Tolling and Pricing Program - Links to Tolling and Pricing Program Home

 

banner

Table of Contents

Executive Summary

Chapter 1

Chapter 2

Chapter 3

Chapter 4

Chapter 5

Chapter 6

Appendix


Click images to view a larger version

Back to Top
executive summary header

I-95 between downtown Miami and I-595 near Fort Lauderdale is one of the most heavily traveled portions of urban interstate highway in America. Weekday traffic volumes between the Golden Glades Interchange and S.R. 112 reach as high as 300,000 vehicles per day, resulting in high levels of congestion in morning and afternoon peak hours.
figure ES-1 location map
In an attempt to examine means of possibly improving capacity and operating conditions on this heavily traveled section of I-95, the Florida Department of Transportation selected Wilbur Smith Associates to perform a comprehensive traffic and revenue study of proposed managed lanes in the corridor. This report documents the results of that study, which was conducted in phases primarily in 2005 and 2006.

Figure ES-1 shows the corridor covered by the analysis. Between I-595 to the north and the Golden Glades Interchange in northern Miami-Dade County, the project would involve simple conversion of existing single-HOV lanes in each direction to high-occupancy toll (HOT) operation. South of the Golden Glades Interchange, six different project configurations and operational scenarios were tested as part of the alternatives analysis. These ranged from simple conversion of existing HOV lanes to HOT to deployment of reversible HOT lanes to construction of an elevated roadway with up to four lanes of additional high-speed traffic.

Managed lanes involve the creation and preservation of a portion of total capacity which would be generally free from congestion. Depending on operating scenario, the managed lanes may be open toll-free to high-occupant vehicles, while non-HOVs would be permitted to use the lanes for a toll charge. Tolls would not be charged on existing toll-free general purpose lanes. All tolls would be collected electronically, and demand and congestion in the managed lanes would be managed by raising or lowering toll rates depending on time and direction of travel.

Study Overview

The study included the development of a detailed operations profile, including quantification of average travel speeds and bottleneck areas. Extensive surveys were undertaken, including both origin-destination and extensive stated preference surveys. These surveys provide useful information regarding values of time and propensity to use the toll lanes from among travelers currently using I-95.

Several focus groups were also undertaken to gauge citizen reaction to the proposed managed lane concept. A detailed economic review was also completed culminating in the development of updated socioeconomic forecasts and new trip tables for use in the analysis.

Three levels of models were developed to analyze the different alternatives:

  • The regional SERPM model was used to estimate total demand in the future in the I-95 corridor;
  • A micro-model, calibrated by hour and travel direction, was used to estimate the share of traffic such that could be expected to use the managed lanes versus the toll-free general purpose lanes; and
  • A detailed microsimulation model of the entire 20-mile corridor was used to identify pockets of congestion and bottleneck locations.

Six different project options were analyzed in the study. This included development of preliminary traffic and revenue estimates, optimum tolls and impacts on operating conditions along the corridor. After review of these, FDOT selected two alternatives as the most promising. These were then subjected to a refined traffic and revenue analysis, culminating in forecasts of annual net revenue.

Current Conditions

Traffic volumes along portions of I-95 are extremely high, reaching as much as 300,000 per typical weekday. The entire section of the corridor between S.R. 112 and the Golden Glades Interchange has average weekday volumes of 290,000 or more. North of the Golden Glades Interchange, average daily traffic ranges from 192,000 to 285,000, just south of the I-595 Interchange.

Essentially, this entire portion of I-95 includes eight general purpose lanes, four in each direction, plus two HOV lanes, one in each direction. Vehicles with two or more occupants are currently allowed to use HOV lanes during peak periods.

Traffic is heavily constrained during peak periods and even during midday hours volumes decline only marginally from their peak levels. Morning peak hour speeds were found to be as low as 21.3 MPH in speed and delay runs conducted in 2005. Afternoon peak congested speeds averaged 27.6 MPH during peak hours.

Focus groups were conducted early in the study to assess I-95 traveler reactions to the potential managed lanes. Almost 200 people participated in different groups. A majority of single-occupant vehicle operators and transit participants like the idea of managed lanes, while HOV users were less positive. However, 76 percent of both SOV and HOV participants acknowledged they would use the managed lanes at least occasionally, with at least 20 percent saying they would use them all the time. Only 11 percent of the focus group participants said they expected to never use the managed lanes because they did not want to pay tolls.

Preferred Alternatives

Each of the six alternatives was subjected to preliminary analysis under two alternative operating assumptions:

  • A definition of carpools equal to two or more occupants; and
  • A definition of carpools equal to three or more occupants.

This carpool definition was found to significantly influence traffic and revenue potential.

Based on the preliminary analysis, it was clear that some of the options would not be operationally viable unless the definition of carpool was changed from two to three. Vehicles with two or more occupants simply make up a significantly high proportion of total traffic that essentially fill the managed lanes under certain scenarios leaving little or no room to sell to non-HOV traffic.

The preliminary analysis also showed that the directional imbalance over most sections of I-95 was not sufficient to support full reversible managed lane operations. Scenarios which would simply convert two single directional HOV lanes to reversible operations were found to negatively impact operations on I-95 by taking away an HOV lane in the minor travel direction while demand necessitated this additional capacity. The study also found that directionality is opposite at the northern end of the corridor, in Broward County, and the southern portion of the corridor, south of the Golden Glades Interchange. In the morning peak hour, for example, heavy traffic is northbound in the Broward County portion of the corridor and southbound in most of the Miami-Dade County portion of the corridor.

Based on the results of the alternatives analysis, FDOT selected two alternatives as most promising. These included Alternatives 5 and 6. As shown in Figure ES-2, Alternative 5 would involve the simple conversion of existing HOV lanes to HOT operation north of the Golden Glades Interchange. South of the interchange, a three-lane elevated roadway would be constructed. The elevated roadway would operate as express toll lanes and all traffic using these lanes would be required to pay a toll. Existing HOV lanes at current roadway level would be retained for carpool traffic. A moveable barrier would be provided on the elevated roadway to permit two lanes to be operated in the major travel direction and one lane in the minor direction.

figure ES-2 conceptual project configuration alternative 5
Text Box: Figure ES-2  Conceptual Project Configuration – Alternative 5

As shown in Figure ES-3, Alternative 6 would be similar to Alternative 5 except the elevated roadway would have a full four-lane cross section. Two lanes would be operated in each direction at all times and a fixed median barrier would be provided. Once again, the existing HOV lanes, and all four general purpose lanes in each direction would be retained south of the Golden Glades Interchange.

Under either alternative, tolls would be collected using SUNPASS electronic toll collection only. Electronic toll gantries would be constructed at several locations along the project corridor. In the single-lane sections of the project, separate lanes would be provided in electronic toll zones to distinguish between toll-free carpool and toll paying non-carpool traffic.

Variable tolls would be used, with highest toll rates assessed during peak travel periods. The project was analyzed assuming tolls would vary by distance traveled, with a per mile rate assigned for each hour of the day for each travel direction.

In 2010, the assumed opening year of the project, morning peak-hour toll rates were estimated at $0.40 per mile in the southbound direction under both project alternatives and $0.45 and $0.40 per mile in the northbound direction under Alternatives 5 and 6, respectively. Rates as low as $0.15 per mile were used in midday and other off-peak hours. Optimum toll rates were based on those which would produce maximum revenue potential while still ensuring free-flow conditions in the managed lanes. As congestion would grow in future years, optimum toll rates would also increase in future analysis years, such as 2020 and 2030.

figure ES-3 conceptual project configuration alternative 6 Text Box: Figure ES-3  Conceptual Project Configuration – Alternative 6

 

Estimated Traffic and Revenue

The managed lanes were estimated to serve as much as 38,000 vehicles per typical weekday along the elevated roadway section, generally in the vicinity of 103rd Street, in 2010 under Alternative 5. This would increase to as much as 49,000 vehicles per weekday by 2020 and beyond.

Weekday traffic estimates in the single-lane northern portion of the project reached as much as 22,000 toll paying vehicles per day, in addition to toll-free vehicles with three or more occupants under Alternative 5. Slightly higher volumes were found under Alternative 6, although more capacity would be provided under this option. In general, traffic on the elevated section of the project was constrained based on relatively high toll rates which were needed to manage demand on the single lane portions of the project.

Table ES-1 shows estimated net revenues for Alternative 5. Annual toll revenues have been adjusted to reflect ramp-up over the first three years of operation. They have been further adjusted to reflect inflation, nominally assumed at 2.5 percent per year from 2005 on.

Annual toll revenue under Alternative 5 would be expected to increase from $17.7 million in 2010 to $77.6 million in 2020 and $130.3 million in 2030. Operations and maintenance costs associated with toll collection only would be expected to increase from just under $5 million in the opening year to just under $10 million by 2030. This would result in annual net revenue ranging from $12.8 million in 2010 to $120.9 million by 2030.

Table ES-2 shows estimated annual net revenue for Alternative 6. This scenario is found to have slightly higher toll revenue potential and slightly lower operating costs. The lower operating cost was due to the elimination of the need for management of reversible operations of the elevated roadway section.

Annual toll revenue under Alternative 6 would increase from just under $20 million in 2010 to more than $137 million by 2030. Net revenue would increase from about $15 million in 2010 to $129 million by 2030.

Table ES-1
Estimated Annual Net Revenue
Alternative 5

Year
Annual Revenue in Future Year
Total Operating & Maintenance Costs (1)
Annual Net Revenue
2010(2)
2011(2)
2012(2)
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
$17,738,000
26,078,000
33,436,000
39,029,000
43,052,000
47,490,000
52,385,000
57,785,000
63,741,000
70,312,000
77,559,000
81,691,000
86,043,000
90,626,000
95,454,000
100,539,000
105,895,000
111,536,000
117,477,000
123,735,000
130,327,000
$4,929,000
5,289,000
5,597,000
5,835,000
6,016,000
6,204,000
6,399,000
6,600,000
6,809,000
7,025,000
7,250,000
7,446,000
7,647,000
7,853,000
8,065,000
8,282,000
8,506,000
8,736,000
8,973,000
9,216,000
9,465,000
$12,809,000
20,789,000
27,839,000
33,194,000
37,036,000
41,286,000
45,986,000
51,185,000
56,932,000
63,287,000
70,309,000
74,245,000
78,396,000
82,773,000
87,389,000
92,257,000
97,389,000
102,800,000
108,504,000
114,519,000
120,862,000

(1)
Operating and Maintenance Costs are in
(2)
Annual transactions and revenue estimates have been adjusted to reflect ramp-up during the first three years of operation.

 

 

Table ES-2
Estimated Annual Net Revenue
Alternative 6

Year
Annual Revenue in Future Year
Total Operating & Maintenance Costs (1)
Annual Net Revenue
2010(2)
2011(2)
2012(2)
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
$19,280,000
28,385,000
36,446,000
42,603,000
47,061,000
51,986,000
57,426,000
63,436,000
70,074,000
77,407,000
85,508,000
89,665,000
94,025,000
98,596,000
103,390,000
108,417,000
113,688,000
119,216,000
125,012,000
131,091,000
137,464,000
$4,285,000
4,650,000
4,964,000
5,203,000
5,383,000
5,571,000
5,766,000
5,970,000
6,183,000
6,406,000
6,638,000
6,810,000
6,987,000
7,169,000
7,355,000
7,546,000
7,742,000
7,944,000
8,150,000
8,362,000
8,580,000
$14,995,000
23,735,000
31,482,000
37,400,000
41,678,000
46,415,000
51,660,000
57,466,000
63,891,000
71,001,000
78,870,000
82,855,000
87,038,000
91,427,000
96,035,000
100,871,000
105,946,000
111,272,000
116,862,000
122,729,000
128,884,000

(1)
Operating and Maintenance Costs are in
(2)
Annual transactions and revenue estimates have been adjusted to reflect ramp-up during the first three years of operation.

Summary

In conclusion, this comprehensive traffic and revenue study showed that managed lanes can have a significant positive impact on meeting future traffic growth and reducing congestion levels on this heavily used segment of urban interstate. Among other conclusions, the study showed:

  • Single-lane HOT options or sections are less viable than multi-lane options; the single-lane sections on the northern end of the project often dictate high toll rates to manage demand;
  • Traffic and revenue potential on the managed lanes would be enhanced with the change in the definition of carpools from two to three occupants;
  • There appears to be insufficient directionality of demand during peak hours to warrant full reversible express roadways;
  • The managed lanes would provide significant time savings of as much as 38 minutes for a through trip between I-595 and downtown Miami;
  • The managed lanes would improve operating speeds in the general purpose lanes, particularly in the later years such as 2030;
  • The most viable portions of the project would appear to be between the Golden Glades Interchange and downtown Miami – FDOT may wish to give consideration to constructing only a partial alternative since the two most promising options involve construction of an elevated roadway south of this location and retention of only single-lane HOT lanes north;
  • The construction of managed lanes on I-95 would provide enhanced opportunities for express bus service in the corridor, which appeared to engender significant support during focus group discussions;
  • Traffic and revenue needs on the lanes, as with most managed lanes projects, are extremely sensitive to small changes in basic assumptions, overall demand in the corridor, values of time and other parameters; and
  • The preferred project alternatives would generate substantial net revenues which would be able to support a significant portion of project cost.

Back to Top
chapter 1 header

I-95 is a critical north-south freeway serving Miami-Dade, Broward and other counties along the east coast of Florida. The route extends from a southern terminus at U.S. Route 1 in Miami north as far as the Canadian border, and is one of the most heavily traveled corridors in America. Within South Florida, it provides an important backbone for commuter and recreational traffic, with traffic volumes exceeding 200,000 vehicles per day in many locations.
 
Over the years, traffic growth on I-95 has been strong, and congestion has continued to grow. This is in spite of major improvement programs along I-95, including the addition of one High Occupancy Vehicle (HOV) lane along the entire length between Miami and Fort Lauderdale and points north.

Nonetheless, congestion has continued to worsen, consistent with trends in major urban areas throughout the U.S. It is increasingly difficult to widen roadways such as I-95 due to right-of-way constraints, environmental and other factors.
 
Managed lanes, such as HOT lanes or express toll lanes, are rapidly emerging as a potential solution to congestion in several urban areas around the country. The first project of this nature, the S.R. 91 Express Lanes in Orange County, California, has been an overwhelming success, providing significant congestion relief and an opportunity for motorists to get out of congestion by paying a variable toll. HOT lane projects have also been successful in San Diego, California; Minneapolis, Minnesota and Houston, Texas with other projects recently opening in Denver and elsewhere.

With minimal opportunity for future widening of I-95 in Miami-Dade and Broward Counties, the Florida Department of Transportation (FDOT) submitted an application to the FHWA Value Pricing Pilot Program for a comprehensive traffic and revenue study of possible managed lanes along portions of I-95. That study was approved and a team headed by Wilbur Smith Associates (WSA) was selected to undertake the study in mid-2004.
 
This report summarizes the results of that study, and the numerous aspects of the analysis undertaken. It incorporates the results of several interim reports and technical memoranda which have been submitted throughout the course of the study.

Study Objective

The objective of the study was to make a comprehensive examination of traffic and revenue potential of proposed managed lanes on I-95, generally between Fort Lauderdale and Miami, under a variety of physical and operational alternatives. The study was performed at a level of detail consistent with that expected for a study intended for use in project financing. However, in this case, more detailed PD&E studies will be needed in advance, hence ultimate project financing is probably some time away.

The study included an extensive program of market research, including focus groups, stated preference surveys and a limited travel pattern survey. A detailed operations profile was developed for the I-95 corridor since motorists’ willingness to pay tolls to use managed lanes is dependent on operating conditions in the adjacent toll-free general purpose lanes. Six project alternatives were analyzed, using a corridor “micro-model,” and operations simulation model. From the six primary alternatives, two preferred options were selected and refined estimates of traffic and revenue potential were developed.

Project corridor

As shown in Figure 1-1, the proposed managed lanes were evaluated in a corridor on I-95 extending from just south of I-595 in Fort Lauderdale to just south of the I-95 junction with S.R. 836 and I-395, near downtown Miami. This covers a total length of 18.4 miles, about 12 miles of which would be located in Miami-Dade County and the remainder in Broward County.

Along this stretch, I-95 carries well over 200,000 vehicles per day, with weekday traffic approaching 300,000 on several sections. Heaviest traffic is generally immediately south of the Golden Glades Interchange; the junction of Florida’s Turnpike, I-95 and the Palmetto Expressway (S.R. 826). Immediately north of the Golden Glades Interchange traffic is somewhat lower on I-95 but increases again to more than 275,000 vehicles per day south of I-595.

figure 1-1 location map
Text Box: Figure 1-1  Location Map

For the most part, as shown in Figure 1-2, this section of I-95 includes a total of ten travel lanes, five in each direction. Four of these lanes in each direction are full-time general purpose lanes, while the leftmost lane in each direction is reserved for high-occupant vehicles, currently defined as vehicles with two or more occupants, during peak periods. An HOV treatment is used, which permits continuous access and egress to the HOV lanes and full general purpose use during off-peak hours and on weekends.

figure 1-2 typical existing I-95 configuration
Text Box: Figure 1-2  Typical Existing I-95 Configuration

In the Broward County portion of the corridor, directional demands are relatively balanced during the morning and afternoon peak period and may slightly favor the northbound direction in the A.M. peak close to Fort Lauderdale. In the southern portions of the project, in Miami-Dade County, there is a slightly greater directional split, favoring the southbound direction in the morning peak and the northbound in the afternoon peak. However, there are very high levels of demand in the off-peak direction as well, in both counties.

