumption is still limited.

The 10-year plan launched in 2002 by IHS America has a goal of saving at least 1 billion
gallons per year. This plan has four programmatic themes, two of which could result in
significant energy consumption reductions in the future, advanced crash avoidance, and
advanced transportation management. We discuss these next.

Savings from Advanced Crash Avoidance and Route Guidance Technology:
Technologies for advanced crash avoidance include adaptive in-vehicle electronics,
which is forecast to reduce fuel consumption by acceleration and deceleration smoothing,
automatic response in stop-and-go driving, anticipatory throttle and transmission
adjustment in varying road and terrain conditions, and enabling safe movement of
platoons of tightly spaced trucks and cars [2]. We do not have estimates of savings nor
costs in conjunction with this category, and have excluded it from our total IHS potential
savings.

In addition to advanced crash avoidance, deployment of route guidance technologies to
reduce miles wasted from erroneous route choices and traffic delays will make an impact,
e.g., in the trucking industry. This product has been accounted for in the trucking section,
and it is clear it could reduce total VMT by reduction of out-of-route miles across all
vehicle classes. Eliminating these would amount to gross fuel savings in the form of
lower VMT. Here the potential is large, as between 3% and 10% of trucking miles driven
are wasted due to poor route-planning [7]. As light vehicles travel a large fraction of their
VMT on familiar routes during commuting, we have used a potential of 0.25% VMT
reduction for light vehicles.

Savings from Advanced transportation management: The advanced transportation
management theme is also forecast to have significant effect on vehicle energy use via
tools to manage vehicle flows adaptively, within the physical infrastructure and across
jurisdictions and modes. This relies on area-wide surveillance and detection, real-time


                                                                                                     4
data capture and analysis of traffic flow data, and predictive capabilities. Highlights of
the better-understood opportunities now follow.

Traffic signal control has been around for a long time, and has recently become a
significant component of IHS. Studies reveal fuel efficiency benefits ranging between 1.6
and 50%, with most results less than 20% [2]. However, it is not clear whether these
percentages would be applicable against national fuel consumption figures, although it is
probably reasonable that the lower figure could be. While it is again not clear what the
costs are, traffic signal installation would likely have very small costs on a CSE basis.

Improved incident management, via surveillance, verification, better emergency dispatch,
changeable message signs, and early notification to upstream drivers, decreases fuel
consumption by reducing the delay and congestion associated with blocked traffic. It isn't
clear by how much, but it is clear it could be important. For a limited initiative, Maryland
calculated fuel savings of 4.1 million gallons per year.

Ramp metering is a technology that safely spaces vehicles margining onto a highway
while minimizing speed disruptions to existing flows, with the most significant benefit
being time savings and with a mixed fuel consumption impact since ramp metering
causes vehicles on ramps to stop and go, which increases fuel consumption, while it
causes smoother flow on the freeway, resulting in consumption reduction [2]. It appears
more detailed studies are needed to understand how the effects interact and how fuel
consumption is affected [2].

For the three technologies above, TTI simulated full adoption--incident management,
signal coordination, and ramp meters--on all 75 cities' major roads vs. actual 2001
deployment. The 2001 actual deployment ranged from 0% to ~60%, depending on what
measure and where. Full deployment would have reduced the wasted time and fuel by an
incremental 15%, equivalent to five years' congestion growth or 0.85 billion gallons of
roadway fuel in 2001. This full implementation equates to 1.69 billion gallons per year in
savings versus the AEO 2025 baseline, or some 0.11 Mbbl/day, i.e., 0.4% of the AEO
2025 total crude consumption. While the benefits of HOV and bus lanes and of public
transport are also quantified in [4], we have generally excluded mode switching from our
impact estimates in this section and elsewhere. Of course, if such a switch involves
greater convenience for the user, this exclusion becomes a clear conservatism.

Electronic toll collection (ETC) saves time and reduces stop-and-go traffic, with analyses
showing that fuel savings offset operating costs, with ETC systems moving five times as
many vehicles as conventional toll lanes, and ETC systems significantly lowering toll
plaza operational costs and reducing delay by 85% [8]. While the following data from
New Jersey Turnpike ETC build-out does not permit a complete net benefits analysis,
they are worth noting. 1.2 million gallons were saved per year at NJTP at the 27 toll
plazas employing ETC. Moreover, $2.7 million were saved from reduced handling costs
of fare media, and revenues were increased by 12% after automated fare collection
implementation [8]. If capital costs were of this order, the CSE would be approximately
zero. The study evaluated NJTP's E-ZPass electronic toll collection system by measuring


                                                                                             5
traffic counts, queue lengths, lane configurations, and transaction times during peak
periods at 27 toll locations. Field observations were evaluated against toll collection
records and 24-hour total queue length and average vehicle-class delay before and after
E-ZPass deployment at each station were generated. Toll plaza delay was reduced by
approximately 85% for a total savings of 2,091,000 vehicle-hours per year. Passenger car
delay was reduced by 1.8 million hours per year; truck delay was reduced by 291,000
hours per year; and E-ZPass user delay was reduced by 1,344,000 hours per year. User
cost savings as a result of delay reductions were estimated at $19.0 million per year for
passenger cars and $6.1 million per year for trucks for a total annual savings of $25.1
million. User cost savings related to fuel consumption were estimated at $1.5 million for
passenger cars, and $400,000 for trucks.

