willingness to pay for high-occupancy toll lanesceprofs.tamu.edu/mburris/papers/trr 2297 - paying...

47

Transportation Research Record: Journal of the Transportation Research Board, No. 2297, Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 47–55.DOI: 10.3141/2297-06

M. Burris, S. Nelson, P. Gupta, and Y. Cho, Zachry Department of Civil Engineer-ing, Texas A&M University, 3136 TAMU, College Station, TX 77843-3136. P. Kelly, 615 North 600 West #8, Provo, UT 84601. Corresponding author: M. Burris, [email protected].

current conditions. Another option is dynamic tolling, in which the toll price is adjusted frequently based on real-time traffic conditions. Dynamic tolling tends to be less politically palatable than simple variable tolling since users do not know the exact toll they will pay until they arrive at the facility. However, it may be that because of similar traffic from one day to the next, the dynamic toll may be approximately the same price at a given time of day every day. In that case it would be plausible that an agency would simply utilize a variable tolling strategy to avoid any potential political backlash from dynamic tolling.

The goal of this research is to improve the understanding of driver behavior in HOT-lane corridors and how variable dynamic tolling truly is. The specific objectives were to

• Determine travelers’ willingness to pay for travel time savings (TTS) on HOT lanes including the distribution of these values over the toll-paying population and• Determine the uniformity of the toll price on dynamically priced

facilities.

Literature review

There are many studies of travelers’ willingness to pay for TTS with many proposed behavioral and travel characteristics influencing the value of travel time (VOT) and variations in an individual’s willing-ness to pay. Trip-related factors include trip purpose, trip length, integrating trip chain, duration, mode, TTS, the time of day the trip is made, the ratio of toll to travel cost, comfort level, safety, conve-nience, and travel time reliability. Socioeconomic factors include income level, age, gender, household composition, and level of edu-cation. Among these diverse factors, trip purpose and income level are crucial in determining an individual’s willingness to pay (1).

Many different studies examining travelers’ VOT exist; however, this study will focus specifically on MLs and HOT lanes. In this case travelers are able to make last-second decisions on which lanes to choose and can see the two options available to them. In addition, MLs and HOT lanes only have a small percentage of the freeway’s capacity available for SOVs to purchase, so it is likely the SOV drivers with the highest VOT are the ones who will choose to pay for the use of the lanes. Thus, empirical analysis of these travelers may reveal VOTs higher than is typical for a freeway facility.

Stated-preference surveys are one method of determining what factors go into a traveler’s decision-making process when he or she decides whether to use MLs. A survey was conducted of users of the Katy Freeway MLs in Houston, Texas, by Patil et al. (2). It was shown that users who are late for an appointment, have an important

Willingness to Pay for High-Occupancy Toll Lanesempirical analysis from i-15 and i-394

Mark Burris, Scott Nelson, Pete Kelly, Partha Gupta, and Youngjae Cho

This research examined the willingness of travelers to pay for travel time savings (TTS) on the I-394 MnPASS express lanes in Minneapolis, Minnesota, and the I-15 express lanes in San Diego, California. The find-ings from both facilities indicate that many travelers are willing to pay a toll for a small TTS. The revealed preference values of time for the I-394 paying customers averaged $73/h in the morning and $116/h in the afternoon. Travelers on the I-15 express lanes received slightly higher TTS than on the I-394 and typically had a lower willingness to pay. Their revealed preference values of time averaged $49/h in the morning and $54/h in the afternoon. On the basis of the magnitude of these val-ues, it is likely that travelers are paying for more than just TTS, possibly travel time reliability also. This finding means that these lanes likely have added value to travelers beyond time savings. The variability of the dynamic toll prices on the I-394 MnPASS express lanes in Minnesota and the I-15 express lanes in San Diego was also examined. Toll rates during the morning and afternoon peak hours varied considerably, with tolls ranging from $0.50 to $8.00. Conversely, off-peak times showed little to no variation. These trends were similar for different days of the week, and even from year to year. These trends were common across both I-394 and I-15 express lanes, although more variation was found on I-15. Tolls varied little during off-peak periods.