The existing HOV lanes operate continuously over the entire length of the study corridor. An HOV bypass elevated section has been constructed through the Golden Glades Interchange. Currently, the HOV lanes end at S.R. 112 on the south, with a direct connection to and from the west on S.R. 112. There are no HOV lanes in the short section between S.R. 112 and S.R. 836 on I-95.

Chapter 2 of this report presents a detailed description of current traffic levels and the traffic operational profile on I-95. While traffic is especially heavy during peak periods, high levels of traffic exist almost throughout the daylight hours, making the corridor a potentially good candidate for managed lanes or other form of tolled express lane facilities.

 

Scope of Work

The comprehensive study included an extensive program of market research and outreach. This included computer based stated preference surveys, which were intended to measure motorists’ willingness to pay tolls and general propensity to use the proposed managed lanes. A series of focus groups were undertaken, targeted to represent different segments of the travel population in the I-95 corridor, to assess public opinion and general acceptability of the concept. A limited origin-destination survey was undertaken, using digital imaging of vehicle license plates, and a mailback survey technique. The survey resulted in a relatively small sample of observed travel patterns on the corridor; this was supplemented by a detailed select link analysis of the trip patterns in the regional travel demand model to get a clear understanding of travel patterns and characteristics.

A detailed traffic operations profile was compiled over the entire length of the study corridor. Two permanent automatic traffic recorder stations were located in the corridor, and these provided a full array of daily, seasonal, and hourly traffic variations by travel direction. In addition, hourly ramp counts were obtained from FDOT at all interchanges, and complete peak versus off-peak operating profiles were established over the entire length of the corridor.

Speed and delay runs were operated, using GPS technology, along I-95 at various times of day to measure current patterns of delay, congestion points and to provide a basis for calibration of the detailed traffic simulation models used in the study.

Future-year global travel demand estimates along the I-95 corridor were developed using the Tri-County Southeastern Regional Planning Model (SERPM). Included with this model was a set of socioeconomic growth forecasts for Miami Dade, Broward County and the overall modeling area. Given the comprehensive nature of this study, WSA retained an independent economic subconsultant, Washington Economics Group, Inc. (WEG) to conduct an independent review of the underlying socioeconomic forecast. Some minor revisions were made in the socioeconomic estimates and new trip tables developed for use in the analysis.

A three-tiered modeling approach was used. The SERPM model was used to estimate total demand in the corridor; that is the total amount of traffic demand on I-95 itself. A market share “micro-model” was developed, using a tight “window” of the SERPM model, to estimate the share of traffic between the toll-free general purpose and the tolled managed lanes for each project configuration. A third model was a VISSIM micro-simulation program to estimate changes in travel speeds and travel times under varying shares of traffic between the toll-free and tolled managed lanes.

Six project alternatives were evaluated ranging from simple conversion of the existing HOV lanes to HOT operation, to a possible multi-lane elevated roadway to be constructed south of the Golden Glades Interchange. A more detailed description of each option is provided later in this chapter.
 
For each of the six physical alternatives, two operational alternatives were considered, one assuming the definition of carpools remains at two or more occupants and a second scenario assuming the regional definition of carpools is increased to vehicles with three or more occupants. Because existing HOV lanes are growing increasingly full, it may be necessary to redefine HOV traffic in the future to ensure free-flow conditions will continue in HOV lanes, regardless of what changes may be introduced on I-95.

After the initial traffic and revenue analysis for all six options, two preferred alternatives were identified. Refined traffic and revenue estimates were developed for these alternatives and are presented later in this full report. A limited number of sensitivity tests were performed, such as reduced growth or alternative values of time, to estimate the potential impact of changes in certain basic study assumptions. Finally, the work was documented in a series of interim submittals, including WSA and subconsultant reports and technical memoranda, and finally compiled into this report.

Description of Alternatives

As noted above, six alternative configurations were fully evaluated in the study. These are briefly described below.

Alternative 1

As shown in Figure 1-3, Alternative 1 would involve simply converting the existing single HOV lanes in each direction to HOT operation from I-595 on the north to S.R. 112 on the south. The lanes would be operated in their current locations, but for purposes of this analysis, it was assumed that they would be converted to HOT operations 24 hours per day. In addition, there would be limited points of access and egress, unlike the open access system in use with the HOV lanes today. Specific access point assumptions for each alternative are described subsequently.

South of S.R. 112, new HOV lanes would be constructed through the I-195 and I-395 Interchanges. This would likely take the form of elevated HOT lanes over this relatively short distance. The new section of elevated lanes would be operated in the same way as the remaining portions of the HOT lanes.

As noted above, this alternative was considered under two operating scenarios, one assuming HOV was defined as two or more persons, while the second option defined as HOV with having three or more occupants. This has a significant impact on operational viability, since there are considerably less three-occupant vehicles in the traffic stream than two-occupant vehicles. In fact, the study found that if the definition of HOV is not increased above two occupants, there will be little or no capacity available in the HOV lane to sell to non-HOV traffic; hence, this alternative may not be practical under the HOV-2 scenario.

 figure 1-3 conceptual project configuration alternative 1  
Figure 1-3
Conceptual Project Configuration – Alternative 1

Alternative 2

As shown in Figure 1-4, Alternative 2, north of the Golden Glades Interchange, would be identical to Alternative 1. In fact, for the most part, all six scenarios were assumed to be the same north of the Golden Glades Interchange; that is two single HOT lanes, one in each direction, with controlled points of access. South of the Golden Glades Interchange, the two HOT lanes would be converted to reversible operation, with two lanes operating southbound in the morning peak and the two lanes operating northbound in the afternoon peak. Under this scenario, there would be no HOV or HOT lane provided in the minor travel direction during peak hours or in either direction during off-peak hours.

Similar to Alternative 1, the HOT lanes would be extended south of S.R. 112 and into the northern periphery of downtown Miami, presumably on an elevated section. This section, too, would include two reversible lanes operated northbound or southbound during the different peak periods.

 figure 1-4 conceptual project configuration alternative 2 
Figure 1-4
Conceptual Project Configuration – Alternative 2

Alternative 3

Figure 1-5 shows the proposed configuration for Alternative 3. Again, north of the Golden Glades Interchange, only a single HOT lane in each direction would be operated. South of the Golden Glades Interchange, a third lane would be added in the median, together with a moveable barrier wall. This would permit two HOT lanes operating in the major direction and one major HOT lane operating in the minor direction. Figure 1-5 shows a typical morning peak condition, with two lanes operating southbound and one operating northbound. This would be reversed in the afternoon peak.

South of S.R. 112, an elevated extension of the lanes was assumed, also to be three lanes, the center lane of which would be reversible. This alternative has the advantage of providing additional HOT lane capacity in the major direction, while not taking away any capacity in the minor direction or during off-peak hours.

 figure 1-5 conceptual project configuration alternative 3  
Figure 1-5
Conceptual Project Configuration – Alternative 3

Alternative 4

As shown in Figure 1-6, Alternative 4 would feature construction of a new elevated roadway south of the Golden Glades Interchange. In this case, the new lanes would be in addition to the existing two HOV lanes on the lower roadway. Hence, the existing 10-lane cross section would be retained, with the leftmost lanes reserved for HOV operation, at least during peak periods. The elevated lanes, which are assumed to be reversible in Alternative 4, would be tolled express lanes. All vehicles using these lanes would be required to pay a toll, regardless of vehicle occupancy. However, carpools could still continue to use toll-free the existing HOV lanes south of the Golden Glades Interchange.

North of the Golden Glades Interchange, Alternative 4 would operate the same as all alternatives; a single HOT lane in each direction as far north as I-595. The elevated roadway would extend into downtown Miami, with fixed points of access between the express lane roadway and the general purpose lanes. For purposes of analysis, it was assumed that the toll-free HOV lanes would also be extended south of S.R. 112, into the northern edge of downtown Miami.

 figure 1-6 conceptual project configuration alternative 4  
Figure 1-6
Conceptual Project Configuration – Alternative 4

Alternative 5

Figure 1-7 shows the hypothetical proposed configuration for Alternative 5. This is the same as Alternative 4, except that the elevated roadway would be constructed with three lanes, two operating in the major direction and one operating in the minor direction. Again, all traffic using the elevated roadway would be required to pay a toll, regardless of occupancy. However, the existing toll-free HOV lanes located on the existing general roadway would remain available for carpool traffic.

 figure 1-7 conceptual project configuration alternative 5   
Figure 1-7
Conceptual Project Configuration – Alternative 5

Alternative 6

As shown in Figure 1-8, Alternative 6 would be identical to Alternative 5, except that the elevated roadway would be constructed with a total of four lanes, two in each direction. A fixed barrier would be provided on the elevated roadway, together with fixed points of transition between the tolled express lanes and the general purpose lanes at various locations between the Golden Glades Interchange and downtown Miami. Where the express lanes were constructed, the HOV lanes would also remain available, toll free for carpools and buses. North of the Golden Glades Interchange, the same two HOT lanes would be operated, as with all project alternatives.

figure 1-8 conceptual project configuration alternative 6  
Figure 1-8
Conceptual Project Configuration – Alternative 6

Access Assumptions

Figure 1-9 shows the proposed HOT/express lanes access assumptions for each project alternative. As noted above, the northern portion of the project would be identical under all six of the operational alternatives. In this section, single HOV lanes would be converted to HOT operation, with one lane operating in each direction. Fixed points of access and egress would be constructed immediately south of I-595, between Sheraton Street and Hollywood Boulevard and between Ives Dairy Road and Miami Gardens Drive. Fixed access and egress slip ramps, probably delineated simply with striping, would be provided at these locations.

The lower portion of Figure 1-9 shows access assumptions for the southern part of the proposed project. Different access assumptions were used for different project alternatives, as shown.

Alternative 1 would involve conversion of the existing HOV lanes to HOT operation, with a limited number of fixed access points. Also note, shown in green, is the extension of the HOT lanes south of S.R. 112 to a point just to and slightly beyond the I-395/S.R. 836 Interchange. Intermediate access points would be constructed to and from the north at NW 135th Street and NW 95th Street. Access to and from the south would be implemented at NW 103rd Street and NW 69th Street. Note that there would be no immediate access at Florida’s Turnpike, due to potential weaving problems in the immediate vicinity of that interchange.

 figure 1-9 proposed HOT/express lanes access assumptions
Figure 1-9
Proposed HOT/Express Lanes Access Assumptions by Project Alternative

Alternative 2 would involve reconstruction of the HOV lanes south of Florida’s Turnpike to reversible HOT operation. This scenario would also include the southerly extension to I-395 and beyond. In this case, it would be necessary to add intermediate access points in the vicinity of Florida’s Turnpike, since the lanes north of Florida’s Turnpike would not be reversible. This additional access would provide access to and from the reversible lanes and bi-directional lanes.
 
Alternative 3 would include similar access arrangements, although no access to and from the north just south of Florida’s Turnpike. This is because under this scenario, the HOT lanes south of the Golden Glades Interchange will always be operated in both directions, although with one additional lane in the major travel direction.

Finally, access for Alternatives 4, 5, and 6 are shown at the bottom of Figure 1-9. In this case, south of the Golden Glades Interchange, the existing HOV lanes would remain in use, and actually be extended to I-395. These toll-free lanes, shown in purple in the figure, would continue to be available, in addition to the elevated toll lanes. The elevated lane configuration is shown in green, including direct access points to and from Florida’s Turnpike, as well as the vicinity of 119th /125th Street and 79th/82nd Street. The middle access points would be designed to provide transition between the main lanes and the elevated express lanes as well as a local ramp option direct on to existing interchange ramps. Full direct access would also be provided, in both directions, at I-195/S.R. 112 and I-395/S.R. 836, to and from the elevated express toll lanes.

Order of Presentation

Chapter 2 provides a description of the existing operating traffic profile along I-95 within the project limits. Chapter 3 summarizes the corridor growth potential and the results of the independent economic review. A summary of the market research program, included the stated preference surveys and focus group results, is provided in Chapter 4. Preliminary traffic and revenue estimates for the various project alternatives are summarized in Chapter 5 while more detailed information for the preferred options are presented in Chapter 6, together with sensitivity test results.

Back to Top
chapter 2 header

The evaluation of traffic and revenue potential for managed lanes such as those proposed for I-95 requires the development of a detailed traffic and operations profile. Motorists’ willingness to pay a toll to use managed lanes, or HOT lanes, is, of course, dependent on levels of congestion in the competing toll-free general purpose lanes. Hence, it is important to consider not only daily traffic levels but also hourly traffic distributions, directional splits and vehicle composition.

This chapter presents a summary of the detailed traffic and operations profile developed for I-95 for use in this study. It includes an analysis of daily and hourly traffic variations, vehicle occupancy rates, a general assessment of travel patterns and characteristics and historical traffic trends. In addition, extensive speed-delay runs were made during peak, off-peak and shoulder hour conditions along I-95; a summary of this is also included.

figure 2-1 2004 average daily traffic
Figure 2-1
2004 Average Daily Traffic

Average Daily Traffic

Average daily traffic for I-95 and other major freeway facilities in Miami-Dade and Broward Counties were obtained from the FDOT traffic data base. Figure 2-1 displays typical average daily traffic volumes from the year 2004 at selected locations throughout the region.
 
On I-95, average daily traffic volumes range from just over 150,000 south of S.R. 836 near downtown Miami, to over 283,000 just south of the Golden Glades Interchange junction with Florida’s Turnpike. These reflect average daily volumes; weekday traffic levels are somewhat higher.
 
In general, the highest traffic levels were found between S.R. 112 and Florida’s Turnpike and just south of I-595 in Broward County. Traffic on I-95 drops below 200,000 in extreme northern Miami-Dade County; just north of the split with Florida’s Turnpike.

Figure 2-1 shows daily traffic volumes on other freeway facilities as well. At the north end of the project corridor, I-595 carries as much as 185,000 vehicles per day while Florida’s Turnpike carries between 62,100 and 105,400 vehicles per day. The Palmetto Expressway, S.R. 826, carries between 135,000 and 216,000, depending on location. S.R. 836 and S.R. 112, both of which are tolled facilities operated by the Miami-Dade Expressway Authority, carry heavy volumes near or well above 100,000 vehicles per day; both facilities intersect with I-95 near the south end of the corridor.

Both the I-195 and I-395 causeway corridors to Miami Beach carry roughly 100,000 vehicles per day. In short, freeway traffic levels in Southeast Florida are exceptionally high; and the I-95 corridor is highest among these.

I-95 Weekday Traffic Profile

Traffic counts were obtained from FDOT at selected locations along I-95 and on all interchange ramps within the project limits. In a few cases, FDOT counts were supplemented by machine counts performed by the WSA Team. This enabled the development of a balanced weekday traffic profile from the south end of the corridor to the north end of the corridor along I-95.
 
The weekday traffic profile, at 2004 levels, is shown in Figure 2-2. Two control count locations were used to anchor the traffic profiles;

  • A continuous count station south of Pembroke Road in Broward County; and
  • A portable count location south of NW 151st Street in Miami Dade.

Beginning at the north end of the project, weekday traffic levels reached 285,000 just south of Griffin Road. This volume decreases to just over 246,000 at the Miami-Dade County line. Weekday traffic further decreases to about 193,000 just north of the Golden Glades Interchange, but then increases again to 290,000 just south of 151st Street.
 
The heaviest traffic volumes were found between NW 95th Street and S.R. 924 (119th Street), at approximately 302,000 vehicles per weekday. While volumes remain over 275,000 as far south as S.R. 112, dropping to 242,000 north of S.R. 836 and to about 191,000 south of S.R. 836.
 
Weekday ramp volumes are also shown in Figure 2-2. For purposes of this presentation, directional ramp volumes were summed and then averaged by northbound and southbound directions, to achieve balanced flows.

 figure 2-2 2004 average weekday traffic 
Figure 2-2
2004 Average Weekday Traffic

Table 2-1
Annual Traffic Trends
I-95 South of Pembroke Road

Year
Average Daily Traffic
Percent Change
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
205,600
212,800
214,100
214,900
213,700
228,500
232,600
233,200
239,400
243,300
252,000
---
3.5%
0.6%
0.4%
-0.6%
6.9%
1.8%
0.3%
2.7%
1.6%
3.6%

Average Annual Percent Change:
1995-2000 - 2.1%
2000-2005 - 2.0%
1995-2005 - 2.1%

Historical Traffic Trends

Historical traffic trends are shown in Table 2-1 for a location south of Pembroke Road. FDOT operates a continuous count station at this location, which shows that average daily traffic increased from 205,600 in 1995 to 252,000 in 2005, an average annual increase over the 10-year period of 2.1 percent. Annual traffic growth has been relatively stable over the 10 years, with year-to-year fluctuation but five-year average growths virtually the same between 1995 and 2005.

