TEMPORARY LOSSES OF HIGHWAY CAPACITY AND IMPACTS ON PERFORMANCE: PHASE 2

Title | Contents | Acknowledgements | Exec. Summary
1. Intro | 2. Approach | 3. Crashes | 4. Breakdowns | 5. Work Zones | 6. Weather | 7. Signal Timing
8. RR Crossings | 9. Toll Facilities | 10. PUD | 11. Results Summary | 12. Next Steps | 13. References

9.  Toll Facilities

9.1  Methodology

The delay caused by toll facilities is an ever-increasing concern to the transportation community. Today, over 115 authorities operate nearly 250 facilities in 31 states. These facilities comprise more than 900 toll plazas. In the U.S, toll roads account for 4,800 miles of both rural and urban roads. Drivers passing through these toll plazas perform over five billion transactions, transactions that result in approximately 5.7 billion dollars of revenue annually.

Unlike other TLC impacts, the method used to determine delay from toll facilities is not directly based on estimating capacity reduction and modeling vehicle queue build-up and dissipation. This is because the total capacity of a toll facility is difficult to determine using available data. Although information on the number of toll lanes is generally available, all lanes are not always open. Without knowing the number of lanes open to traffic, it is difficult to determine capacity. Therefore, an alternative method was devised comprised of the following steps:

Step 1.  Identify all toll collection facilities and their locations

Step 2.  Determine the number of transactions at each facility

Step 3.  Determine the type of toll collection used at each facility

Step 4.  Estimate the total delay at each facility based on the number of transactions and the toll collection types used

Step 5.  Sum the delay for all collection facilities

These steps are described in the sections that follow.

9.1.1  Identifying Toll Facilities and Their Locations

The first step in estimating toll facility delay was to identify all toll facilities on freeways and principal arterials. The primary data source for this information was Toll Facilities in the United States: Bridges –Roads – Tunnels – Ferries published by the Federal Highway Administration's Office of Highway Policy Information (OHPI) (FHWA 2003).  This report contains selected information on all toll facilities in the United States, such as name, location, financing/operating authority, toll direction, mileage, and type of toll collection. The information is based on a survey of facilities in operation, financed, or under construction as of January 1, 2003.  TLC2 used the toll facilities listing in this report as its "master list" of toll facilities. Another document entitled "United States Toll Facilities" by Traffic Technologies, Inc. was used as a supplement to the OHPI report.[8]

9.1.2  Determining the Number of Transactions at Each Facility

Once all toll facilities on freeways and principal arterials were identified, the next step was to determine the number of transactions performed at each. The most detailed and reliable source of this information are the web pages and annual reports of the toll facility operating authorities.  However, some toll facility operating authorities regard the toll transaction information as business sensitive. In these instances, transaction information can only be obtained from the authority's legal department using the Freedom of Information Act (FOIA). For this study, transaction information was obtained for about 83 percent of the facilities within the study's scope.

For instances where transaction information from the toll authority was not available or could not be obtained, Annual Average Daily Traffic (AADT) information from HPMS was used as a surrogate for the number of daily transactions. As a final option, annual total toll transaction information from the document "United States Toll Facilities" was used if information was not available from the above sources.

9.1.3  Determining Toll Facility Type

In order to properly assign delay to a given facility, the methods used to determine toll rates and collect tolls must be considered.

Open vs. Closed Systems

For long toll roads, two common methods are used to determine the toll rate. The first is the open system. In the open system, there are many toll facilities along the main-line toll road, and drivers pay a toll at each facility they encounter.  For this type of facility, each transaction represents one vehicle paying toll at one facility. The other method of toll collection is called the closed system. In a closed system, typically used with ticketed toll facilities, the driver stops and receives a ticket stamped with the location of the entrance to the toll facility. He/she stops again upon exiting the facility and pays the toll, which is based on the point of entry and point of exit along the facility route.  Therefore, in the closed system, each transaction represents one vehicle stopping twice.

Toll Collection Method

Three types of toll collection are used at modern toll facilities:

Manual, or manned, toll facilities. The most common method is the manual tollbooth, where drivers pay the toll to an attendant who then raises a gate to permit the vehicle to pass. While this method is the most common, it causes the most delay of any of the facilities (the cost to maintain a toll collector at each booth also makes it expensive).

Coin-Basket Facility. The coin-basket facility uses an unmanned booth where drivers stop at the tollbooth and toss the exact change in coins into a basket.  The machine determines whether the correct amount of toll has been paid and, if so, raises a gate to permit the vehicle to pass. This system causes less delay than a wholly manual system. However, drivers must have exact change and must "hit" the basket with it. Otherwise, drivers can become stuck in the queue. Though common, these systems have begun to decline in popularity in recent years.

Electronic Toll Collection (ETC) systems. In the ETC system, drivers subscribe to a service and are given a transponder. Toll facilities are outfitted to detect the transponder and subtract the toll money from the driver's account when his/her vehicle passes the booth. This system creates the least amount of delay, since drivers must merely slow down, rather than stop, for payment.