The current economic and political climates have encouraged the use of alternative means of obtaining funding for transportation infra-structure projects. Many agencies are turning to tolling to help pay for these projects. Tolling allows an agency to recover some of the cost of the project directly from users. One option is managed lanes (MLs), which are typically lanes added to an existing roadway and are controlled to remain free-flowing. MLs can be operated by using sev-eral management strategies including high-occupancy vehicle (HOV) and high-occupancy toll (HOT) strategies. HOV lanes are lanes purely reserved for multioccupant vehicles, and HOT lanes typically allow single-occupant vehicles (SOVs) to travel in the MLs for a toll and higher-occupancy vehicles to use the lanes toll free.

Several options exist for determining the toll to charge on these facilities. One option is variable tolling, in which a set rate is charged at a certain time of day and cannot be modified in real time based on

48 Transportation Research Record 2297

appointment, or are worried that they will be late if they use the general purpose lanes (GPLs) will have the highest willingness to pay for the MLs. In addition, it was shown that as a traveler’s household income increases, the likelihood of using the MLs increases (2). A more recent survey of travelers on the Katy Freeway MLs also examined their willingness to pay for TTS and reduced travel time variabil ity. The study compared the stated preference survey responses of travelers to real-world data on ML usage. The stated preference results include a willingness to pay of $22/h for TTS and $28/h for reduced variability. Analysis of the real-world data found an average willingness to pay of $51/h of TTS on the lane, surprisingly close to the combined value for TTS and improved travel time reliability of the lanes (3).

A panel survey of users of San Diego’s I-15 express lanes (named FasTrak) conducted by Supernak et al. indicated that FasTrak users were predominantly using FasTrak for the TTS (4). However, results also indicated that travel time reliability was an important factor in choosing the express lanes, especially during the morning peak period (4). Brownstone and Small, with the data from SR-91 and I-15 HOT lanes, estimated the VOT to be 95% to 140% of the median value of travel time (5).

Sullivan found that a traveler’s primary reason for using the SR-91 express lanes (91-X) was TTS (6). However, one-third of the respon-dents cited reasons other than simply TTS. Easier driving and safety ranked high among the other reasons noted by drivers who paid to use the lane during off-peak periods. About 58% of express lane users believed that these lanes were safer than the free lanes. A very small percentage of the respondents cited reasons like enjoyment in passing others, feeling prestige, or a low risk of a speeding ticket. Less congestion, less aggressive driving, no large vehicles, better enforcement, and better emergency response were among the rea-sons why the respondents thought that 91-X lanes were safer than free lanes. Fourteen percent of the 91-X respondents, however, believed that the express lanes were less safe due to high speed, aggressive driving, no shoulder, and less enforcement. Fifteen percent of the respondents had a company-provided transponder and did not have to pay to use it, which could be a reason why some travelers used express lanes during the nonpeak period (6).

Study Location

i-394 MnPaSS express Lanes

I-394 runs roughly east–west between downtown Minneapolis and the outlying suburb of Wayzata. This route serves as one of the primary routes for commuters to traverse between the western

suburbs and points east, including downtown Minneapolis (see Figure 1).

The operation of the I-394 HOT lanes is similar to the opera-tion of other HOT lanes in that SOVs have the option to pay to use the MnPASS lanes. HOVs, mass transit, emergency vehicles, and motorcyclists can use the lanes free of charge. Tolls are only charged during peak periods in the peak direction for SOVs, so SOVs can use the MLs during off-peak hours free of charge. MnPASS determines the toll prices in real time based on the level of service in the MLs. The Minnesota Department of Transportation has placed sensors in the pavement at ½-mi intervals, which are used to determine important traffic information including current volume and whether the volume is increasing or decreasing. Dynamic toll pricing is utilized to help ensure Level of Service C, which roughly equates to a free-flow speed of 50 to 55 mph, for the MLs. When traffic volumes are high in the MLs, MnPASS charges a higher toll in an effort to discourage drivers from using the managed facility to keep traffic flowing at 50 to 55 mph. When traffic volumes in the MLs are low, low toll prices are charged to encourage the use of the MLs. Toll prices are calculated every 3 min on the basis of traffic conditions and sensor information, and thus can potentially change every 3 min. MnPASS considers four possible toll options for each 3-min period and chooses to charge the lowest price among these four. The current toll is posted on electronic signs located just upstream of entrances to the MLs. The tolls range from $0.25 to $8 and average $1 to $4 during peak hours (7).