The nominal 2 percent average annual rate of traffic growth in Broward County is considered typical of the overall corridor. This rate of growth is lower than some other facilities in the region, likely due to the fact that traffic volumes are already exceptionally high and congestion routinely occurs, especially south of the Golden Glades Interchange.

Hourly Traffic Variations

Hourly traffic variations are shown in Figure 2-3 at two locations along I-95; south of Pembroke Road and south of 151st Street. Hourly traffic is shown for a typical weekday condition, separately by northbound and southbound directions.
 
At the northern location (south of Pembroke Road) in Broward County, peak hour traffic in both directions is relatively balanced, with slightly heavier southbound volumes in some hours, but only by a small margin. Peak hour volumes reach about 8,600, per direction, close to the capacity of I-95 at that location, which includes four general purpose and one HOV lane in each direction.
 
It is interesting to note, however, that hourly demands decrease only a small amount during shoulder and even mid-day hours. In fact, in the southbound direction, all hours between 7:00 A.M. and 6:00 P.M. were found to be in the range of 7,500 to 8,600 vehicles per hour.
 
At the Miami-Dade County selected location, south of 151st Street, slightly more directional peaking is shown, with slightly heavier volumes southbound in the morning peak and northbound in the afternoon peak. Total hourly traffic flows are also slightly higher, with peak direction/peak hour volumes reaching as much as 9,600 vehicles per hour, at or above the nominal capacity of this section of road.
 
Once again, traffic in the southbound direction shows relatively small variations between hours of the day. Northbound traffic does show a late morning dip prior to a steady increase in traffic during the afternoon hours.

Perhaps most important to note is the lack of pronounced directional splits at either of these locations. Some of the options studied for the managed lanes included use of reversible or variable lanes, northbound versus southbound. The relatively small directional imbalances may tend to make these options less effective.

 figure 2-3 I-95 hourly traffic distributions  
Figure 2-3
I-95 Hourly Traffic Distribution

Peak Period Profile

Figures 2-4 and 2-5 show A.M. and P.M. peak period traffic profiles along all mainline sections and ramps of I-95 within the project limits. Hourly volumes are shown by direction at all locations. Figure 2-4 shows A.M. peak period (6:00-9:00 A.M. - 3 hours) traffic volumes along I-95. As noted above, directional imbalances are relatively small, but would favor the southbound direction in the A.M. peak period south of the Golden Glades Interchange. Interestingly, within Broward County the major direction is northbound in the A.M. peak period, although relatively small directional imbalances are shown.

This is another important finding of the study, inasmuch as the north and south ends of the proposed managed lanes would have opposite directional peaking characteristics. This further complicates potential pricing strategies needed to manage demand along the full length of the facility and add further question regarding the viability of reversible lanes. It is noted that none of the six alternatives studied included reversible lanes in the Broward County portion of the project.

Figure 2-5 provides the P.M. traffic profile, again at 2004 levels. This covers the three-hour P.M. peak period from 3:00 – 6:00 P.M. In this case, directional flows are slightly higher in the northbound direction, south of the Golden Glades Interchange, and relatively balanced in the Broward County section. Three-hour volumes currently reach 27,000 – 29,000 in some locations, reflecting an average of 9,000 – 10,000 vehicles per hour per direction. These are exceptionally high traffic levels, and are indicative of the levels of congestion which were found in extensive speed studies performed as part of the analysis.

figure 2-4 2004 average weekday traffic a.m. peak
Figure 2-4
2004 Average Weekday Traffic = A.M. Peak Period (6:00-9:00 A.M.)

figure 2-5 2004 average weekday traffic a.m. peak
Figure 2-5
2004 Average Weekday Traffic = P.M. Peak Period (3:00-6:00 P.M.)

Figure 2-6 presents the profile information on a single hour basis for A.M. peak, mid-day and P.M. peak conditions. Northbound and southbound traffic volumes are shown in separate colors. This is a convenient way of showing the distribution of peak traffic loadings over the length of the facility, while also pointing out directional splits wherever these may exist.
 
In the A.M. peak hour, southbound volumes south of the Golden Glades Interchange are shown to be between 9,000 and 10,000 per hour at most locations. Northbound volumes in the A.M. peak are considerably lower. The nominal capacity of the general purpose lanes in this area would be in the range of 8,000 vehicles per hour, recognizing the presence of trucks in the traffic stream. Traffic volumes above 8,000 need to be accommodated in the HOV lane; it is clearly shown that for many sections, the HOV lanes and general purpose lanes would likely be at capacity, south of the Golden Glades Interchange.

 figure 2-6 hourly demand profile by direction  
Figure 2-6
Hourly Demand Profile by Direction

Immediately north of the Golden Glades Interchange, traffic levels drop off considerably, to as low as 5,000 vehicles per hour per direction. This is due to the significant amount of traffic which leaves I-95 for Florida’s Turnpike and Palmetto Expressway. Traffic levels build steadily moving north, with the increased commuting demand to and from Fort Lauderdale.

Similar patterns are shown in the P.M. peak hour, although slightly lower peaking levels are experienced. Also, the directional imbalance south of the Golden Glades Interchange is not as pronounced as in the A.M. peak hour. Traffic is shown to be quite balanced directionally in the Broward County section.

Hourly traffic volumes in the mid-day hours are also shown to be quite high, particularly south of the Golden Glades Interchange. Even in mid-day hours, total southbound volumes, for example, exceed 8,000 vehicles per hour on several sections. The leftmost lane is not restricted to HOV during these hours; hence the total roadway capacity is close to 10,000 per hour per direction. Nonetheless, these relatively high volumes even during mid-day hours suggest that with future traffic growth, there may well be a need for managing demand even in off-peak hours.

Vehicle Class Distributions

FDOT maintains vehicle classification counts at its continuous count stations. One of these stations is located on I-95 within the project study area, just south of Pembroke Road in Broward County. Table 2-2 shows the percent distribution between passenger cars and commercial vehicles at three points over the last 10 years.

Passenger cars, as a proportion of total vehicles, have remained relatively constant over the last 10 years, increasing from 93.4 percent in 1995 to 93.8 percent in 2003. The total commercial vehicles are generally split evenly between single and multi-unit trucks, with multi-unit trucks representing 3.2 percent in 2003 while light trucks represented 3.0 percent. In analyzing the various project alternatives, trucks were assumed to be prohibited from any of the HOT lane sections of the project alternatives, although theoretically all vehicle classes might be eligible to use the elevated lanes in Alternatives 4, 5 and 6 since separate HOV lanes are provided.

Table 2-2
Vehicle Class Distributions

 
Commercial Vehicles
 
Year
Passenger Cars
Single Unit
Multi-Unit
Total
Total Vehicles
1995
1999
2003
Count
Percent
Count 
Percent
Count
Percent
192,091
93.4%
199,813
93.5%
224,537
93.8%
8,004
3.9%
7,480
3.5%
7,189
3.0%
5,336
2.6%
6,411
3.0%
7,668
3.2%
13,545
6.6%
13,891
6.5%
14,857
6.2%
205,636
100.0%
213,704
100.0%
239,394
100.0%

Source: Florida Department of Transportation

Travel Patterns

A limited travel pattern and characteristic survey was undertaken as part of the study. License plate images were obtained at a pedestrian overpass location south of the Golden Glades Interchange and on selected ramp locations along the corridor.

figure 2-7 license plate distribution by zip code

 

 

 

figure 2-8 license plate distribution by zip code

 

 

 

Text Box: Figure 2-7  License Plate Distribution by Zip Code  All I-95 Survey Locations    Text Box: Figure 2-8  License Plate Distribution by Zip Code  Golden Glades Park and Ride Survey Locations

Figure 2-7 shows the proportional distribution of license plate by zip code of vehicle registration. As might be expected, vehicle registrations are most heavily concentrated along the I-95 corridor itself, in eastern Miami Dade and Broward Counties. The heaviest concentrations were found in close proximity to the county line.

License plate surveys were also conducted at a major park-and-ride in the vicinity of the Golden Glades Interchange. Figure 2-8 shows the zip code distributions of license plates from that location; these are heavily concentrated along the eastern portion of the Broward/Miami-Dade County line, just north of the Golden Glades Interchange. That park-and-ride facility provides express bus service to and from downtown Miami; consistent with the distribution pattern of vehicle registrations.

To provide a general assessment of the relative distribution of trip patterns along this section of I-95, WSA performed a select link analysis at the northern and southern ends of the corridor, using the base year trip tables from the SERPM travel demand model. These are graphically shown in Figure 2-9.
 
The left side of the figure displays the distribution of trips to and from the north end of the corridor, at a select link location just south of I-595. Of trips using I-95 in Fort Lauderdale, 67 percent are traveling to or from local interchanges north of the Golden Glades Interchange. About 8 percent of the traffic is oriented to the Golden Glades Interchange itself, mostly the Palmetto Expressway, while another 8 percent is found to be using local interchanges between the Florida Turnpike and S.R. 112. About 17 percent of the trips on I-95 south of I-595 continue through the entire length of the corridor, to or from S.R. 112, S.R. 836 or points south.

The right side of the graphic shows the distribution of travel patterns from a select link location just north of S.R. 112, near the south end of the project. About 40 percent of I-95 traffic in this location comes from local interchanges between S.R. 112 and the Golden Glades Interchange. Almost one-fourth of traffic is using Florida’s Turnpike or the Palmetto Expressway, while about 36 percent is traveling to local interchanges in Broward County, or to I-595 and north. Through trips represent about 18 percent, comparable to the proportion of through trips at the northern select link location.

figure 2-9 daily select link travel patterns Text Box: Figure 2-9  Daily Select Link Travel Patterns

 

Vehicle Occupancy Distribution

Vehicle occupancy levels are extremely critical to this analysis, recognizing that all of the alternatives have at least a portion of the project which will be operated as HOT lanes. As noted previously, each of the alternatives was analyzed under two operational scenarios, one assuming carpools with two or more occupants would gain toll-free access while another limited this to vehicles with three or more occupants.

Figure 2-10 shows the estimated vehicle occupancy distribution along the I-95 corridor for typical weekday conditions. Data is shown separately for northbound and southbound directions and for A.M. and P.M. peak period conditions. This data was obtained from the 2004 I-95 High Occupancy Vehicle Lane Monitoring report, produced for FDOT. The occupancy distributions were obtained at a location near NW 79th Street in Miami-Dade County. In that study, periodic counts were obtained of vehicle occupancies in both the HOV lane and immediately adjacent general purpose lane.

WSA reviewed this information and made adjustments to reflect the total volume of traffic in all four general purpose lanes. This resulted in the development of occupancy distributions across all five lanes at this location, which is shown in Figure 2-10.

figure 2-10 vehicle occupancy distribution
Text Box: Figure 2-10  Vehicle Occupancy Distribution

 

As might be expected, the highest category of occupancy distribution is the single-occupant vehicle, ranging from 80.5 to 84.1 percent of total traffic. Vehicles with two occupants generally represent between 14.7 and 17.9 percent of the total. Vehicles with three or more occupants represented a very small percentage of traffic, as low as 0.9 percent in the northbound A.M. peak hour to 2.7 percent in the southbound direction.

It is important to note that vehicles with two or more occupants represent an extremely high percentage of total HOV traffic. This suggests that if the definition of carpools was increased from two to three occupants, a significant amount of additional capacity would be opened up in the HOV lanes. This is an important consideration in this study analysis.

Trip Characteristics

A limited trip characteristics survey was undertaken, using a mailback approach from vehicles observed in the I-95 corridor. Figure 2-11 presents a summary of the trip purpose and trip frequency distributions which were found. These were from survey respondents, all of which were driving on I-95 during the period of the survey.

Trips to and from work was the primary peak period trip purpose, with about 74 percent of respondents indicating this. Other trip purposes were relatively evenly distributed during the peak period. About one-third of off-peak users were also commuters, although another one-third were conducting company or personal business.

In terms of trip frequency, two-thirds of peak period travelers made the trip five or more times per week. Only 15 percent were infrequent travelers. As might be expected, during off-peak hours, high frequency trips declined to 28 percent of the total, while low frequency trips (one or less per week) increased to 39 percent of the total.

figure 2-11 trip characteristics distribution Text Box: Figure 2-11  Trip Characteristics Distribution

 

Travel speed profiles

Congestion levels are extremely important when considering managed lanes projects. An extensive program of speed-delay studies were undertaken, using GPS measuring devices, in A.M. peak, P.M. peak and off-peak conditions. A total of 55 speed-delay runs were operated in Miami-Dade and Broward Counties, in either the northbound or southbound direction, on I-95. Six to eight runs were made in each major direction in A.M. and P.M. peak periods in Miami-Dade County; approximately five runs were made in each peak period in Broward County.

For each speed and delay run within Miami-Dade County, vehicles entered I-95 at the Ives Dairy Road Interchange and continued to a point south of I-395 in downtown Miami, or in the reverse direction. Speed and delay runs in Broward County were also conducted between Ives Dairy Road and I-595. The typical Miami-Dade County run covered about 13.5 miles while the typical Broward County portion covered about 8.3 miles.

GPS technology was used to track actual travel speeds, by distance traveled, over the length of each corridor. This is useful in showing bottleneck locations and specific areas where speeds drop below posted speed limits.

Miami-Dade Speed Profiles

Figure 2-12 shows southbound A.M. peak period speed profiles for six speed-delay runs in Miami-Dade County. For each run, the observed operating speed is tracked, by mile post and interchange location. All of these speed runs cover a distance of about 13.5 miles. The total run time and overall average operating speed is shown for each of the six discrete runs. The earliest run began at 6:02 A.M., and was found to have an overall average operating speed of 60.4 MPH. In general this reflects LOS D travel conditions, and only for a short distance did speeds fall below 50 MPH.

However, by the run beginning at 6:29 A.M., a significant degradation of speed is shown, particularly south of the Golden Glades Interchange. Here speeds are shown to drop to as low as 10 MPH or less, with an overall average speed of 38.6 MPH. By 7:13 A.M., average speeds dropped to 36 MPH, with the vast majority of delays experienced south of the Golden Glades Interchange. Speeds continue to decrease, dropping to a low of 21.3 MPH for a speed and delay run which began at Ives Dairy Road at 8:28 A.M. In that case, almost the entire distance between the Golden Glades Interchange and I-395 was run at 20 MPH or less. In fact, the average operating speed between 167th Street and I-395 was found to be just 17.3 MPH, with 33 minutes required to go less than 10 miles.

By 9:30 A.M., conditions had begun to improve, the overall average speed was found to be about 50.6 MPH. Only in a short section between 62nd Street and S.R. 112 did high levels of congestion occur.

figure 2-12 I-95 Miami-Dade county speed profile a.m. peak
Text Box: Figure 2-12  I-95 Miami-Dade County Speed Profile – A.M. Southbound Peak

 

Figure 2-13 shows comparable information in the northbound direction, for the P.M. peak period. In this case, the travel time runs began in downtown Miami just south of the I-395 Interchange and extended north to Ives Dairy Road. Beginning at 4:10 P.M. congestion is found over most of the length of I-95 south of the Golden Glades Interchange. The overall operating speed was computed at 31.1 MPH, operating speeds improved north of the Golden Glades Interchange, since there is a significant reduction in traffic north of Florida’s Turnpike.  By 4:44 P.M., average speeds dropped to just 19.8 MPH, with continual delays and travel speeds between 10 and 20 MPH for 9 continuous miles south of the Golden Glades Interchange. In fact, in this section of the run alone, speeds averaged just 14.6 MPH, with over 35 minutes required to travel only 8.5 miles. Free flowing managed lanes in this section of I-95 would save travelers 25 minutes or more.
 
Average speeds progressively improved as the P.M. peak period progressed, reaching about 30 MPH at 6:17 and 41.5 MPH at 7:06. Over time, the length of congested travel decreases, with the heaviest congestion lasting the longest time at a section between 95th Street and 151st Street.

figure 2-13I-95 Miami-Dade county speed profile p.m. peak
Text Box: Figure 2-13  I-95 Miami-Dade County Speed Profile – P.M. Northbound Peak

 

Table 2-3 presents a summary of average times and average speeds for all travel time runs performed in Miami-Dade County. This includes both peak and off-peak as well as northbound and southbound directions. As previously noted, heavy congestion exists in the southbound morning peak and northbound evening peak periods. In the off-peak travel directions, relatively free-flow conditions exist in the morning while some degradation of service is experienced in the southbound direction in the P.M. peak. Mid-day operating speeds are generally between 60 and 64 MPH, with the exception of one southbound run.