One could estimate the impact of nationwide ETC if the following data were available:
NJTP relative traffic density and fraction of traffic, fraction of NJTP plazas involved in
the specific savings, and what fraction of nationwide plazas with what traffic densities
have ETC today. Unfortunately these data are not easy to come by. The NJTP is 148
miles in length, but it is unclear what IHS lane-mile-count is or how long it is and what
its relative traffic density is per lane-mile. Alternatively, if we assume there are about as
many toll collection points that could benefit from this initiative in the rest of the 75
major metropolitan areas in mainland U.S. as along the 27 toll collection points, that
these points are not yet built out, and that the average benefits would be approximately
the same as for the 27, it would appear that some 1.2M gal/y x 75 ~ 100 million
additional gallons could be saved from E-ZPass introduction nationwide if at the density
seen with the NJTP.

Savings from other measures: Several other measures that are not easily quantifiable are
now listed. Given the magnitude of the potential savings, we believe some of these will
make an impact in our SOA technology suite. We estimate that these will, together with
the conventional technology suite, amount to approximately 0.12M bbl saved per day in
2025. The measures include signal priority modeling for bus rapid transit, saving ~5% of
IHS fuel [2]. Simulations for intelligent cruise control show 8.528.5% fuel savings when
10% of vehicles in a lane have intelligent cruise control [2]. Very close vehicle spacing
simulations show very close headways produce drafting effects that can save 515% of
fuel [2]. Another technology is a vehicle classifier [9]. This is connected to a PC, and
detects when a vehicle, particularly for trucks, approaches an intersection, then relays that
detection to the PC, which in turn delays a light change from green to red to give the
truck time to clear the intersection safely and without the energy expenditure of stopping.
The literature indicates that each truck stop consumes one-third of a gallon of fuel, a
direct saving of this system where deployed, indicating a system payback from counting
just two (saved time and fuel, the former worth about four times the value of the fuel) of
five benefits at a typical intersection of just under two years [9, p. 8]. Further
development in routing algorithms as a result of operations research applied to the
`traveling salesman problem' will no doubt lead to further efficiencies by eliminating
backtracking and overlapping routes. For example, Waste Management, Inc., recently
eliminated 761 trucks (or, we estimate from their 10-K, between 5% and 10% of IHS
fleet of trucks) and $91M in annual operating costs by deploying an optimization


                                                                                                6
algorithm to reroute IHS fleet [10]. DARPA-derived technologies for agent-based
computing infrastructure (the Agent based Logistics (ALP) protocol) may also contribute
significantly to this area [11].

Costs of Savings from IHS: Little data exists on the costs of savings. Nevertheless, Table
7-1 has estimated such a cost, but on a fuel basis only. While little data exists on costs of
systems-based IHS, there are four reasons for why we believe a carefully implemented
set of IHS technologies would be very cost effective overall and that the costs related to
the energy savings are insignificant and, unlike the fuel-based estimates in the Table of
3345¢/gal of fuel, approximately zero.

This first reason does not relate to energy savings, but to the value of the time that is
wasted during congestion. Reducing this waste of time more than pays for the investment
in the technology. In other words, irrespective of fuel savings, investment in IHS R&D
and deployment is something that will occur anyways due to the value of the saved time.
Of TTI's ~$70b in congestion-related waste, ~$63b is time, so the 15% saving that
reduced fuel burn by 1.2b gallons, or some $1.6b, reduces wasted time by another order
of magnitude, or about $10b. For this reason, we ascribe zero cost of energy saving--as
long as wasted time from congestion is reduced, IHS investments will be made
irrespective of any energy savings made. The NJTP experience is another indicator of
high cost effectiveness without regard to fuel, as a 12% revenue boost was seen from
ETC alone.

Second, the majority of the measures are quite simple installations that are largely fixed
and involve few physical installations relative to the number of vehicles influenced; only
a small fraction is vehicle-based technologies (which would imply a large number of
installations). A small number of relatively low-cost installations affecting an entire stock
of vehicles are likely to be cost effective.

Third, because these fixed technologies involve relatively little hardware, and mostly
involve automated and IT-centric systems with a relatively low embedded R&D cost,
they are likely to have relatively low up-front cost. For a given savings impact, this factor
tends to reduce amortized costs relative to hardware-heavy and vehicle-based
technologies.