Within the study area, I-394 contains two unique HOT-lane configurations and operation schemes:

1. A nonreversible section of HOT lanes with one lane per direc-tion, adjacent to the GPLs, is in place from Wayzata Boulevard (west of I-494) to just west of MN-100. During the workweek, westbound SOV traffic is charged to use the HOT lanes between 2:00 and 7:00 p.m. Eastbound SOV traffic is charged a toll to use the HOT lanes between 6:00 and 10:00 a.m. During off-peak times and on weekends, the lanes are opened to all traffic toll-free.

2. A separated two-lane reversible section is operated from just west of MN-100 to I-94 in downtown Minneapolis. This section is in operation Monday through Friday to accommodate eastbound traffic between 6:00 a.m. and 1:00 p.m. and westbound traffic between 2:00 p.m. and 5:00 a.m. with SOVs being charged a toll at all times during operation. The separated reversible lanes are not accessible during nonpeak hours.

Entrance and Exit points

FIGURE 1 I-394 MnPASS express lane corridor. (Source: MnPASS.)

Burris, Nelson, Kelly, Gupta, and Cho 49

I-15 Express Lanes

The I-15 express lanes in California are a 16-mi stretch of MLs running north–south from SR-163 in the south to Centre City Park-way in Escondido. This project consists of four HOT lanes (two per direction), which will eventually span 20 mi of the I-15 corridor from SR-163 to SR-78 in San Diego County. The project is being constructed in three segments. The middle segment, which spans 8 mi between SR-56 and Centre City Parkway, was completed in March 2009 and is currently in service. The north segment, which will reach SR-78, is scheduled for completion in 2011, and the south segment, which already has two functional lanes, will be expanded to four lanes by 2012 (8) (see Figure 2). Vehicles can enter the HOT lanes by using intermediate access points and direct access ramps at certain locations along the corridor.

On the express lanes, vehicles with two or more occupants, motorcycles, transit vehicles, and certain clean air vehicles are per-mitted to enter free of charge, whereas solo drivers pay to use the lanes (9). Dynamic pricing ensures continuous free flow by adjust-ing the per mile toll rate up or down every few minutes depending on the amount of traffic in the MLs. The toll depends on where one enters and exits the FasTrak express lanes and the volume of vehicles in the toll lanes. The toll is posted on electronic variable message signs located just upstream of entrances to the FasTrak lanes and is charged by using FasTrak transponders (9). The toll is charged on a per-mile basis, and the total toll amounts can range from $0.50 to $8. Under normal peak-hour conditions the highest toll is $4 (10).

The operation time for the I-15 express lanes during the study period was as follows (9):

• The middle 8-mi (two lanes in each direction) portion from SR-56 to Rancho Bernardo Road is always open in both directions and• The operations on the south 8-mi portion (two reversible lanes)

from SR-163 to SR-56 are– Monday to Thursday: southbound, 5:30 a.m. to noon, and

northbound, 1:00 p.m. to 7:00 p.m., and– Friday: southbound, 5:45 a.m. to 11:00 a.m., and northbound,

Friday, 5:00 p.m., to Monday, 5:00 a.m.

DATA

I-394 MnPASS Express Lanes

Tolling Data

This research used tolling data from 2008 for the I-394 HOT lanes. This information included the date, time, entry location, exit location, and toll paid for every toll-paying vehicle during this year (roughly 1 million total transactions).

Loop Detector Data

Loop detectors were used to collect traffic speed and volume data. The loop detector data are the property of the Minnesota Department of Transportation. These data are available to the public through the department website (www.dot.state.mn.us/tmc/trafficinfo/developers.html). Detector data are provided to the public in real time in XML format, which is updated every 30 s. The XML files contain vol-ume, occupancy, speed, and flow data for each detector in the Twin Cities metropolitan area. With this website, data from all mainline GPL and HOT-lane detectors located in the study section were obtained.

The total number of detectors on the GPLs is 44 in the westbound direction and 51 in the eastbound direction. There are 12 detectors on the westbound nonreversible HOT lane, 15 on the eastbound nonreversible HOT lane, and 11 detectors in the reversible section.

I-15 Express Lanes

Toll Payment Data

The San Diego Association of Governments provided toll payment data for each paying traveler of the FasTrak lanes between March 2009 and June 2010. The toll data included the starting and ending plaza, total toll amount paid, entry time, and exit time. All told, there were approximately 2.5 million toll transactions during that period.