Table 2-3
Summary of Average Speeds
Miami - Dade County (Ives Dairy Rd - I-395)
13.5 Miles

 
SOUTHBOUND
NORTHBOUND
 
Start Time
Minutes
Average Speed
Start Time
Minutes
Average Speed
AM Peak
6:02
6:29
7:13
7:42
8:28
9:06
9:30
13.4
21.0
22.5
32.0
38.0
27.0
16.0
60.4
38.6
36.0
25.3
21.3
30.0
50.6
6:38
7:38
8:27
9:36
12.6
12.9
13.0
13.2
64.3
62.8
62.3
61.4
Mid-Day
10:10
11:24
12:34
1:09
3:23
13.0
13.7
13.0
12.6
15.3
62.3
59.1
62.3
64.3
52.9
11:33
12:24
1:33
13.5
13.0
12.8
60.0
62.3
63.3
PM Peak
3:52
4:18
5:40
6:29
15.1
23.5
14.8
15.1
53.6
34.5
54.7
53.6
4:10
4:44
5:20
5:57
6:17
6:56
7:06
26.0
41.0
40.0
35.0
27.0
22.0
19.5
31.1
19.8
20.3
23.1
30.0
36.8
41.5

Broward County Speed Profiles

Figure 2-14 shows peak period operating speed profiles in the Broward County portion of the study area. In this case, both northbound A.M. peak and southbound P.M. peak runs are shown.
 
For the most part, relatively little delay was found in the northbound direction, with the three peak period runs shown all averaging speeds in the vicinity of 60 MPH or higher. In the southbound P.M. peak, some congestion was found, with average speeds in the range of 25-30 MPH for the 8.3 mile trip between I-595 and Ives Dairy Road. However, by 4:34 P.M., still within the P.M. peak period, average speeds had improved to about 52 MPH.

figure 2-14 I-95 broward county speed profile
Text Box: Figure 2-14  I-95 Broward County Speed Profile

 

A summary of all travel time runs performed in Broward County are shown in Table 2-4. In general, operating conditions in the Broward County portion of the project are not as congested as in the Miami-Dade portion of the project. However, conditions are indicative of a “near breakdown” operation; that is, as traffic demands continue to grow, even modestly in future years, the length and duration of congested sections will likely increase. Also, even small incidents will likely cause a breakdown in normal operations.

Table 2-4
Summary of Average Speeds
Broward County (Ives Dairy Rd - I-595)
8.3 Miles

 
SOUTHBOUND
NORTHBOUND
 
Start Time
Minutes
Average Speed
Start Time
Minutes
Average Speed
AM Peak
6:14
7:11
7:36
8:01
8:45
7.8
7.7
8.5
7.8
7.6
63.9
64.7
58.6
63.9
65.6
6:18
7:18
7:43
8:09
9:01
7.5
8.0
8.6
8.2
7.8
66.4
62.3
58.0
60.7
63.9
Mid-Day
11:35
12:23
1:20
8.0
7.7
7.9
62.3
64.7
63.0
11:30
12:36
1:11
7.8
7.9
7.4
63.9
63.0
67.3
PM Peak
3:41
4:16
4:34
5:29
6:51
19.5
17.2
9.5
10.9
7.6
25.5
28.9
52.4
45.7
65.6
3:58
4:24
5:24
6:16
7.9
7.8
7.7
7.6
63.0
63.9
64.7
65.6

Back to Top
chapter 3 header

A detailed review of the socioeconomic characteristics of the area affecting I-95 was completed in 2005. Due to the importance of socioeconomic data used in the modeling of future travel demand, an independent economics consultant – The Washington Economics Group (WEG) was contracted to provide a detailed review of the most recent socioeconomic forecasts for Miami-Dade and Broward Counties.

Socioeconomic data is one of the most important input variables used in the traffic forecast modeling process. It is the primary factor in determining the number of vehicle trips that are generated within any given area. The intensity and relative orientation of housing and employment development influence the manner in which travel patterns may change over the projection period.

Details of the review and methods used in developing the updated projections were provided by WEG in a separate report submitted previously. However, a general overview of the patterns of growth that are expected to occur over the next 25 years in Miami-Dade and Broward Counties is included in this Chapter.

Population, Household and Employment Projections

The results of the WEG analysis are projections for population, housing, and employment in Miami-Dade and Broward Counties at discrete years – 2010, 2020, and 2030. Table 3-1 provides a summary of these projections on a county basis together with historical data beginning in 1980. Population and household unit forecasts through 2020 for Miami-Dade are expected to continue to grow at rates which are consistent with those during the previous 20 years. Beyond 2020, the rate of growth is forecasted to be 0.7 percent per year. Average annual job growth for Miami-Dade is projected at a rate similar to population growth, an average of 1.1 percent per year through 2020. Beyond 2020, the estimated rate of growth is reduced to 0.5 percent per year.

Table 3-1
Socioeconomic Trends and Projections

Miami-Dade County

Category
Population
Households
Employment
1980
1,625,509
609,830
AAPC
1.8
1.3
1990
1,937,194
692,355
AAPC
1.3
1.1
2000
2,204,713
774,339
1,183,282
AAPC
1.3
1.2
1.1
2010
2,503,991
870,475
1,320,134
AAPC
1.4
1.3
1.1
2020
2,873,373
990,607
1,479,598
AAPC
0.7
0.7
0.5
2030
3,096,134
1,066,939
1,558,727

Broward County

Category
Population
Households
Employment
1980
1,018,257
417,517

AAPC
2.1
2.4

1990
1,255,488
528,442

AAPC
2.5
2.2

2000
1,601,918
654,199
651,360

AAPC
1.0
0.9
0.9

2010
1,774,811
714,294
714,805

AAPC
1.3
1.0
1.3

2020
2,020,635
791,675
813,966

AAPC
0.5
0.4
1.0

2030
2,128,313
827,134
901,407

AAPC - Average Annual Percent Change
Sources: 1980, 1990 and 2000 from FDOT, 2010, 2020 and 2030 from WEG.

Broward County forecasts of future population and housing units are somewhat lower than was experienced from 1980 to 2000. From 2000 to 2020 the rate of growth is approximately one-half of past experience. As discussed in more detail in the WEG report, this change is based on a number of factors directly related to the fact that the supply of vacant land for residential development in Broward County is substantially depleted. Employment growth in Broward County is forecasted to be 0.9 percent per year through 2010 and 1.3 percent annually from 2010 to 2020. Beyond 2020, the rate of growth is forecasted to be 1.0 percent per year.

The forecasts for each category by county are shown graphically in Figure 3-1. The overall change from 2000 to 2030 of each of the forecast categories is also shown.

figure 3-1 population and household growth forecast  Text Box: Figure 3-1  Population and Household Growth Forecast

 

Forecasts by Regional Area

In developing its forecast, WEG divided the counties into 13 geographical areas – 8 within Miami-Dade County and 5 within Broward County. The 13 areas are displayed in Figure 3-2.
 
figure 3-2 economic analysis area
Text Box: Figure 3-2  Economic Analysis Area

 

Table 3-2 shows the population projections for each of the 13 regional areas. In Miami-Dade County, population is forecasted to grow by almost 300,000 people between 2000 and 2010, and by more than 369,000 between 2010 and 2020. Beyond 2020, forecasted growth in population over the next ten years is just under 223,000. In Broward County, population is expected to increase by 173,000 between 2000 and 2010, and by about 246,000 during the following ten years. From 2020 to 2030, population is forecasted to grow much slower at 108,000.

Table 3-2
Population Forecast by Subregional Area

Miami-Dade
Reg. Area
North East
North West
N. Central East
N. Central West
S. Central East
S. Central West
South East
South West

Subtotal

2000
378,411
261,098
533,351

116,873
473,044

269,917
105,822
66,197

2,204,713

Increment
28,780
28,978
54,010

30,342
30,102

56,844
51,974
18,248

299,278

AAPC
0.7
1.1
1.0

2.3
0.6

1.9
4.1
2.5

1.3

2010
407,191
290,076
587,361

147,215
503,146

326,761
157,796
84,445

2,503,991

Increment
35,214
23,250
53,114

43,801
31,669

64,092
82,444
35,798

369,382

AAPC
0.8
0.8
0.9

2.6
0.6

1.8
4.3
3.6

1.4

2020
442,405
313,326
640,475

191,016
534,815

390,853
240,240
120,243

2,873,373

Increment
25,432
15,758
41,468

22,326
27,820

49,629
25,457
14,871

222,761

AAPC
0.6
0.5
0.6

1.1
0.5

1.2
1.0
1.2

0.7

2030
467,837
329,084
681,943

213,342
562,635

440,482
265,697
135,114

3,096,134

Broward
Reg. Area
North
N. Central East
N. Central West
South East
South West

Subtotal
Total

2000
334,798

377,057

292,485
363,105
234,473

1,601,918
3,806,631

Increment
32,785

41,793

25,629
51,228
21,458

172,893
472,171

AAPC
0.9

1.1

0.8
1.3
0.9

1.0
1.2

2010
367,583

418,850

318,114
414,333
255,931

1,774,811
4,278,802

Increment
34,855

33,303

53,124
56,152
68,390

245,824
615,206

AAPC
0.9

0.8

1.6
1.3
2.4

1.3
1.4

2020
402,438

452,153

371,238
470,485
324,321

2,020,635
4,894,008

Increment
18,654

23,955

10,186
54,883
0

107,678
330,439

AAPC
0.5

0.5

0.3
1.1
0.0

0.5
0.7

2030
421,092

476,108

381,424
525,368
324,321

2,128,313
5,224,447

Within Miami-Dade County, the most rapid rates of future population growth are expected to be in the southeastern sector well south of the I-95 Corridor. Population is expected to increase from just under 106,000 in 2000 to over 240,000 by 2020. The north-central west sector, generally along Florida’s Turnpike Homestead Extension in Northwest Miami-Dade County, also is forecasted to show high levels of growth. Along the immediate project corridor itself, more modest rates of growth are forecast overall. These include the northeast and north-central east Miami-Dade County sectors, generally forecasted to grow at or below 1 percent per year in population.

In Broward County, the most rapid rates of growth are forecast in the southeast and southwest portion. The southeast sector, which is within the immediate vicinity of the I-95 Corridor, is expected to see relatively high levels of growth through 2020, with population increasing from just over 363,000 to over 470,000 by 2020. By 2030, population in the southeastern section of Broward County is expected to reach more than 525,000. The relative distribution of population growth by traffic zone is shown in Figure 3-3. Those traffic zones shaded in progressively darker shades of brown indicate areas of highest population growth. Along the immediate I-95 Corridor, this includes traffic zones within the vicinity of Hollywood within Broward County as well as some zones in far northeastern Miami-Dade County near North Miami Beach and Adventura.

The central core of the I-95 project, generally between the Golden Glades Interchange and Downtown Miami, is expected to experience only modest traffic growth as this portion of the area is largely built out.

figure 3-3 distribution of population growth 2000-2030
Text Box: Figure 3-3  Distribution of Population Growth (2000-2030)

 

Table 3-3 shows the household projections for each of the 13 regional areas. Because household and population growth are intertwined, the pattern of housing growth shown over the forecast period is consistent with that of the population growth pattern. The ratio of new persons to new housing units is approximately 3.0 throughout the forecast period for both counties. Over the forecast period, Miami-Dade County is expected to add almost 293,000 housing units, while Broward County is expected to add around 173,000 new units.

Table 3-3
Household Forecast by Subregional Area

Miami-Dade
Reg. Area
North East
North West
N. Central
East
N. Central
West
S. Central
East
S. Central
West
South East
South West

Subtotal

2000
133,419
88,944

200,535

39,583

168,696

86,950
34,278
21,934

774,339

Increment
8,688
9,139

18,493

9,821

9,449

17,652
17,191
5,713

96,146

AAPC
0.6
1.0

0.9

2.2

0.5

1.9
4.1
2.3

1.2

2010
142,107
98,083

219,028

49,404

178,145

104,602
51,469
27,647

870,485

Increment
11,395
7,612

18,606

14,505

10,826

19,475
26,290
11,413

120,122

AAPC
0.8
0.8

0.8

2.6

0.6

1.7
4.2
3.5

1.3

2020
153,502
105,695

237,634

63,909

188,971

124,077
77,759
39,060

990,607

Increment
8,658
5,050

15,202

7,508

10,056

15,635
9,402
4,821

76,332

AAPC
0.6
0.5

0.6

1.1

0.5

1.2
1.1
1.2

0.7

2030
162,160
110,745

252,836

71,417

199,027

139,712
87,161
43,881

1,066,939

Broward
Reg. Area
North
N. Central
East
N. Central
West
South East
South West

Subtotal
Total

2000
135,701

166,100

118,954
150,906
82,538

654,199
1,428,538

Increment
11,100

15,427

8,283
19,235
6,050

60,095
156,241

AAPC
0.8

0.9

0.7
1.2
0.7

0.9
1.0

2010
146,801

181,527

127,237
170,141
88,588

714,294
1,584,779

Increment
10,811

11,601

17,211
20,641
17,117

77,381
197,503


AAPC
0.7

0.6

1.3
1.2
1.8

1.0
1.2

2020
157,612

193,128

144,448
190,782
105,705

791,675
1,782,282

Increment
5,678

7,344

3,034
19,403
0

35,459
111,791

AAPC
0.4

0.4

0.2
1.0
0.0

0.4
0.6


2030
163,290

200,472

147,482
210,185
105,705

827,134
1,894,073

Figure 3-4 shows the distribution of household growth by traffic zone. Total incremental numbers of households between 2000 and 2030 are depicted in progressively deeper shades of color.

figure 3-4 distibution of household growth 2000-2030
Text Box: Figure 3-4  Distribution of Household Growth (2000-2030)

 

As with population, the largest number of new household units in the immediate corridor is anticipated in the southeastern portion of Broward County, generally in the Hollywood and Hallendale areas. Pockets of strong growth are also shown in portions of Miami Beach and North Miami Beach, while the immediate section of I-95 between the Golden Glades Interchange and Downtown Miami is forecasted to have only modest levels of new housing development.

Table 3-4 also shows the employment projections for each of the 13 regional areas. Over the forecast period, Miami-Dade County is expected to add more than 375,000 new jobs, while Broward County can expect to add 250,000. The employment forecast consists of industrial, commercial, and service related employment categories. In Miami-Dade County, 1.7 percent of job growth is forecasted to be industrial while 35.2 percent is forecasted to be commercial, with the remaining 63.1 percent falling into the service employment category. Broward County industrial employment growth is forecasted to account for 11.2 percent of the total employment growth, while commercial and service related employment growth is forecasted to account for 32.6 percent and 56.2 percent, respectively.

Table 3-4
Total Employment Forecast by Subregional Area

Miami-Dade
Reg. Area
North East
North West
N. Central
East
N. Central
West
S. Central
East
S. Central
West
South East
South West

Subtotal

2000
143,039
88,919

384,759

178,431

290,709

48,196
26,816
22,413

1,183,282

Increment
15,778
11,972

41,376

15,996

30,530

10,589
7,386
3,225

136,852

AAPC
1.1
1.3

1.0

0.9

1.0

2.0
2.5
1.4

1.1

2010
158,817
100,891

426,135

194,427

321,239

58,785
34,202
25,638

1,320,134

Increment
8,458
12,725

34,903

22,625

55,243

7,853
10,420
7,237

159,464

AAPC
0.5
1.2

0.8

1.1

1.6

1.3
2.7
2.5

1.1

2020
167,275
113,616

461,038

217,052

376,482

66,638
44,622
32,875

1,479,598

Increment
5,294
7,125

25,578

20,081

8,350

4,984
3,933
3,784

79,129

AAPC
0.3
0.6

0.5

0.9

0.2

0.7
0.8
1.1

0.5

2030
172,569
120,741

486,616

237,133

384,832

71,622
48,555
36,659

1,558,727

Broward
Reg. Area
North
N. Central
East
N. Central
West
South East
South West

Subtotal
Total

2000
125,630

231,554

91,786
151,296
51,094

651,360
1,834,642

Increment
17,370

11,137

9,684
14,942
10,312

63,445
200,297

AAPC
1.3

0.5

1.0
0.9
1.9

0.9
1.0

2010
143,000

242,691

101,470
166,238
61,406

714,805
2,034,939

Increment
20,345

15,256

19,036
16,602
27,922

99,161
258,625


AAPC
1.3

0.6

1.7
1.0
3.8

1.3
1.2

2020
163,345

257,947

120,506
182,840
89,328

813,966
2,293,564

Increment
27,821

14,989

10,411
20,338
13,882

87,441
166,570

AAPC
1.6

0.6

0.8
1.1
1.5

1.0
0.7

2030
191,166

272,936

130,917
203,178
103,210

901,407
2,460,134

Within Miami-Dade County, relatively high levels of employment growth are forecasted in the south-central west, southeast and southwest sectors. These areas are generally south of the I-95 Corridor. The most rapid rate of employment growth is forecasted generally east of U.S. Route 1 in the Homestead area. Heavy employment growth is also anticipated in areas west of the HEFT. Miami-Dade sectors in the immediate vicinity of I-95 show more modest levels of employment growth, generally averaging 1 percent per year through 2010 and lower rates thereafter. This is already a heavily built out section of the county.

Within Broward County, high levels of employment growth are forecasted for the southwest and north-central west sectors. Both of these sectors are along the I-75 Corridor, and considerably west of I-95. Along I-95 itself, about 50,000 jobs are expected to be added in the southeast Broward sector, with more than 65,000 added in the north sector.