Finally, IHS technologies generally influence operational fuel burn for entire vehicle
fleets in a given location. Because this influence is exerted on groups of technologies that
are already installed (i.e. existing vehicles), it does not involve changes in existing
infrastructure nor vehicles. It is therefore quite plausible that average implementation
costs would be quite small per gallon saved for this category relative to vehicle-based
technologies.




                                                                                            7
REFERENCES

1   Interlaboratory Working Group. 2000. Scenarios for a Clean Energy Future (Oak
    Ridge, TN: Oak Ridge National Laboratory and Berkeley, CA: Lawrence Berkeley
    National Laboratory), ORNL/CON-476 and LBNL-44029, November. Available
    online at http://www.ornl.gov/sci/eere/cef/

2   Shaheen, S.A. and Finason, R., "Intelligent Transportation Systems," in Encyclopedia
    of Energy, pp. 487-496, vol. 3, 2004

3   David Schrank and Tim Lomax, "2003 Urban Mobility Report," Texas
    Transportation Institute, The Texas A&M University System, September 2003.
    Available online at http://mobility.tamu.edu/ums/report/

4   David Schrank and Tim Lomax, "2003 Urban Mobility Report, Volume 2: Five
    Congestion Reduction Strategies and Their Effects on Mobility," Texas
    Transportation Institute, The Texas A&M University System, September 2003.
    Available online at http://mobility.tamu.edu/ums

5   David Schrank and Tim Lomax, "APPENDIX B, Methodology for 2003 Urban
    Mobility Report, Volume 2: Five Congestion Reduction Strategies and Their Effects
    on Mobility," Texas Transportation Institute, The Texas A&M University System,
    September 2003. Available online at http://mobility.tamu.edu/ums

6   See http://IHSdeployment2.ed.ornl.gov/IHS2002/default.asp

7   Kenworth Truck Company, "White Paper on Fuel Economy", October 2001,
    downloadable at www.kenworth.com/Kenworth_White_Paper.pdf.

8   US DOT, "New Jersey: Intelligent Transportation Systems," FHWA, section on IHS
    Benefit Data, excerpt online at http://www.IHS.dot.gov/staterpt/NJ.HTM

9   Srinivasa R. Sunkari, Hassan A. Charara, and Thomas Urbanik II, "Reducing Truck
    Stops at High-Speed Isolated Traffic Signals," Texas Transportation Institute,
    September 2000. Available online at http://tti.tamu.edu/product/catalog/reports/1439-
    8.pdf Also in Texas Transportation Researcher, Volume 37, Number 2 (2001),
    Available online at http://tti.tamu.edu/researcher/

10 Sharon Begley, "Did You Hear the One About the Salesman Who Traveled Better?",
   The Wall Street Journal, 23 April 2004.

11 Sherry Marcus et. al., "Predictive Assessment in the Advanced Logistics Project"
   Final Status Report, Defense Advanced Research Projects Agency (DARPA), order
   G531, Available online at
   http://cougaar.org/docman/view.php/16/30/Predictive_Assessment_Final_Report.pdf




                                                                                        8
Table 7­1: ESTIMATED HIGHWAY FUEL SAVINGS FROM INTELLIGENT HIGHWAY SYSTEMS
Selected Technologies


                                    Current          CW            SOA
                                     2000           2025           2025

Baseline data
                                                                              Percent of truck, lt truck, &
Calculated non-ITS redct'n                         27%             58%        auto use saved, 2025
Total Highway fuel use                20,444         23,747          13,783   T Btu/yr
Implied 25-yr change                               1.162           0.674      ratio
SOA/CW savings ratio                                                2.0       ratio

Estimated savings
   Subtotal, TTI figs (*)            0.85           0.99            1.15      b gal/y
   ETC                               0.10           0.12            0.13      b gal/y
   Adv Routing (**)                  0.30           0.35            0.40      b gal/y
   Other                             0.20           0.23            0.27      b gal/y
Total savings (various units)        1.45           1.68            1.95      BN gal/yr
                                         181            210             244   T Btu/yr
                                     34.5           40.1            46.5      M bbl/y
                                     0.09           0.11            0.13      M bbl/d
Percent reduction, ITS              0.89%          0.89%           1.77%      % of 2025


Estimated costs
All-in cost of saved energy (***)                  $14.01      $     18.73    $/bbl product
                                               $        0.33   $       0.45   $/gal product
Energy-related CSE                                   $0             $0        $/bbl
                                                     $0             $0        $/gal


(*) These include: incident management, signal coordination, ramp meters

(**) Advanced Routing Assumptions:
                                                                              Percent VMT rduct'n light
                                    0.25%          0.25%           0.25%      vehicles
Total non-truck                       14,978         11,750          10,893   T Btu/yr
                                    119.8           94.0            87.1      b gal/y

  (***) All-in costs here assumed equal to avg non-IHS highway transport CSE.