Vehicle Detector Station Data

Several vehicle detector stations (VDS) were used to collect traffic speed and volume data. A VDS is a logical grouping of detectors of a particular type at a location (11). VDS data for a particular sec-tion contain speed and flow of all the lanes in a particular direction at that section. The VDS data are the property of the California Department of Transportation Performance Measurement System

FIGURE 2 I-15 express lanes. (Source: http://fastrak.511sd.com.)


(Caltrans PeMS) and are available to the public through the PeMS website (pems.dot.ca.gov). Data from all GPL and HOT-lane VDS located in the study section were obtained from this website. The total number of VDS on the GPLs is 15 in the northbound direction and 14 in the southbound direction. The volume and speed data for all the GPL and HOT-lane VDS were obtained from the website for every 5 min of every day from March 2009 to June 2010.

There were only one VDS on the northbound HOT lanes and one VDS on the southbound HOT lanes, and therefore there was no good source of travel time data on the HOT lanes. However, the toll transaction data contained the entry and exit time and location of each toll-paying vehicle. The HOT-lane travel time of an individual traveler was easily found by using the difference between these two times. Also, since the distance between any entry and exit plaza combination was known, the average speed of the traveler could be calculated easily on the basis of the sections they traveled. By using the speeds of all the toll-paying travelers through any section within any 5-min period, the average HOT-lane speeds were calculated for that time period.

data reduction and cLeaning

For both I-394 and I-15 the data reduction involved calculating the average GPL speed, the average GPL volume, and the average HOT-lane speed value for each section of the highway. Since 5-min time intervals were used to extract loop detector data for the morn-ing and afternoon operating times, the average toll amount and the average number of paying travelers was also computed over 5-min intervals.

For speeds, readings less than or equal to zero and greater than or equal to 100 mph were removed from the analysis. A volume of less than zero was treated as an error in the data. The detector data were then averaged (weighted by number of vehicles) for each sec-tion of the freeway. The output included the average GPL speed and volume and the average HOT-lane speed and volume. A similar pro-cedure was performed to examine the toll data; no clearly erroneous data were found in the toll data set.

For the toll variability analysis, the tolling scheme for the I-394 corridor split the HOT-lane users into three distinct groups with each group paying a different toll based on where they traveled within the system. The first group consisted of users who traveled the entire length. These users essentially traveled both halves of the MnPASS system, incurring the maximum toll possible. The second group consisted of users who traveled on the separated reversible lanes and paid a lesser toll. The third group consisted of users who paid a lesser toll by traveling only within plazas west of the sepa-rated reversible lanes. The toll variability analysis procedure varied somewhat for the I-15 express lanes since these lanes used a toll-per-mile scheme. With the total toll amounts paid by the user and the distances traveled based on their starting and ending plazas, the toll rate in dollars per mile was determined.

reSuLtS

i-394 MnPaSS express Lanes (2008 only)

The morning operation time for MnPASS is from 6:00 to 10:00 a.m. In this analysis the travel time required from west of Carlson Park-way to near I-94 was examined, a distance of 12.4 mi. This distance is essentially the entire length of the HOT lanes but does not account for any possible advantage or disadvantage for these travelers at the far east end of the lanes when they exit and merge with GPL traffic. Overall, 79% of the travelers in the HOT lanes traveled 12.4 mi in less than 12.0 min, whereas only 54% of all travelers in GPLs spent less than 12.0 min to travel that distance (see Figure 3). The find-ings support the hypothesis that speeds on HOT lanes are faster and more reliable than those on GPLs. The small difference between GPL and HOT-lane speeds resulted in very small TTS. Thirty-five percent of travelers on the MnPASS lanes paid for an average TTS of less than a minute.

The toll ranged between $0.25 and $8.00 during the a.m. period in 2008. Approximately 50% of the toll transactions on the HOT lane were for $1.25 or less (Figure 4a). Because the travel time dif-ferences between the HOT lanes and the GPLs were very small, and

FIGURE 3 GPL and HOT-lane travel time on I-394 (6:00 a.m. to 10:00 a.m., eastbound).


the toll was $1.25 or less for 50% of the HOT lane trips, the willing-ness to pay results indicate that morning travelers have a wide range of willingness to pay. Overall, 7.5% of the travelers were willing to pay for the HOT lane during the entire morning period. Also, the median value of TTS for the toll-paying travelers was extremely high, approximately $73/h (Figure 4b).