The general distribution of employment growth between 2000 and 2030 is shown graphically in Figure 3-5. In general, the darker shades of green indicate areas of high employment growth.
 
As shown in Figure 3-5, employment growth is expected to be especially high between the HEFT and Palmetto Expressways, especially west of Miami International Airport. Other pockets of high employment growth include sectors along I-75 in Broward County and in the area between Fort Lauderdale and Hollywood within Broward County. Relatively strong employment growth is also shown in areas east of I-95 in the City of Miami.

figure 3-5 distibution of employment growth 2000-2030
Figure 3-5
Distribution of Employment Growth (2000-2030)

Summary

Considerably more detailed information about the economic analysis is included in the report prepared by WEG. That report provides an excellent summary of economic drivers in the county and a detailed rationale underlying the WEG review and adjustment of socioeconomic forecasts.

The economic forecasts were provided to WSA at the traffic analysis zone level. This was then used to develop updated trip matrices reflecting future travel demands at 2010, 2020, and 2030 levels. The socioeconomic forecasts provided by WEG varied slightly in some cases from the baseline forecasts used in the original SERPM model, although the differences were not significant.

The reader is encouraged to review the WEG report in its entirety for a more detailed description of Miami-Dade and Broward Counties economic forces and conditions.

Back to Top
chapter 4 header

Market research is an important element that provides key analysis inputs when conducting any comprehensive traffic and revenue study. As part of the overall I-95 Managed Lane Comprehensive Traffic and Revenue Study work program two (2) market research approaches were employed.
 
The first was in the form of stated preference surveys which were aimed at gauging the willingness of motorists to pay for a given amount of time savings. This value of time was derived through the stated preference survey process conducted on drivers currently utilizing the portion of I-95 which comprises the study corridor. The second was a public outreach approach which included the conduct of focus groups which tested the will and reaction of Miami-Dade County citizens to potential future strategies planned for the I-95 project corridor. This chapter provides a summary of the results of these two market research approaches.

Stated Preference Surveys  

The concept of the driver’s value of time is central in determining the revenue estimates for travel on a tolled facility. To understand its role in the estimation of tolled facility traffic and revenue assessments, one must broadly sketch out the overall modeling process. In broad terms, the overall modeling process can be described as follows:

  1. Demand for travel in a corridor is estimated.
  2. A fraction of that travel is allocated to transit; the remainder is allocated to the road network.
  3. Travel times are estimated on the tolled facility and its competing non-tolled facilities for various travel movements (origin-destinations). The amount of time savings offered by the tolled facility is estimated. The portion of the corridor traveling population that is willing to pay the prevailing toll for the estimated time savings is allocated to the tolled facility. From this, traffic and toll revenue are calculated for the tolled facility. This last step is reiterated within an equilibrium assignment process to allow for multiple alternatives for travelers.

To accomplish this assessment, one must know what the traveling population is willing to pay for a given amount of time saving. The value of time for the I-95 corridor in Miami was derived from stated preference surveys conducted of current drivers in the corridor by WSA’s subconsultant, Resource Systems Group (RSG). The full report summarizing the survey and its results is included as Appendix B. This chapter provides a summary of that report.

Survey Administration Plan

The survey was administered by RSG at various locations in Miami-Dade and Broward counties between June 18 and June 27, 2005. Intercept locations were chosen to access a diverse sample of travelers across income levels, race and ethnicity, and age. Two Miami area public relations firms, Communikatz and Charesse Chester & Associates (CCA) provided assistance in identifying key segments of the area population and developing strategies to enlist participation of these respondents.
 
Specific survey sites within each targeted location were venues with high pedestrian traffic in a setting where respondents are most likely to have time to stop for the survey. Locations that are close to the study corridor tend to have the highest incidence of people who have made a recent trip that used or could have used the study portion of I-95. Survey sites included driver’s license offices, shopping centers and retail locations, community and recreation facilities, colleges and community education centers.

The survey was developed in English, Spanish, and Creole languages to support a diverse sample of respondents. Attendants at each of the intercept sites were fluent in English, Spanish, and Creole, in addition to having expertise regarding the computer-based survey instrument.

Table 4-1 lists the intercept administration sites by date. More than 1,000 qualifying respondents completed the survey at intercept sites during the 11 days of administration.

Table 4-1
Intercept Survey Sites

Date
Friday, June 16, 2006
Sunday, June 18, 2006
Monday, June 19, 2006
Tuesday, June 20, 2006
Wednesday, June 21, 2006
Thursday, June 22, 2006
Friday, June 23, 2006
Saturday, June 24, 2006
Sunday, June 25, 2006
Monday, June 26, 2006
Tuesday, June 27, 2006
Wednesday, June 28, 2006
Thursday, June 29, 2006
Location
County Square Driver’s License Office
Aventura Mall
Dania Jai-Alai
Galleria Mall
Lauderhill Mall
Aventura Mall
Cobb Theater at Miami Lakes and Dolphin Mall
Galleria Mall
Lauderhill Mall
Florida Career College
Fort Lauderdale Driver’s License Office
Office Depot
Pembroke Road Driver’s License Office
Florida Career College
Miami Gardens Driver’s License Office
Miami Northside Driver’s License Office
Fort Lauderdale Driver’s License Office
Miami Central Driver’s License Office
Miami Northside Driver’s License Office
Bunche Park Driver’s License Office
County Square Driver’s License Office
Miami Northside Driver’s License Office
Galleria Mall
Lauderdale Lakes Driver’s License Office
North Dade Justice Center Driver’s License Office
Regions Bank
Aventura Mall
Cobb Theater at Miami Lakes and Dolphin Mall
Office Depot
Aventura Mall
Cobb Theater at Miami Lakes and Dolphin Mall

Miami Lakes Technical Education Center
North Miami Adult Education Center
Turner Technical Adult Education Center
County Square Driver’s License Office
County Square Driver’s License Office
Miami Central Driver’s License Office
Bunche Park Driver’s License Office
Bunche Park Driver’s License Office
Fort Lauderdale Driver’s License Office
Lauderdale Lakes Driver’s License Office
Toussant L'Overture Elementary School
County
Miami-Dade
Miami-Dade
Broward
Broward
Broward
Miami-Dade
Miami-Dade
Broward
Broward
Broward
Broward
Miami-Dade
Broward
Broward
Miami-Dade
Miami-Dade
Broward
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Broward
Broward
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Miami-Dade
Broward
Broward
Miami-Dade

Survey Approach

The stated preference survey was designed to identify the travel patterns and preferences of drivers who currently use or could reasonably use I-95 between I-595 and S.R. 836. The survey approach employed a computer-assisted self-interview (CASI) technique. The survey instrument was programmed using customized software developed by RSG for administration via laptop computers at activity sites and online through RSG’s SurveyCafe.com website. The customized computer-based survey software adapts to the trip characteristics of each respondent, making the questionnaire realistic for each individual, and it and presents complex ideas in a simple manner. Electronic validation of each question eliminates item nonresponse and prevents the entry of invalid inputs. Responses to each question are stored directly to a database, where they are available for review and tabulation.

Survey Questionnaire

The stated preference survey questionnaire was comprised of four main sections:

  • Trip description - These questions collected detailed information about each traveler’s most recent trip in the corridor. Information collected in this section included: origin, destination, whether the trip was based at home, day of week, time of travel, direction of travel, trip purpose, trip frequency, mode of transportation, vehicle occupancy, delays experienced, and current SunPass® usage, among others. If travelers did not use I-95 for this trip, they were asked why they did not use I-95 and whether they chose their time of day purposely to avoid congestion, if applicable.
  • Project Information – The second section of the survey provided a description of the proposed managed lanes, including the location and configuration of the lanes, the managed lanes concept, and requirements for all vehicles using the managed lanes to be equipped with a SunPass® transponder.
  • Stated Preference Scenarios – Respondents who traveled by automobile were presented with hypothetical stated preference scenarios based on the trip information they provided in the first section of the survey. In each of eight scenarios, respondents were asked to choose one of three options for a hypothetical future trip:
1. Driving alone on the new I-95 managed lanes with a toll.
2. Driving in a two- or three-person carpool on the new I-95 managed lanes with a toll or toll-free.
3. Driving toll-free on the I-95 general purpose lanes (for current I-95 users) or on their current route (for travelers who could have but did not use I-95)
  • For the carpool option, Option 2, occupancy requirements for free travel on the managed lanes and other parameters were varied based on the respondents’ current reported occupancy level. Travel times, toll costs, and vehicle occupancy are also key attributes of the travel options. These attributes were presented at different values or “levels” in each of the respondent’s eight scenarios, determined using characteristics of the respondent’s recent trip, such as the distance traveled or potentially traveled on I-95 and any tolls paid on the reported trip.
  • Debrief and Demographic Questions – In the last section of the survey, participants were asked follow-up questions designed to provide insight about why certain choices were made in the stated preference section, including why some respondents never selected the new managed lanes options; how new information might affect previous choices; and testing usage of potential new BRT and van services. Standard demographic information such as household size, gender, employment status, age, zip code, and household income was also collected in this section. At the end of the survey, respondents were also given the opportunity to provide free-form comments about I-95.

Survey Results

A total of 1,251 respondents completed the stated preference survey; 87 percent from respondents at intercept sites and the remainder through the Internet-based survey. The full details of the survey responses are provided in the RSG report in Appendix B. Some of the more significant responses to the survey questions are shown below.

Income and Demographic Characteristics of Survey Respondents

The sample was comprised of 59 percent men and 41 percent women. The median age of the sample was in the category of 35-44 years old; 85 percent were employed either part-time or full-time. A small portion (2 percent) reported no vehicle in the household but most (69 percent) had one or two vehicles. The median household income of the sample was between $20,000 and $39,999 per year, and nearly two-thirds reported household income less than $60,000 per year. This is consistent with the median household income for Miami-Dade and Broward Counties collected in the 2000 Census.

Time of Day and Purpose

figure 4-1 trip purpose by time of day
Text Box: Figure 4-1  Trip Purpose by Time of Day

Peak travel periods were defined as weekday trips between the hours of 6:00 A.M. and 9:00 A.M. or 4:00 P.M. and 7:00 P.M. Fifty percent of respondents reported a trip in the peak period, with 33 percent in the morning peak and 17 percent in the afternoon peak. Respondents making off-peak trips on weekdays were asked whether they were traveling outside peak periods to avoid traffic delays; 63 percent said they were.

Slightly more than half (53 percent) of all trips made during peak periods were work trips. Figure 4-1 shows the distribution of trip purpose across the survey sample by period.

Trip Duration

The median total travel time of trips in the sample was 45 minutes. The median of travel time spent on I-95 was in the category of 30-44 minutes. The median distance between entry/exit points on I-95 that were chosen by respondents (estimated distance traveled on I-95) was in the range of 9 to 13 miles. Figure 4-2 shows the distribution of total travel time.

Trip Frequency

figure 4-2 total travel time
Text Box: Figure 4-2  Total Travel Time

The frequency of travel for the trips described varied according to trip purpose. Almost 70 percent of trips to and from work occurred at least five times per week, while over 90 percent of shopping, social or recreational, and personal trips occurred less than five times per week and almost half of these types of trips occurred less than once per week (Figure 4-3)

HOV Lanes

All non-transit respondents who used I-95 for their trip were asked whether or not they used the HOV lane during their reference trip, regardless of their current vehicle occupancy. Half (52 percent) of all high occupancy vehicles (HOVs) used the HOV lane, and 21 percent of SOVs reported using it. A few travelers (5 percent) were not sure whether they used the HOV lane or not.

Current SunPass® Usage

After being provided with a description of a SunPass® transponder, respondents were asked whether they currently own one or if they plan to purchase one in the future. Twenty-eight percent of respondents said that they currently owned a SunPass® transponder; while 22 percent said they planned to purchase one. Half of the sample (50 percent) said that they did not have a transponder and did not plan to purchase one.

figure 4-3 trip purpose by frequency
Text Box: Figure 4-3  Trip Purpose by Frequency

Stated Preference Debrief Questions

After finishing the stated preference portion of the survey, the 36 percent of respondents who never choose the managed lanes option were asked why not. Respondents could select multiple reasons. Although 26 percent were opposed to paying a toll, 50 percent of respondents said that the time savings presented to them was not worth the toll cost, and 40 percent of respondents said that the toll was too high (Figure 4-4).

Modeling

The stated preference data from the survey were compiled into an ALOGIT dataset used to support estimation of the coefficients of a multinomial logit-based mode choice model and later estimation of individual coefficients for each respondent in the sample using a Hierarchical Bayes estimation technique1. Data from the choice experiments were expanded into a dataset that contained eight observations for each of the 1,233 automobile respondents, yielding a total of 9,864 observations. These eight choice observations consist of the data from individual survey screens, each of which was one experiment of a 16-experiment orthogonal design with three choice alternatives customized by current vehicle occupancy:

figure 4-4 reason for not selecting a managed laned lane alternative
Text Box: Figure 4-4  Reason for Not Selecting a Managed Lane Alternative

Single occupant vehicles:

  • I-95 new managed lanes driving alone with a toll paid using SunPass®;
  • I-95 new managed lanes in a carpool with SunPass® and a possible toll; and
  • Existing general-purpose lanes (or current non-I 95 route) with no toll.

Two-occupant carpools:

  • I-95 new managed lanes driving alone with a toll paid using SunPass®;
  • I-95 new managed lanes in a two-person carpool with SunPass® and a possible toll; and
  • Existing general-purpose lanes in a two-person carpool with no toll.

Carpools with three or more occupants:

  • I-95 new managed lanes driving alone with a toll paid using SunPass®;
  • I-95 new managed lanes in a three-person carpool with SunPass® toll-free; and
  • Existing general-purpose lanes in a three-person carpool with no toll.

A set of diagnostic model runs was conducted to ensure that the data were consistent with the original stated preference experimental design and to identify data outliers. Outliers in the data were identified in several ways, including identification of extreme values in the input data—very long travel times, unrealistic calculated vehicle speeds on I-95, and very short distances traveled on I-95 (less than 2 miles)—as well as the estimation identification of responses with low choice probabilities (for which the model showed a likelihood of selecting the option that was chosen of less than 0.025). Ultimately, 8,768 observations (data from 1,096 respondents) were used to estimate the model results in this report.

Traveler Segmentation

A total of four segments based on time of day and trip purpose were identified and modeled separately. The dataset is divided into four segments:

  • Peak period work trips;
  • Peak period non-work trips;
  • Off-peak period work trips; and
  • Off-peak period non-work trips.

Work trips included only travelers who were making a trip to or from work. Business-related travel was classified as non-work.

Two different peak specifications were tested: peak time of day only, and peak time of day and direction. In the first specification, peak period trips were defined as trips made by respondents who traveled on weekdays between 6:00 and 9:00 A.M. or 4:00 and 7:00 P.M. In the second specification, peak period trips were defined as southbound weekday trips between 6:00 and 9:00 A.M. or northbound weekday trips between 4:00 and 7:00 P.M. Trips made during the peak period in a reverse-commute direction were classified as off-peak. In both specifications, all weekend trips were considered off-peak. Ultimately, peak period by time of day only was found to provide the best model fit, and was used for the final specification.

General Model Structure

The quantitative relationships between the variables that affect choices and choice probabilities are specified in the utility functions. Coefficients for the variables in these utility functions can be statistically estimated using maximum likelihood procedures. A single general model structure was estimated for all model segments. The time coefficient applies to all options and the cost coefficient applies to the options that would or could possibly have a toll (managed lanes driving alone or in a two-person carpool). Some other coefficients are option-specific.

All model segments include the same variables, and separate coefficients values were estimated for each individual in the sample. The coefficients estimated in the model are shown in Table 4-2.

Table 4-2
Model Coefficients

Coefficient
Time
Cost
Distance
Two-person carpool
Three-person carpool
Carpool inertia
General purpose
lanes constant
Alternate route
constant
No SunPass®
constant
Units
minutes
dollars*
miles
(0,1)
(0,1)
(0,1)
(0,1)
(0,1)
(0,1)
Description
Total travel time in minutes applied to all alternatives
Toll cost in dollars divided by the log of household income in thousands of dollars divided by occupancy ^0.6
Log of total trip distance in miles applied to the managed lanes alternatives
Dummy variable applied to the alternatives shown as a two-person carpool
Dummy variable applied to the alternatives shown as a three-person carpool
Constant to capture willingness of current carpoolers to remain in a carpool
Constant applied to the general-purpose lanes alternative if the respondent’s trip used I-95
Constant applied to the alternate route alternative if the respondent’s trip did not use I-95
Constant to capture disutility of having to obtain SunPass® to use the managed lanes, applied to the general purpose lanes or alternate route alternative

The travel time coefficient is applicable to all alternatives and represents the sensitivity to total travel time in minutes.

Toll cost is applicable to all tolled alternatives and represents sensitivity to total trip toll cost in dollars. A respondent’s sensitivity to toll cost can vary with their income level. To account for this, several income effects on toll cost were tested. These effects included cost divided by a code representing the household income category, log of a household income code, household income in thousands of dollars, and the log of household income in thousands of dollars. Toll cost divided by the log of household income in thousands of dollars was found to be most statistically significant and was included in the model.