Like the transaction data, information on toll collection types was obtained from the annual reports of facility operating authorities or from other information on their web sites. Toll collection method data was available for about 83 percent of facilities. For the remaining facilities, these shares were based on the average shares for the 83 percent for which data was available.

9.1.4  Estimating Delay at Each Facility

Delay from toll collection can be divided into three types: (1) "process" or "service" delay resulting from the time it takes to make the fee transaction, (2) "slow down" delay from slowing down and speeding up at the facility, and (3) queue delay in cases where a line of vehicles is waiting in line.

For this study, process time for manual and coin-basket facilities was assumed to be 6 seconds. This is loosely based on data from a report entitled Electronic Toll and Traffic Management (ETTM) Systems - A Synthesis of Highway Practice (NCHRP 1993). According to that study, manual collections facilities can process about 350 to 450 vehicles per hour per lane, while coin-basket facilities can process about 500 to 600. The 6-second process time corresponds with a capacity of 600 vehicles per hour per lane. It was assumed that ETC facilities can process a transaction in half as much time, although it may actually take even less.

Slow down delay for manual and coin-basket facilities was assumed to be 10 seconds. This was loosely based on the fact that it takes most vehicles from 5 to 12 seconds to reach a speed of 60 miles per hour from a complete stop—although this obviously varies due to driver behavior and vehicle performance. It was assumed that a similar amount of time was required to come to a complete stop. The study also assumed that slow down time for ETC facilities is about half of that for manual and coin-basket facilities (5 seconds) since vehicles need only slow down while driving through the facility, rather than coming to a complete stop.

For this study, queue delay was assumed to take place only at manual collection facilities during peak hours. Data collected by the New York City Metropolitan Transportation Authority (MTA) shows that peak-hour queue time at toll bridges and tunnels under their authority ranged from 18 to 29 seconds in 2002 and the first quarter of 2003. This study, therefore, assumed that average queue time at manual facilities during peak hours was 20 seconds. It was also assumed that no queue delay is experienced during non-peak hours at these facilities. Several studies in the literature show that ETC facilities reduce delay to the point that no, or almost no, queue delay was experienced (Wilbur Smith & Assoc. 2001; Vollmer Associated LLP 2000; Saka 2002). Therefore, it is assumed that queue delay for ETC transactions is zero.

The per-vehicle delay assumptions used in this study are summarized in the table below.

In this study, total delay at a given toll facility was calculated as the sum of the delay for (1) the manual collection lanes during peak hours, (2) manual collection lanes during off-peak hours, and (3) electronic toll collection lanes during peak and off-peak hours—delay for coin-basket lanes is calculated the same as for manual lanes. The following equation was used:

Equation 26:

where

Delay_MP = average process delay per vehicle for manual collection lanes

Delay_MS = average slowdown delay per vehicle for manual collection lanes

Delay_MQ = average queue delay per vehicle for manual collection lanes

Delay_EP = average process delay per vehicle for ETC lanes

Delay_ES = average slowdown delay per vehicle for ETC lanes

Transactions_Total = total number of transactions at a given facility

Hrs_Peak = number of hours a day a facility is closest to its maximum capacity

Hrs_Total = total number of hours a day the facility is open

Lanes_ETC = number of lanes at a given facility that utilize an ETC system

Lanes_Total = total number of toll collection lanes at the facility (both manual and ETC)

9.2  Results

The study estimates that 21 million vehicle-hours of delay were caused by toll collection facilities in 1999. The average delay per transaction was 11.9 seconds. This resulted in 33.6 person-hours of delay.  Nearly three-quarters of this delay, was incurred at manual and coin-basket (i.e., non-ETC) collection facilities.

Figure 27. Over three quarters of the delay from toll facilities in 1999 occurred in lanes using non-electronic collection methods.

9.3  Reliability

9.3.1  Methodology

The methodology for estimating toll facility delay is simple and straightforward. It is accorded a high level of confidence.

9.3.2  Data & Key Assumptions

The service rates for both manual and electronic toll collection system are based on many recently published studies.

Slowdown and speed resumption delays are based on acceleration rates for all passenger vehicles from recently published reports. Therefore, this information is qualified as having a high degree of confidence.

Assumptions regarding queue formation and queue delays are based solely on a study of toll facilities operated by the New York City Metropolitan Transportation Authority; there is little information on queue delay at other toll facilities. Therefore, this information is qualified as having a low degree of confidence.

The total number of transactions and the percent of transactions processed by ETC are primarily based on toll operator annual reports.  Data were available for 83% of toll facilities. Other transaction information is based either on traffic volume information or older (1 to 2 year-old) transaction information. This information is accorded a medium to high degree of confidence.

8. The United States Toll Facilities document was taken from a web site that is no longer operational.

Previous | Table of Contents | Next