The same analysis was performed on all travelers during the peak hour of morning traffic (7:30 to 8:30 a.m.). Results were similar although there was slightly more TTS and slightly higher tolls. During the morning peak hour, the average proportion of paying travelers rarely exceeds 13%. The tolls ranged between $0.25 and $8.00. Approximately 50% of the toll transactions during the morn-ing peak period were for $2.00 or less. The travel time difference between HOT lanes and GPLs remains small even during the peak period. The median value of TTS for the toll-paying travelers was $78/h, slightly higher than that for the entire morning.

Next, the entire afternoon period (2:00 to 7:00 p.m.) and the afternoon peak hour (5:00 to 6:00 p.m.) were examined for the westbound direction. The results were similar to the morning results.

Overall, 4.7% of the travelers were willing to pay for the HOT lane during the entire afternoon period. The median value of TTS for the toll-paying travelers was $116/h for the entire afternoon and $124/h during the peak. Approximately 70% of paying HOT-lane customers were paying for 1 min or less of TTS during the afternoon period.

The users of I-394 were sorted into four classifications based on their amount of HOT-lane usage in 2008: rare (less than or equal to 12 trips), occasional (greater than 12 and less than or equal to 104 trips), moderate (greater than 105 and less than or equal to 260 trips), and frequent (greater than 260 trips) to determine whether a user’s VOT varied on the basis of the frequency of their HOT-lane usage. It was shown from the results that there was essentially no difference in a user’s VOT based on their frequency of use (see Figure 5).

The amount of variation in the toll price for a given 5-min period for all weekdays in 2008 was examined. It was found that peak-hour operation periods had much higher toll prices and variation in toll prices than off-peak periods. Also, it was shown that the toll vari-ances tended to be higher in the a.m. peak than in the p.m. peak. This

(a) (b)

FIGURE 4 Distributions on I-395 (6:00 a.m. to 10:00 a.m., eastbound): (a) tolls and (b) VOT.

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350 400

Cum

ulat

ive

Per

cent

age

of T

rave

lers

VOT ($/hr)

RareOccasionalIrregularFrequent

FIGURE 5 VOT distribution for users of I-395 HOT lanes by frequency of usage.


finding was expected since the median toll amounts were generally greater during the a.m. peak than during the p.m. peak. Five off-peak time periods were analyzed. For the off-peak times, the toll variances were small or even nonexistent (Figure 6).

Drivers who only used the eastern and western halves of the facil-ity paid lower tolls than those who used the entire facility. Accord-ingly, the variance in the toll price paid by partial-length users tended to be less than the variance paid by entire-length users but followed similar trends.

Examining different days of the week also showed similar results for all days of the week, but with Mondays through Thursdays hav-

ing slightly higher toll variation than Fridays since there were more congestion and higher toll prices on those days.

i-15 express Lanes (april 2009 to June 2010)

The Monday through Thursday morning operation period for I-15 is from 5:30 a.m. until noon. It was shown that roughly 80% of travel-ers using the GPLs spent 15 min or less to travel the entire length of the study area, and over 95% of travelers using the HOT lanes spent 15 min or less to travel the same length (Figure 7). The median toll

Per

cen

tag

e o

f U

sers

0

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

5.50

6.00

6.50

7.00

7.50

10

20

30

40

50

60

70

80

90

100

Toll Paid ($)

5:30 PM (Peak)

6:30 AM (Off-Peak)

FIGURE 6 Toll variation on I-395 for users who traveled entire length (2008).

0 5 10 15 20 25 300

10

20

30

40

50

60

70

80

90

100

Travel Time (min)

Cum

ulat

ive

Per

cent

age

of T

rips

HOTGPL

FIGURE 7 GPL and HOT-lane travel time on I-15 (5:30 a.m. to noon, southbound).


paid by express lane users was $1.20, and the median toll per mile paid was $0.15/mi (see Figure 8a). The median TTS for express lane users was 1.16 min. The median VOT was $49.22/h, and 7.22% of travelers were willing to pay for express lane service (excluding HOVs, which travel toll free) (see Figure 8b).

The same analysis was performed on travelers during only the morning peak hour of operation on Monday through Thursday (7:30 a.m. until 8:30 a.m.). During this period, the median TTS for express lane users was 1.05 min. The median VOT was $82.42/h, and 14.3% of travelers were willing to pay for express lane service (excluding HOVs).