Various occupancy effects on toll cost were also tested to account for cost sharing among occupants traveling in HOVs. Toll cost divided by occupancy to the power of 0.6 was found to be statistically significant and included in the model.
 
Managed lanes distance is a positive coefficient for managed lanes options, indicating an increased likelihood of using the managed lanes for longer distance trips.

The two- and three-person carpool variables represent the sensitivity to number of vehicle occupants. The carpool inertia coefficient captures the increased likelihood of choosing carpool for travelers who are already in a carpool.

The general-purpose lanes constant and alternate route constants are applied to the third alternative and capture utility or disutility of the non-managed lanes alternative that cannot be explained by the other variables in the model such as time, cost, and occupancy.

No SunPass® transponder is a variable included on the general purpose lanes that shows an increased likelihood of using general purpose lanes (rather than managed lanes) for travelers who do not currently have SunPass® and do not plan to acquire it.
 
Several other variables and constants were tested beyond what is presented in this final specification, but were found to be insignificant or to worsen the overall fit of the model to the data.

Model Coefficients and Value-of-Time

After testing hundreds of multinomial logit model specifications using ALOGIT, a final specification was selected and used for the individual coefficient estimation using a Hierarchical Bayes technique. This produces individual choice model coefficients for each respondent in the sample, which in-turn provides implied values of travel time savings for each individual respondent as well as the distribution of values of travel time savings across the sample.

The implied value that travelers place on their time when making a route choice decision on their reported trip is determined using the marginal rate of substitution of the travel time and toll cost coefficients. In addition to using the individual-specific time and cost coefficients, the individuals’ reported income level and vehicle occupancy are used in the computation of individual value of time. The median values of time for the segments range from $7.65 per hour to $9.30 per hour (Table 4-3).

Table 4-3
Median Values-of-Time

 
Median Value-of-Time
Travel Segment
Peak  work
Peak  non-work(1)
Off-peak  work
Off-peak  non-work
$ per hour
$7.65
$9.30
$8.37
$8.06
$ per minute
$0.128
$0.155
$0.140
$0.134

(1) Adjusted from raw data indicated $13.01 per hour and $0.22 per minute.

Focus Groups

Public involvement is a crucial element of any public project. Recognizing the importance for public involvement WSA engaged Kelley Swofford Roy, Inc. (KSR), a Florida based marketing firm to assist with the public outreach task of the overall I-95 Managed Lane work program.

Focus Group Administration Plan

The public outreach program associated with the  Proposed I-95 Managed Lane analysis relied on focus groups to test the will and reaction of Miami-Dade County citizens to potential strategies planned for the project corridor portion of I-95. The value of input and reaction from individual Miami-Dade citizens will play a critical role in the decision-making process as FDOT considers future potential strategies for I-95.

FIGURE 4-5 focus group participant distribution
Text Box: Figure 4-5  Focus Group Participant Distribution

A total of 196 citizens participated in 15 sessions conducted between October 27, 2004 and February 28, 2005. Six of the sessions, including 81 participants, were conducted among I-95 roadway users who drive primarily in single occupancy vehicles (SOV) to commute to and from work. Four of the sessions, included 55 participants, were conducted among I-95 roadway users who drive primarily with one or more passengers and use the I-95 HOV lanes to commute to and from work. Both the participants in the SOV and HOV lanes confirmed as a prerequisite that they drive at least 10 miles on I-95 in Miami-Dade County on most work days. Additionally, five of the sessions, including 60 participants were conducted among Miami-Dade residents who use Miami-Dade transit as their primary means to commute to and from work.        

In total 30 percent of the 196 participants were Hispanic, 35 percent Caucasian, 29 percent African American and 6 percent Creole as indicated in Figure 4-5. Each of the sessions averaged close to two hours in length. Twelve of the 15 focus groups were conducted in English, two in Spanish, and one in Creole. One of the Transit panels was composed exclusively of African heritage participants. The two Spanish panels, one an SOV panel and the other a Transit panel, recruited participants whose first language is Spanish.

A total of 96 men and 100 women comprised the 196 participants. In the SOV panels there were 41 men and 40 women. In the HOV panels there were 26 men and 29 women. In the Transit panels there were 29 men and 31 women. Occupations were diverse and very representative including clericals, government, professionals, and a few students.Participants for the sessions were selected from each of the major zip code groupings within Miami-Dade County to insure a broad countywide sampling of opinions. Because the selection process was stratified in this way the quantitative conclusions and percentages represent the participants’ opinions on the topics covered in the groups and no specific conclusions can be made on the margin of error if these results were projected countywide.

Focus Group Survey Methodology

The survey methodology utilized in the focus groups was developed by Dr. William R. Roy. The Foquand(sm) process is more formalized than traditional qualitative focus groups. It uses much larger focus panels averaging from 12 to 14 participants. Under this methodology participants recorded their reactions to a topic area, both quantitatively using rating scales and qualitatively by completing open-end questions, before the open discussion begins on that topic area.  This made it possible to assess individual perceptions and opinions before the discussion begins to modify those reactions. As a result the individual reaction of each of the participants is included in this analysis, not just the comments of the discussion participants, as is the case in traditional focus groups. This provides a much broader base of analytic information than is possible through the traditional focus group approach.

Percentages shown subsequently in this section are always based on the number of respondents, which can change from one question to another because all respondents did not always record their answers to all questions. On rating questions, the median was used for comparative purposes. This was arrived at by arraying all respondents’ ratings from the highest to the lowest and then identifying the rating that is halfway down the scale as the median. For example if there were 51 responses to a rating question, the rating given by the 26th respondent down the rating scale would be the median. The percentage of all respondents who gave a rating at the high end of the scale that is a 10, 9, or 8 as well as the percentage of all respondents who gave a rating at the low end of the scale that is a 3, 2, or 1 was also examined. In most cases the majority of ratings tend to skew toward one end of the scale or another. However in some cases we find a “polarized” response. This means that there were substantial differences of opinion with nearly equal ratings at both the high and low ends of the scale.
 
The full report summarizing the focus group results is included as Appendix C of this document. This section provides a summary of that report.

Summary of Focus Group Results

As stated previously, the focus group session participants represented a broad cross section of Miami-Dade I-95 single occupant vehicle drivers (SOV), I-95 two or more occupant vehicles (HOV) and Transit users, most of whom did not use I-95 for their Transit commute. The panels were conducted between October 27, 2004 and February 28, 2005. Approximately 30 percent of the participants were Hispanic, 35 percent Caucasian, 29 percent African American, and 6 percent Creole. A brief summary of the most important findings follows:

  • 48 percent of SOV and 36 percent of HOV participants drive cars equipped with SunPass®. The majority of both SOV and HOV participants used I-95 to commute to work, occasionally into other counties. 86 percent of the SOV and 91 percent of HOV participants indicated that their trip on I-95 takes them into Broward and Palm Beach counties, at least occasionally;
  • 58 percent of HOV participants use car or vanpools compared to 16 percent for SOV participants, who are reluctant to add passengers. Most HOV/SOV users would not favor mass transit unless schedules/service improved significantly;
  • When asked to speculate on what it would take to get SOV and HOV participants to consider using mass transit “more routes/destinations” and “better schedules” were mentioned the most;
  • Both SOV and HOV users were generally satisfied with I-95. The biggest negative in using I-95 is “too much traffic,” for which the HOV lane provides a satisfactory peak hour alternative. SOV participants also said I-95 was “quicker.” Both SOV and HOV participants mentioned “more exits than toll roads” as a positive;
  • SOV users gave higher marks to FDOT’s care of I-95 than HOV users. Most liked the idea of extending the HOV lane’s length, but opposed extending hours of operation to 24/7;
  • When the concept of “managed lanes based on open road tolling” was introduced in the panels, the majority of SOV and Transit participants liked the idea while HOV users were more polarized and far less positive, because it would disrupt their use of the HOV lanes. However, 76 percent of both SOV and HOV participants acknowledged they would use the managed lanes, at least occasionally, with nearly 20 percent saying they would use them all the time, if they could afford to. Only 11 percent of SOV participants and 20 percent of HOV participants say they will never use these lanes because they “don’t want to pay any tolls”;
  • When the topic of specific toll levels for the managed lanes was introduced, beginning with a $6 one way toll, most SOV and HOV participants became more hesitant because of cost with SOV showing more price elasticity than HOV participants, who showed some interest in adding passengers to travel free. The toll discussion turned off more HOV than SOV participants toward the “managed lanes” concept;
  • The “managed lanes” toll discussion produced increased interest in considering express bus service or adding passengers or considering a vanpool if it was convenient as an alternative to paying high tolls, even though 54 percent of the SOV participants and 25 percent of the HOV participants still said they would “most likely continue to commute by myself in my car”;
  • Participants were given more information about the economic and job potential associated with mixed use development areas along the I-95 “managed lanes.” When, following the description of these economic development areas, participants were asked to evaluate a series of statements about the benefits of “managed lanes,” overall reactions were consistently much more positive than negative;
  • When participants were given a list of statements about the degree to which the addition of the I-95 managed lanes will improve the community, most agreed it would be very positive;
  • 48 percent of the Transit participants commute by transit part of the time and get a ride with friends or in a car or van pool the rest of the time. 44 percent use transit or transit plus walking all the time. 40 percent use transit because it is “less stressful;” 38 percent because it “saves money;” 37 percent because they have “no car,” and 32 percent because it is the “easiest way to get there.” Overall, Transit participants were “satisfied” with MDT; and
  • While many Transit participants were happy with their current jobs, they acknowledged that transit limitations restricted their career options. Many saw the “managed lanes” concept as a way to improve and expand transit service, thereby leading to improved career and living opportunities. 56 percent said better public transportation choices are “very important” in their ability to get a better job.

Back to Top
chapter 5 header

As discussed in previous chapters, a total of six project alternatives were subjected to preliminary traffic and revenue analysis. This included the development of multi-level detailed traffic models, an assessment of optimum toll levels, estimation of traffic volumes by time of day and travel direction, estimation of revenue potential and a general assessment of operational benefits. Each of the six physical project alternatives were considered under two operational scenarios:

  • Assuming vehicles with two or more occupants could use the facility toll-free (HOV-2+); and
  • Assuming vehicles with three or more occupants could use the facility toll-free (HOV-3+).

In practice, Alternatives 1, 2 and 3 were found to be impractical in peak periods under the HOV-2+ scenario, and parts of the full analysis were undertaken for these alternatives only under the HOV-3+ option.

This Chapter is intended to concisely summarize the results of this extensive alternatives analysis. Detailed study findings were presented to the Department, together with a comparative analysis to identify the most promising options. Those most promising alternatives were subjected to more refined, detailed analysis; the results of which are presented in Chapter 6.

In addition to an overview of the analytical approach used, this Chapter summarizes the results of the initial alternatives analysis. A general summary of traffic and revenue potential is included for each alternative. More detailed information regarding estimated traffic for each alternative is presented in a series of graphics included in the Appendix to this report.

Modeling Approach

figuire 5-1 modeling approach
Text Box: Figure 5-1  Modeling Approach

Figure 5-1 presents an overview of the methodology used to develop estimates of traffic and revenue for each of the six project alternatives. A traffic and revenue study attempts to answer three overall questions:

  • How much demand exists in the corridor today;
  • How much will the demand grow in the future; and
  • What share of traffic can be expected to use the toll facility and what will drivers be willing to pay?

A detailed profile of existing demand was collected as part of the study and was summarized previously in Chapter 2. This included hourly traffic profiles by travel direction, travel time surveys, vehicle occupancy counts and more. This became the basic foundation of the travel demand models used in the analysis.

The overall modeling approach used in the study actually required the development of three independent models.
These included:

  • Regional model for global demand estimates – The global demand is an estimate of the total amount of traffic that would be expected to be using the I-95 corridor, under both the existing or improved conditions. These estimates were based on the results of the independent economic growth assessment for the region that was performed as part of the study, and made use of the Tri-County SERPM regional model.
  • Travel time simulation model - A traffic model of the I-95 Managed Lane corridor was developed using the VISSIM microsimulation software to identify changes in travel time and delay on different segments of the general purpose lanes under various alternative project configurations. Motorists’ willingness to pay a toll to use express toll lanes depend heavily on congestion levels and traffic backups in the toll-free general purpose lanes. The simulation model is an enhanced tool to provide a better indication of delay conditions under varying levels of demand.
  • The market share micro-model – This model is used to estimate the share of total traffic on I-95 that would choose the managed lanes versus the toll-free general purpose lanes under varying operating conditions and toll rates. Micro-model trip tables were calibrated directly to the observed base year hourly control volumes on I-95 mainline and ramp locations, as discussed above. The share of corridor traffic in the managed lanes is based on several factors, including location of access points, differences in configurations, time savings offered by the managed lanes and, of course, the toll rates being charged.

Global Demand Estimates

The corridor global traffic demand is defined as the total traffic traveling in the I-95 Managed Lane corridor. Global demand estimates were prepared using trip tables developed using modified socioeconomic forecasts with the SERPM regional travel demand model. The Washington Economics Group, Inc. (WEG), a Miami-Dade County based economic consultant, conducted an independent review of the underlying socioeconomic forecast used to develop the SERPM model trip tables. Based on this review, revisions were made to the SERPM socioeconomic data sets at traffic zone level. These revised socioeconomic data sets were then used to develop revised trip tables for use in this study analysis.

The regional travel demand model was used in two ways: to provide the base travel patterns for the micro-model subarea and to develop growth characteristics for the micro-model subarea.

The calibration process for the regional model used for this study included the following steps:

  • Develop trip tables at 2004, 2010, 2020, and 2030 levels for a.m., midday, p.m., and overnight time periods. Divide trip tables into SOV, HOV-2, HOV-3+ and truck components;
  • Adjust assignment parameters including link speeds and capacities, and speed/flow relationships; and
  • Extract micro-model subarea travel information from base and future years.

The SERPM network included the latest Transportation Improvement Program (TIP), as defined by the MPO’s in the three county areas that are covered by the model. Specific modifications to this plan related only to assumptions regarding the six Managed Lane (ML) alternative project configurations. As noted earlier, a set of independent economic forecasts were developed for use in this study. These housing and employment forecasts were used to develop updated trip tables for use in this study.

The subarea trip tables used in the micro-model were initially extracted from regionwide traffic assignments at base-year (2004) levels. These trip tables were used as “seed matrices” in a calibration process that adjusted the trip tables to traffic volumes representing the average hourly volume (for each of the time periods) for I-95 ramps and mainlines for the analysis intervals used in the micromodel, which are smaller than those used in the regional model. The hourly traffic volume profile summarized previously was used to identify appropriate analysis intervals for use in this study.

The analysis periods used in the windowed model and the micro-model have been defined as follows:

  • A.M. 1 – A. M. Pre-Peak Shoulder – 6:00 - 7:00 A.M.
  • A.M. 2 – A.M. Peak Period – 7:00 - 8:00 A.M.
  • A.M. 3 – A.M. Post Peak Shoulder – 8:00 - 9:00 A.M.
  • MD – Midday Period – 9:00 A.M. - 3:00 P.M.
  • P.M. 1 – P.M. Pre-Peak Shoulder - 3:00 - 4:30 P.M.
  • P.M. 2 – P.M. Peak Period - 4:30 - 6:00 P.M.
  • P.M. 3 – P.M. Post-Peak Shoulder - 6:00 - 7:00 P.M.

The overnight period from 7:00 P.M. to 6:00 A.M. was not analyzed explicitly. The traffic and toll revenue forecasts presented later in this report assume a certain fixed percentage of traffic and revenue will occur during the overnight hours, as well as on weekends.

Future-year (2010, 2020, and 2030) traffic assignments using the regional model were made to identify potential changes in travel patterns in the corridor. Among other things, these travel patterns are likely to be affected by:

  • Forecasted growth in the region;
  • The addition of new capacity to the freeway in the form of the MLs being studied herein;
  • Highway improvements to other freeways in the region; and
  • Changes to the ramp configurations as part of these improvements.

Trip tables representing the micro-model subarea were extracted from each set of runs and compared to those developed for the base year to estimate zonal growth rates, which were then applied to the calibrated base-year subarea matrices.

Travel Time Simulation Model (VISSIM)

Traditional traffic assignment models do not well replicate the impact of merging and weaving maneuvers on freeway capacity, nor can they reflect the impact of downstream queuing on freeway segments. WSA has used a microscopic simulation model called VISSIM to assist in estimating the impacts of the project travel speeds on different segments of the freeway. VISSIM attempts to model each vehicle as a separate entity. The roadway geometry and interaction with other vehicles influences the behavior of each vehicle in the model. A certain level of randomness in vehicle behavior was also introduced.

A series of VISSIM runs were made using differing assumptions on traffic shifts to the MLs for each of the six analysis time periods, at 2010 and 2020 levels. As traffic shifts into the MLs, the amount of traffic in the general purpose lanes would decrease, resulting in lower congestion levels in the general purpose lanes. A total of six runs were made for each of the six primary analysis periods for each direction. Within each time period, for each link, a relationship was developed between the “traffic demand” on the link and its modeled travel speed. By graphing the relationship between traffic demand and travel speed for all six runs for each mainline segment, WSA developed scenario-specific volume-delay curves for each mainline link on the general purpose lanes.

Each link in the micro-model was then tagged with a user code to identify a curve to be used to estimate travel speeds for that link during the micro-model assignment process. Links with less weaving and merging tended to be able to accommodate higher traffic volumes at higher speeds before breaking down. Certain sections of the freeway, which may have a large entering ramp volume, tended to break down at lower demand levels, and also may break down more quickly. Other sections of freeway may appear to break down at relatively low levels of demand, but may actually be affected by downstream congestion and queuing from these downstream bottlenecks.

Market Share Micro-Model

The extracted micro-model subarea used for this study is of a size that covers the freeway from south of the   I-95/S.R. 836 Interchange to north of I-595, and includes arterials and other freeway links within three to four miles on either side of I-95. The micro-model package included six alternative networks and three sets of alternative trip tables that were used to estimate traffic and revenue for the various combinations of project configuration/tolling alternatives.

In the micro-model, travel time between a path using the tolled MLs was compared to travel time on a path using the next best free routes (most likely the general purpose lanes or frontage roads). For each travel movement, the proportion of motorists expected to use the MLs is a function of the computed time savings and the cost to use the lanes (cost-per-minute saved) vs. the value placed on time savings by the motorist value of time (VOT).

The share of each traffic movement that is captured by the MLs is based on an estimate of the assumed distribution of the VOT, also developed from the stated preference surveys. It was assumed that motorists with a VOT greater than the cost per minute saved would tend to choose the MLs while those with a lower VOT would tend not to choose the lanes.

The micro-model relies on developing an equilibrium condition between the toll cost and the estimated time savings. If more traffic uses the MLs, there is less congestion in the free lanes and lower time savings. Less time savings would result in less traffic choosing the MLs. For each toll rate level, there exists an equilibrium point between the level of traffic congestion in the free lanes (time savings) and the amount of traffic willing to pay a toll to save that same amount of time. At low toll levels, there is a higher propensity to use the MLs, and there is a lower congestion level in the free lanes. At higher toll levels, there is less traffic in the MLs and also more congestion in the free lanes.

A full range of toll rates were tested, from $0.05 per mile to $0.60 per mile, for each time period and travel direction. The toll rates chosen for use in the traffic and revenue analysis generally reflect those that maximize revenues for each individual time period. During certain peak periods in the 2020 and 2030 assignments, checks for capacity constraints in the MLs indicated a need to use higher toll rates to manage demand to maintain an acceptable level of service at one or two locations in the system. A higher range of toll rates were tested and chosen in those cases, which is reflected in the traffic and revenues presented subsequently in this Chapter.

Vehicle Categories

The micro-model for each of the three trip tables were separated into five components: SOV work, SOV non-work, HOV-2 work, HOV-2 non-work and HOV-3+, are assigned simultaneously until an equilibrium condition was reached for that particular toll rate. In the tolling alternatives that involved free passage for either HOV-2+ or HOV-3+ traffic, the HOV traffic was allowed free access to the MLs.

Before the analysis began, FDOT determined that trucks would not be allowed to use the MLs. As a result, WSA separated out a portion of the micro-model trip tables to represent trucks. This trip table was assigned to the arterial streets and general purpose lanes only and used as a preload volume for the main equilibrium assignments.

It is noted that the preliminary traffic and revenue analyses conducted for the six project alternatives initially assumed lower proportions of vehicles with three or more occupants than the final analysis (Chapter 6). The lower levels were consistent with currently observed HOV percentages along existing sections of I-95. However, once the two most promising alternatives were identified, trip tables were refined to allow for some increase in the share of vehicles with three or more occupants in future years. This reflected the potential for increases in ridesharing in the future as congestion levels increase and the restricted use of HOV lanes in other parts of the region may be limited to vehicles with three or more occupants. The higher HOV-3+ share assumptions also provided a slightly more conservative platform when estimating traffic and revenue.

Hence, traffic and revenue results for the preferred alternatives as described in Chapter 6 are slightly different than those described in Chapter 5.

Alternatives Evaluated

Table 5-1 presents a summary of the six alternatives analyzed in the analysis, including the different assumptions regarding vehicle eligibility and whether or not tolls are charged or free access is provided.

Alternative 1 would simply use the existing HOV lane capacity, and would convert the HOV lanes to HOT operation. It was initially assessed under both operating scenarios (that is HOV-2+ and HOV-3+ being considered free). Preliminary analysis showed, however, that there was very little capacity available to “sell” to SOV traffic if vehicles with two or more occupants are continued to be allowed toll-free access to the HOT lanes in the future.

Alternative 2 featured the use of two reversible HOT lanes south of the Golden Glades Interchange. This would result in the elimination of one HOV lane (in the minor travel direction). North of the Golden Glades Interchange, the single HOT lane in each direction would continue to be used.

Again, the initial assessment looked at both HOV-2+ and HOV-3+ being free. The preliminary analysis showed that after 2010, sufficient capacity to sell to single-occupant vehicles would be available only during shoulder and off-peak hours; hence, only limited analysis of this option was undertaken. A full analysis of the HOV-3+ alternative was completed.

Alternative 3 involved a three-lane HOT system south of the Golden Glades Interchange, which would incorporate the use of a moveable barrier. This would permit two HOT lanes in the major travel direction while still retaining one HOT lane in the minor direction. The preliminary analysis showed that this scenario would also have difficulty under an HOV-2 definition, at least in the minor travel direction.

Alternative 4 was the first of three alternatives involving the assumed construction of an elevated roadway segment south of the Golden Glades Interchange. In this case, the existing HOV lanes would be retained at the lower roadway level and available for carpools only. Two additional express toll lanes would be constructed, most likely elevated above the existing roadway. In Alternative 4, two reversible express toll lanes would be implemented, with both lanes operated southbound in the morning peak and northbound in the afternoon peak. This scenario would be viable under both the HOV-2+ and HOV-3+ scenarios, recognizing that HOV traffic would be allowed to use the HOV lanes themselves south of the Golden Glades Interchange. This is shown to the right side of Table 5-1.

Alternative 5 would involve the construction of a three-lane elevated roadway and the retention of the existing HOV lanes south of the Golden Glades Interchange. The three elevated express toll lanes would include a moveable barrier which would permit two lanes to be operated in the major direction and one lane to be operated in the minor direction.

Finally, Alternative 6 would be the same as Alternative 5, except four elevated express toll lanes would be constructed, two in each travel direction. The existing HOV lanes would be retained at existing roadway levels south of the Golden Glades Interchange, while north of the Golden Glades Interchange the single HOV lane in each direction would be converted to HOT operation.

Table 5-1
Proposed I-95 Managed Lanes Alternative Evaluation Assumption

ML North of Golden Glades
ML North of Golden Glades
HOV South of Golden Glades
Alternative
1
2
3
4
5
6
Toll Scenario
HOV2+ free
HOV3+ free
HOV2+ free
HOV3+ free
HOV2+ free
HOV3+ free
HOV2+ free
HOV3+ free
HOV2+ free
HOV3+ free
HOV2+ free
HOV3+ free
SOV
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
HOV2
Free
Pay
Free
Pay
Free
Pay
Free
Pay
Free
Pay
Free
Pay
HOV3+
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
SOV
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
Pay
HOV2
Free
Pay
Free
Pay
Free
Pay
Pay
Pay
Pay
Pay
Pay
Pay
HOV3+
Free
Free
Free
Free
Free
Free
Pay
Pay
Pay
Pay
Pay
Pay
SOV
---
---
---
---
---
---
Banned
Banned
Banned
Banned
Banned
Banned
HOV2
---
---
---
---
---
---
Allowed
Banned
Allowed
Banned
Allowed
Banned
HOV3+
---
---
---
---
---
---
Allowed
Allowed
Allowed
Allowed
Allowed
Allowed

Optimum Toll Rates

Using the calibrated “micro-model,” a series of traffic assignments were made at various toll rates at each of three analysis years, 2010, 2020, and 2030. A range of toll rates were tested for each travel time period (e.g., a.m. shoulder, a.m. peak, midday, etc.). Working toll sensitivity curves were prepared for each time increment, showing the relative revenue potential at each of about a dozen alternative toll rates. In addition, traffic assigned to the managed lanes was reviewed for reasonableness, and for potential overloads based on available capacity on each project cross section.

For each analysis year, and each alternative, under both the HOV-2+ (where viable) and the HOV-3+ scenarios, an optimum toll rate was selected. Rates were assessed in terms of cents per mile. For purposes of this analysis, it was assumed that the same per-mile toll rate would be in effect over the entire length of the managed lane facility, in each respective travel direction. Differential optimum rates were often found for northbound versus southbound travel, but an overall consistent rate was used per direction over the entire length of the facility.

In practice, depending on alternative, there were varying amounts of available capacity to sell to non-HOV traffic. As such, optimum toll rates may vary by section of the road. It was often found, for example, that toll rates needed to manage demand in the northern portion of the project, where just one HOT lane was available in each travel direction, determined the optimum toll rate over the entire length of the facility. In a more refined analysis, it may be appropriate to consider variable rates by point within the system.

Table 5-2 shows optimum toll rates for each of the six project alternatives under the HOV-2+ free operating scenario. That is, vehicles with two or more occupants were assumed to be allowed to use the managed lanes without paying a toll.

As shown in Table 5-2, in many cases, there was little or no capacity available to sell to non-HOV traffic under the HOV-2+ condition. This is highlighted in green shading, with no optimum toll rate shown. During this period of time, in the respective analysis years, it was determined that HOV-2+ toll-free demand would reach, or nearly reach, the total available capacity of the HOT lane.

As shown in Table 5-2, under Alternative 1 there was little or no capacity to sell to non-HOV traffic at virtually all daylight hours by 2010 and, of course, beyond. Under Alternative 2, which featured two managed lanes in the major travel direction south of the Golden Glades Interchange, adequate capacity would be available in all periods of 2010, but would generally not be available in peak hours at 2020 and 2030 levels. Similar results are shown for Alternative 3 under the HOV-2+ condition; ultimately yielding a conclusion that Alternatives 1, 2 and 3 would only be operationally viable, over the long-term, if the definition of carpool was increased from two occupants to three occupants.

Table 5-2
Optimum Per Mile Rates
HOV 2+ Operating Scenario

   
Northbound
Alternative
2010
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
-
-
-
-
-
-
-
-
2
0.10
0.45
0.35
0.05
0.10
0.25
0.30
0.15
3
0.20
-
-
-
0.05
0.35
0.30
0.25
4
0.05
0.25
0.10
0.05
0.15
0.25
0.25
0.20
5
0.10
0.35
0.20
0.10
0.10
0.25
0.25
0.20
6
0.10
0.40
0.25
0.10
0.10
0.25
0.25
0.20

   
Southbound
Alternative
2010
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
-
-
-
-
-
-
-
-
2
0.20
0.60
0.40
0.10
0.05
0.15
0.20
0.15
3
0.25
0.60
0.45
0.05
-
-
-
-
4
0.20
0.35
0.30
0.15
0.05
0.10
0.15
0.05
5
0.20
0.35
0.30
0.05
0.05
0.20
0.25
0.15
6
0.20
0.35
0.30
0.05
0.05
0.20
0.25
0.15

 

   
Northbound
Alternative
2020
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
-
-
-
-
-
-
-
-
2
0.30
-
0.40
0.20
0.35
-
-
-
3
-
-
-
-
0.35
-
-
-
4
0.15
0.45
0.25
0.05
0.20
0..35
0.30
0.30
5
0.35
0.55
0.45
0.20
0.20
0.35
0.30
0.30
6
0.35
0.50
0.35
0.25
0.25
0.35
0.30
0.30

   
Southbound
Alternative
2020
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
-
-
-
-
-
-
-
-
2
0.40
-
0.60
0.35
0.15
0.30
-
0.30
3
0.40
-
0.45
0.35
-
-
-
-
4
0.25
0.45
0.45
0.20
0.05
0.25
0.35
0.15
5
0.25
0.45
0.45
0.20
0.20
0.35
0.40
0.30
6
0.25
0.45
0.45
0.25
0.25
0.35
0.35
0.30

 

   
Northbound
Alternative
2030
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
-
-
-
-
-
-
-
-
2
0.45
-
-
0.20
0.45
-
-
-
3
-
-
-
-
0.45
-
-
-
4
0.45
-
-
0.05
0.20
0.35
0.35
0.35
5
0.40
0.60
0.55
0.20
0.20
0.35
0.35
0.35
6
0.35
0.90
0.45
0.25
0.25
0.35
0.35
0.35

   
Southbound
Alternative
2030
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
-
-
-
-
-
-
-
-
2
0.45
-
-
0.20
0.45
-
-
-
3
-
-
-
-
0.45
-
-
-
4
0.45
-
-
0.05
0.20
0.35
0.35
0.35
5
0.40
0.55
0.40
0.20
0.20
0.40
0.45
0.35
6
0.40
0.55
0.40
0.25
0.25
0.35
0.35
0.30

green bgIndicates time periods when Managed Lanes would accommodate HOV's only.

yellow bgIndicates time periods where demand is above available capacity along selected project mainline segments at optimum per mile toll rates.
Per Mile Toll Rates Calculated in 2005 Dollars

For Alternatives 4, 5 and 6, optimum toll rates are shown in almost all operating hours. Only in the major direction peak hour in 2030 was this found to not be viable under Alternative 4, and it would be viable throughout under Alternatives 5 and 6.

Under Alternative 4, for example, in the southbound travel direction, the optimum toll in the a.m. pre-peak shoulder period (between 6:00 and 7:00 A.M.) was found to be $0.20 per mile in 2010. This increased to $0.25 by 2020 and to $0.40 by 2030. In the morning peak hour, between 7:00 and 8:00 A.M., in the southbound direction, optimum tolls ranged from $0.35 per mile in 2010 to $0.55 per mile in 2020.
  
In contrast, northbound toll rates in the morning peak period were found to be considerably lower, while p.m. peak period optimum toll rates were found to be higher, as might be expected. This is because of a slight directional imbalance in favor of the morning peak southbound and the afternoon peak in the northbound direction.

Under Alternative 6, the optimum toll in the morning peak hour was found to be $0.35 in 2010 southbound and $0.40 northbound. The higher northbound rate was actually dictated by peak northbound demands in the single lane sections in Broward County. In the afternoon peak hour, peak period tolls of $0.25 per mile would be sufficient in both travel directions.

By 2030, morning peak hour rates would reach as much as $0.90 in the northbound direction, driven heavily by excess demand in the northernmost portions of the project in Broward County. A more reasonable $0.55 per mile was found as optimum levels in the southbound direction.

In general, optimal tolls increased in proportion with overall travel demand, with lower rates during shoulder and midday periods and higher rates in the peak periods, as might be expected. Also, over time optimal toll rates needed to manage demand and optimize revenue are shown to increase as overall traffic levels continue to increase on I-95 along with congestion levels in the general purpose lanes.

Table 5-3 provides similar information for the six project alternatives under the HOV-3+ free operating scenario. This case reflects a condition where the regional definition of carpools is increased to vehicles with three or more occupants. In this case, all alternatives for all time periods would be operationally viable. Optimum tolls are shown to get relatively high, particularly in the morning peak period northbound, again driven by limited amounts of capacity, primarily in the Broward County section of the project. Optimal toll rates are shown for each alternative and each time period, following a similar pattern to that shown in Table 5-2.