Next, the entire afternoon period (1:00 p.m. until 7:00 p.m.) and afternoon peak hour (5:00 p.m. until 6:00 p.m.) were analyzed. Dur-ing the entire afternoon period roughly 70% of travelers using the GPLs spent 15 min or less to travel the entire length, whereas more

than 95% of travelers using the HOT lanes spent 15 min or less to travel the same length. The median toll paid by express lane users was $1.40 and the median toll per mile paid was $0.18/mi. The median TTS for express lane users was 1.11 min. The median VOT was $54.39/h, and 6.9% of travelers were willing to pay for express lane service (excluding HOVs) (see Figure 9).

It was found that the median VOT was much higher during the peak hour of the afternoon ($71.41/h) than during other times. In addition 11.1% of travelers were willing to pay for express lane service (excluding HOVs) during the peak hour.

The operation of I-15 was different from the operation of I-394 in that the HOT lanes on Friday and the weekends were operated differently than during the rest of the week. Analyses of the morning and afternoon on Friday, as well as of the weekends as a whole, were performed. Results were generally similar to those for the Monday

(a) (b)

FIGURE 8 Distributions on I-15 (5:30 a.m. to noon, southbound): (a) tolls per mile and (b) VOT.

0 20 40 60 80 100 120 140 160 180 2000

0.5

1

1.5

2

2.5

3

3.5

4

Cost of Travel Time Savings ($/hour)

Per

cent

age

of P

ayin

g T

rips

0

10

20

30

40

50

60

70

80

90

100

Cum

ulat

ive

Per

cent

age

of P

ayin

g T

rips

FIGURE 9 VOT distribution on I-15 (1:00 p.m. to 7:00 p.m., northbound).


through Thursday morning periods, although the Friday morning VOT was typically lower (for example, the Friday morning median VOT was $43.49/h) since less congestion was encountered by users.

Weekend periods showed very little congestion on the GPLs; roughly 95% of all travelers spent 15 min or less to travel the entire length. The median toll paid by express lane users was $0.50 and the median toll paid per mile was $0.10. In this case, TTS and VOT were insignificant compared with weekday peri-ods and only 1.5% of travelers were willing to pay for express lane service (excluding HOVs).

A toll variability analysis for I-15, similar to the analysis per-formed on I-394, was performed on toll data from April 2009 to June 2010. The tolling scheme used on I-15 differed from that used on I-394 in that tolls were charged on a per-mile basis. However, toll variability and distribution trends were somewhat similar for both projects. It was found that the toll rate per mile was higher in peak periods than in off-peak periods. Toll variation for peak periods was higher than on I-394. During off-peak times, much more concen-trated and less variable toll distributions were encountered than dur-ing peak times. Although the off-peak periods for I-15 showed more variance than those at similar times on I-394, there was still only a small amount of variation during off-peak periods (Figure 10).

concLuSion

The average travel time on GPLs for both I-394 and I-15, during both the peak and off-peak periods, was not much higher than that on the adjacent HOT lanes. This finding indicated that users were willing to pay for even a relatively small TTS. Combined with tolls that typically ranged between 50 cents and $2, the mean values of time for travelers paying to use the HOT lanes were quite high. For example, on I-394, mean VOT for morning HOT-lane travelers was $78/h and for afternoon travelers it was $116/h.

MnPASS travelers may have several reasons to use the lanes even though the overall TTS are very small. For example, the reversible-lane section may help some travelers avoid bottlenecks entering downtown. However, it was found that a large number of travel-ers paid to use the nonreversible portion of the HOT lanes with similarly high VOTs. It could be that travelers enjoy queue jump-ing and the satisfaction of getting a premium service. Possibly it is the travel time reliability that is the most important. According to a recent survey of MnPASS users, a majority of drivers agreed that MnPASS provides “a fast, safe, reliable commute every time” (12). Also, it may be that since these are the few travelers willing to pay for a premium service, they are at the high end of the spectrum of VOT on this facility. However, because of how very high the VOTs were, the HOT-lane customers on both I-15 and I-394 appear to be paying for more than just TTS.

It was found that the dynamic tolling systems on I-15 and I-394 had a significant amount of variance in the toll prices during peak periods. Off-peak periods showed little to no variation in the toll prices, although I-15 had a slightly greater amount of toll variance. Therefore, it was shown that dynamic tolling charges a significantly different toll than variable tolling during peak periods when volumes are highest.

acknowLedgMentS

This research was funded as independent research by the South-western University Transportation Center, which is funded, in part, by the State of Texas. Much of the research was conducted with the support of the Undergraduate Transportation Scholars Program at the Southwestern University Transportation Center, and the authors are grateful for the support. The authors also thank the Min-nesota Department of Transportation for sharing the I-394 traffic and toll data along with COFIROUTE USA for sending the I-394

0.05

0.15

0.25

0.35

0.45

Per

cen

tag

e o

f U

sers

0

0.00

-0.0

5

0.05

-0.1

0

0.10

-0.1

5

0.15

-0.2

0

0.20

-0.2

5

0.25

-0.3

0

0.30

-0.3

5

0.35

-0.4

0

0.40

-0.4

5

0.45

-0.5

0

0.50

-0.5

5

0.55

-0.6

0

0.60

-0.6

5

0.65

-0.7

0

0.70

-0.7

5

0.75

-0.8

0

0.80

-0.8

5

0.85

-0.9

0

0.90

-0.9

5>0

.95

0.1

0. 2

0.3

0.4

Toll Per Mile ($/mile)

5:30 PM (Peak)

6:30 AM (Off-Peak)

FIGURE 10 Toll variation on I-15.


toll transaction data. In addition, the authors thank the San Diego Association of Governments for sending toll data and the Califor-nia Department of Transportation for making traffic data available through PeMS.

referenceS

1. Jian, M., and T. Morikawa. Theoretical Analysis on the Variation of Value of Travel Time Savings. Transportation Research Part A, Vol. 38, 2004, pp. 551–571.

2. Patil, S., M. Burris, W. D. Shaw, and S. Concas. Variation in the Value of Travel Time Savings and Its Impact on the Benefits of Managed Lanes. Transportation Planning and Technology, Vol. 34, No. 6, Aug. 2011, pp. 547–567.

3. Devarasetty, P. C., M. Burris, and W. D. Shaw. The Value of Travel Time and Reliability of Houston Managed Lane Travelers: Evidence from Stated Preference Survey and Actual Usage Data. Presented at 91st Annual Meeting of the Transportation Research Board, Washington, D.C., 2012.

4. Supernak, J., J. Golob, T. F. Golob, C. Kaschade, C. Kazimi, E. Schraffler, and D. Steffey. San Diego’s Interstate 15 Congestion Pricing Project: Atti-tudinal, Behavioral, and Institutional Issues. In Transportation Research Record: Journal of the Transportation Research Board, No. 1812, Trans-portation Research Board of the National Academies, Washington, D.C., 2002, pp. 78–86.

5. Brownstone, D., and K. A. Small. Valuing Time and Reliability: Assessing the Evidence from Road Pricing Demonstrations. Transpor-tation Research Part A, Vol. 39, 2005, pp. 279–293.

6. Sullivan, E. C. Continuation Study to Evaluate the Impacts of the SR 91 Value-Priced Express Lanes: Final Report. California Polytechnic State University, San Luis Obispo, 2000.

7. MnPASS. Minnesota Department of Transportation, Minneapolis. www.mnpass.org. Accessed July 27, 2011.

8. I-15 Express Lanes Fact Sheet. San Diego Association of Governments. http://www.sandag.org/uploads/publicationid/publicationid_6_1065.pdf. Accessed July 27, 2011.

9. FasTrak—Getting Started. San Diego Association of Governments. http://fastrak.511sd.com/GettingStarted.aspx. Accessed July 2, 2011.

10. Frequently Asked Questions as of May 1, 2007. San Diego Asso-ciation of Governments. http://www.sandag.org/services/fastrak/faqs.asp?classid=29&fuseaction=home.classhome#4. Accessed July 3, 2011.

11. Caltrans PeMS. California Department of Transportation. http://pems.dot.ca.gov. Accessed March 5, 2011.

12. Users of Minneapolis I-394 Toll Express ’Selves Happy, Tolls Soon on I-35W Lanes. TOLLROADSnews. http://www.tollroadsnews.com/node/4292. Accessed Aug. 9, 2009.

The contents of this paper reflect the views of the authors, who are responsible for the facts and the accuracy of the data. The contents do not necessarily reflect the official view or policies of the Southwestern University Transportation Center.

The Congestion Pricing Committee peer-reviewed this paper.

willingness to pay for high-occupancy toll lanesceprofs.tamu.edu/mburris/papers/trr 2297 - paying...

Documents