Table 5-3
Optimum Per Mile Rates
HOV 3+ Operating Scenario

   
Northbound
Alternative
2010
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
0.25
0.50
0.35
0.15
0.15
0.35
0.35
0.35
2
0.15
0.50
0.35
0.05
0.15
0.35
0.35
0.35
3
0.30
0.60
0.35
0.15
0.15
0.35
0.35
0.30
4
0.10
0.30
0.15
0.05
0.15
0.30
0.30
0.35
5
0.25
0.45
0.35
0.15
0.15
0.30
0.30
0.35
6
0.25
0.40
0.30
0.20
0.20
0.30
0.30
0.35

   
Southbound
Alternative
2010
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
0.30
0.55
0.40
0.15
0.15
0.35
0.35
0.35
2
0.20
0.50
0.40
0.15
0.10
0.30
0.40
0.30
3
0.30
0.50
0.40
0.15
0.15
0.40
0.45
0.35
4
0.25
0.40
0.30
0.15
0.05
0.20
0.30
0.20
5
0.25
0.40
0.30
0.15
0.15
0.30
0.35
0.30
6
0.25
0.40
0.30
0.15
0.15
0.30
0.35
0.30

 

   
Northbound
Alternative
2020
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
0.40
0.90
0.55
0.35
0.35
0.50
0.60
0.45
2
0.30
0.90
0.45
0.20
0.35
0.50
0.45
0.40
3
0.45
1.00
0.60
0.45
0.35
0.45
0.50
0.35
4
0.25
0.80
0.40
0.10
0.30
0.35
0.40
0.35
5
0.35
0.80
0.55
0.35
0.30
0.35
0.40
0.35
6
0.35
0.80
0.50
0.30
0.30
0.35
0.40
0.35

Southbound
Alternative
2020
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
0.40
0.90
0.55
0.35
0.35
0.55
0.50
0.45
2
0.35
0.60
0.45
0.30
0.20
0.50
0.50
0.45
3
0.35
0.70
0.45
0.30
0.45
0.55
0.60
0.50
4
0.35
0.60
0.50
0.30
0.10
0.40
0.40
0.40
5
0.35
0.60
0.50
0.30
0.35
0.45
0.50
0.45
6
0.35
0.60
0.50
0.30
0.30
0.40
0.40
0.40

 

   
Northbound
Alternative
2030
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
0.50
1.20
0.80
0.40
0.40
0.80
0.80
0.55
2
0.50
1.20
0.90
0.25
0.30
0.55
0.55
0.45
3
0.70
1.20
0.90
0.50
0.35
0.60
0.60
0.50
4
0.45
1.20
0.90
0.25
0.35
0.60
0.60
0.50
5
0.55
1.20
0.80
0.40
0.30
0.45
0.50
0.40
6
0.55
1.20
0.60
0.35
0.35
0.45
0.50
0.40

Southbound
Alternative
2030
AM1
AM2
AM3
MD1
MD2
PM1
PM2
PM3
Time Periods
6:00-7:00 a.m.
7:00-8:00 a.m.
8:00-9:00 a.m.
9:00 a.m.-3:00 p.m.
9:00 a.m.-3:00 p.m.
3:00-4:30 p.m.
4:30-6:00 p.m.
6:00-7:00 p.m.
1
0.45
1.10
0.55
0.40
0.40
0.55
0.60
0.50
2
0.40
1.10
0.50
0.30
0.25
0.60
0.55
0.55
3
0.35
1.10
0.70
0.30
0.45
0.70
0.80
0.55
4
0.35
0.60
0.55
0.30
0.15
0.55
0.40
0.40
5
0.35
0.60
0.55
0.30
0.40
0.50
0.60
0.45
6
0.35
0.60
0.55
0.35
0.35
0.40
0.55
0.45

yellow bgIndicates time periods when Managed Lanes would accommodate HOV's only.
Per Mile Toll Rates Calculated in 2005 Dollars.

Those periods highlighted in yellow indicate instances where travel demand in the managed lanes on one or two sections of the project would be slightly over nominal free-flow operating capacity but still within the overall theoretical capacity of the lane. In such cases, the optimal toll was selected which generated the best relationship between revenue and traffic service. In all time periods without shading, the optimum toll rates shown would be sufficient to manage demand and ensure a minimum of Level of Service C operating conditions.

Alternative 1

A brief summary of the preliminary traffic and revenue analysis for Alternative 1 is presented in this section. As previously described in Chapter 1, Alternative 1 would involve converting the existing single HOV lanes in each direction to HOT operation. Also as noted previously, under an HOV-2+ definition of carpools, this scenario was found to be generally unviable for HOT operation, since little or no capacity would remain available after serving vehicles with two or more occupants.

The traffic analysis was conducted along the entire length of the facility and with assignment results produced for SOV, HOV-2, HOV-3+ and truck traffic components. Table 5-4 shows the traffic distribution by northbound versus southbound direction, at a representative link on I-95 for morning and afternoon peak hour conditions. Information in Table 5-4 reflects a location between 95th Street and 103rd Street at 2020 traffic assignment levels.

Table 5-4
Peak Hour Corridor Traffic Distribution - Alternative 1

Southbound
Northbound
Operating
Scenario
HOV-2+
Free
 
HOV-2+
Free
HOV-3+
Free
HOV-3+
Free
Peak
Hour
AM
 
PM
AM
PM
Vehicle
Type
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
General
Purpose
7.5
0.8
0.0
0.8
0.9

9.2
1.14
7.1
1.3
0.0
1.3
0.7

9.1
1.09
6.8
1.7
0.0
1.7
0.9

9.4
1.16
6.0
2.2
0.0
2.2
0.7

8.9
1.07
HOT/Toll
Lanes
0.0
1.8
0.1
1.9
0.0

1.9
0.90
0.0
1.8
0.1
1.9
0.0

1.9
0.90
0.9
0.3
0.2
0.5
0.0

1.4
0.67
0.9
0.4
0.1
0.5
0.0

1.4
0.67
HOV
Lane
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
Total
7.5
2.6
0.1
2.7
0.9

11.1
1.10
7.1
3.1
0.1
3.2
0.7

11.0
1.05
7.7
2.0
0.2
2.2
0.9

10.8
1.07
6.9
2.6
0.1
2.7
0.7

10.3
0.99
Percent
HOT/Toll
0.0%
69.2%
100.0%
70.4%
---
17.1%
0.0%
58.1%
100.0%
59.4%
---
17.3%
11.7%
15.0%
100.0%
22.7%
---
13.0%
13.0%
15.4%
100.0%
18.5%
---
13.6%
General
Purpose
7.4
0.7
0.0
0.7
0.9

9.0
1.12
6.5
1.1
0.0
1.1
0.8

8.4
1.04
6.6
1.6
0.0
1.6
0.9

9.1
1.13
5.2
1.9
0.0
1.9
0.8

7.9
0.98
HOT/Toll
Lanes
0.0
1.8
0.1
1.9
0.0

1.9
0.90
0.0
1.8
0.1
1.9
0.0

1.9
0.90
0.9
0.3
0.1
0.4
0.0

1.3
0.62
0.9
0.4
0.1
0.5
0.0

1.4
0.67
HOV
Lane
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
Total
7.4
2.5
0.1
2.6
0.9

10.9
1.08
6.5
2.9
0.1
3.0
0.8

10.3
1.01
7.5
1.9
0.1
2.0
0.9

10.4
1.04
6.1
2.3
0.1
2.4
0.8

9.3
0.93
Percent
HOT/Toll
0.0%
72.0%
100.0%
73.1%
---
17.4%
0.0%
62.1%
100.0%
63.3%
---
18.4%
12.0%
15.8%
100.0%
20.0%
---
12.5%
14.8%
17.4%
100.0%
20.8%
---
15.1%

The table shows traffic assigned to general purpose lanes, the HOT/toll lanes and, where applicable, separate HOV lanes (not applicable under Alternative 1). The table also shows the proportion of total traffic for each category assigned to the HOT/toll lanes, as well as the estimated volume/ capacity ratio on general purpose lanes and, where appropriate, in the HOT lanes.

As shown in Table 5-4, there is sufficient demand in the HOT lanes from HOV traffic alone under the HOV-2+ operating scenario. For example, in the a.m. peak hour, southbound direction, an estimated 1,800 vehicles with two or more occupants were assigned to the HOT lane, and this represented just 69.2 percent of the total.

After adding in vehicles with three or more occupants, total traffic demand in the HOT lane is estimated at 1,900 per hour, above the desired free-flow operating capacity of the lane, even without permitting SOV traffic to buy in.

Similar relationships are shown under the HOV-2 scenario in the northbound direction. In this case, the HOV lane is able to accommodate between 62 and 72 percent of total HOV demand, even without allowing any SOV traffic into the lanes.

It is also noted that under this condition, the general purpose lanes are, in most cases, over capacity. The lower half of Table 5-4 shows the estimated distribution of traffic at 2020 levels under an HOV-3+ scenario. Since there are significantly fewer vehicles with three or more occupants than two or more occupants, a significant amount of capacity is opened up in the HOT lanes for sale to non-HOV traffic. Under this operating condition, both SOV and HOV-2 vehicles are assumed to be required to pay a toll.

In the a.m. peak southbound direction, for example, an estimated 900 SOVs would be expected to buy into the single managed lane at this location, reflecting about 11.7 percent of the total estimated SOV demand in the southbound direction. An additional 300 vehicles with two occupants, about 15 percent of the total, would be expected to buy in. All 200 of the vehicles with three or more occupants would be assumed to use the lane (toll free) bringing total traffic in the HOT lane to about 1,400 per hour. This reflects a volume/capacity ratio of 0.67 and would be expected to provide for free-flow operating conditions.

At the same time, total demand in the four general purpose lanes would be estimated at about 9,400 vehicles. This results in a volume/capacity ratio at this section of 1.16, which would equate to extremely congested conditions.

figure 5-2 average weekday traffic volumes alternative 1
Text Box: Figure 5-2  Average Weekday Traffic Volumes Alternative 1 – HOV+3 Free

Similar relationships are shown for both the morning and evening peak hour conditions in both the northbound and southbound direction. Total traffic in the HOT lane including SOV and HOV traffic, is estimated to range between 1,300 and 1,400 in both travel directions, although significantly different optimum toll rates would be used by travel direction. Finally, it is important to recognize that all trucks are assumed to be in the general purpose lanes and are assumed to be banned from the HOT lanes.

Figure 5-2 shows estimated total weekday toll traffic in the managed lanes on each mainline section between I-595 and I-395. Under Alternative 1, only one HOT lane is available in each direction over the entire length of the facility. The yellow shading at the southern end of the project indicates a short elevated section, which would still have only a single lane each way.

Volumes are shown in thousands, with the blue figures representing estimated traffic at 2010 levels and the red figures representing estimated traffic at 2030 levels. This is total weekday traffic; volumes would vary from hour to hour by travel direction, based on toll rate and amount of available capacity to sell.

In 2010, estimated weekday toll users of the managed lanes would range from about 9,500 at the north end of the facility to a peak of about 28,000 in the vicinity of 95th Street. At the south end of the project, total toll paying traffic would be expected to drop to 19,200.

It should be kept in mind that only toll paying traffic is shown in Figure 5-2. In addition, vehicles with three or more occupants would also use the facility toll-free. Figure 5-2 represents the HOV-3 operating condition; Alternative 1 was not considered to be operationally viable under the HOV-2+ case.

Preliminary estimates of daily toll transactions and annual toll revenue are summarized for Alternative 1 in Table 5-5. In this case, figures are shown only for the HOV-3+ free condition. Total tolled vehicles estimated to use the managed lanes are shown for each of the seven analysis periods, by travel direction. Also shown is the average toll rate assessed to those vehicles and the result in revenue for each period of the day.

For example, in 2010, under the HOV-3+ condition, during the morning peak hour (7:00 to 8:00 A.M.) an estimated 1,810 tolled vehicles would use the managed lanes in the northbound direction, paying an average toll of $4.91. In the southbound direction, toll traffic is estimated at 1,760, paying an average toll of $5.03. It is noted that the tolled vehicles represent the total toll traffic entering the managed lanes at all points of entry, not necessarily the volume at any one location along the roadway. It is also noted that traffic appears to be heaviest in the midday hours, although the toll rate charged in those hours is considerably lower. It should be kept in mind that the midday period reflects six hours of traffic, indicating a total usage in the range of 1,000 vehicles per hour, on average, at much lower toll rates.

Table 5-5
Estimated Toll Transactions and Revenue - Alternative 1

Year 2010 - HOV2+Free
 
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled
0
0
0
0
0
0
0

0
0
0
0
Average Toll
-
-
-
-
-
-
-
Revenue
$0
0
0
0
0
0
0

$0
Tolled
0
0
0
0
0
0
0

0
0
0
0
Average Toll
-
-
-
-
-
-
-
Total
Daily
Revenue
$0
0
0
0
0
0
0

$0
Annual
Average
Total Tolled Transactions
0
0
0
0
0
0
0

0
0
0
Revenue
$0
0
0
0
0
0
0

$0

Year 2010 - HOV3+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled
1,050
1,810
1,520
6,500
2,210
2,300
1,170

16,560
4,654,000
12,751
25,392
Average Toll
$2.53
4.91
3.36
1.39
3.23
3.20
3.18
27132
Revenue
$2,660
8,880
5,110
9,040
7,130
7,350
3,720

$43,890
Tolled
1,190
1,760
1,750
6,520
2,010
2,180
1,010

16,420
4,614,000
12,641
Average Toll
$2.50
5.03
3.52
1.34
3.43
3.44
3.59
Total
Daily
Revenue
$2,980
8,860
6,160
8,760
6,900
7,510
3,630

$44,800
Annual
Average
Total Tolled Transactions
2,240
3,570
3,270
13,020
4,220
4,480
2,180

32,980
9,268,000
25,392
Revenue
$5,640
17,740
11,270
17,800
14,030
14,860
7,350

$88,690
$24,923,000
$68,282

 

Year 2020 - HOV2+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled
0
0
0
0
0
0
0

0
0
0
0
Average Toll
-
-
-
-
-
-
-
Revenue
$0
0
0
0
0
0
0

$0
Tolled
0
0
0
0
0
0
0

0
0
0
0
Average Toll
-
-
-
-
-
-
-
Total
Daily
Revenue
$0
0
0
0
0
0
0

$0
Annual
Average
Total Tolled Transactions
0
0
0
0
0
0
0

0
0
0
Revenue
$0
0
0
0
0
0
0

$0

Year 2020 - HOV3+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled
1,350
1,970
1,720
7,870
2,340
2,190
1,500

18,940
5,322,000
14,581
28,540
Average Toll
$4.09
8.79
5.51
3.24
4.86
5.67
4.27
Revenue
$5,520
17,310
9,470
25,510
11,370
12,410
6,410

$88,000
Tolled
1,400
1,410
1,750
7,730
1,980
2,400
1,460

18,130
5,095,000
13,959
Average Toll
$3.50
8.71
5.02
3.14
5.73
5.15
4.60
Total
Daily
Revenue
$4,900
12,280
8,780
24,270
11,350
12,360
6,710

$80,650
Annual
Average
Total Tolled Transactions
2,750
3,380
3,470
15,600
4,320
4,590
2,960

37,070
10,417,000
28,540
Revenue
$10,420
29,590
18,250
49,780
22,720
24,770
13,120

$168,650
$47,392,000
$129,841

 

Year 2030 - HOV2+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled
0
0
0
0
0
0
0

0
0
0
0
Average Toll
-
-
-
-
-
-
-
Revenue
$0
0
0
0
0
0
0

$0
Tolled
0
0
0
0
0
0
0

$0
0
0
0
Average Toll
-
-
-
-
-
-
-
Total
Daily
Revenue
$0
0
0
0
0
0
0

$0
Annual
Average
Total Tolled Transactions
0
0
0
0
0
0
0

0
0
0
Revenue
$0
0
0
0
0
0
0

$0
0
0

Year 2030 - HOV3+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled
1,870
2,720
2,110
7,870
1,930
2,060
1,590

20,150
5,662,000
15,512
30,087
Average Toll
$5.18
11.58
7.66
3.81
7.98
7.73
5.42
Revenue
$9,690
31,490
16,170
29,950
15,410
15,920
8,620

$127,250
Tolled
1,360
1,410
1,980
7,520
2,570
2,500
1,590

18,930
5,320,000
14,575
Average Toll
$3.99
10.35
5.07
3.73
5.70
6.23
5.19
Total
Daily
Revenue
$5,430
14,640
10,030
28,040
14,660
15,570
8,250

$96,620
Annual
Average
Total Tolled Transactions
3,230
4,130
4,090
15,390
4,500
4,560
3,180

39,080
10,982,000
30,088
Revenue
$15,120
46,130
26,200
57,990
30,070
31,490
16,870

$223,870
$62,910,000
$172,356

Note: Revenues do not consider effects of ramp-up of demand.
All toll revenue calculations in 2005 dollars.

In 2010, it is estimated that a total of just under 33,000 vehicles per typical weekday would use the managed lanes. This would generate average weekday revenue estimated at $88,690 per weekday. This is raised to annual levels by assuming weekend traffic is 25 percent of estimated weekday traffic.

Weekday revenue is estimated at $24,923,000 in 2010. This would be expected to increase to more than $47 million by 2020 and to almost $63 million per year by 2030. It is important to note, however, that relatively small volumes of toll traffic would be estimated to use the managed lanes and, in peak periods, relatively high toll rates would be required to manage demand. For example, by the year 2030, the average toll would be in the range of $10.00 to $12.00 during the morning peak depending on direction.

Alternative 2

Under Alternative 2, south of the Golden Glades Interchange, two reversible HOT lanes would be provided, both assumed to operate in the major direction of travel. In the morning peak, the two lanes would operate southbound, but no HOV or HOT lane would be provided in the northbound direction. In the afternoon peak period, the lanes would operate northbound, with no HOV or HOT lane operated in the southbound direction. North of the golden Glades Interchange, a single HOT lane in each direction would continue to be provided.

Table 5-6 shows the peak hour corridor traffic distribution for Alternative 2. In this case, with the a.m. and p.m. peak hour, usage of the HOT lanes was assumed to be limited to HOV traffic only under the HOV-2+ scenario. Total HOV demand in the HOT lane is estimated at 3,200 vehicles, close to the optimal capacity of free-flow conditions during the morning peak. No traffic is shown in the HOT lanes in the northbound direction a.m. peak since no such lanes would exist. In the afternoon peak, HOV demand is estimated at 3,400, again close to the free-flow operating capacity of the two-lane reversible HOT roadway. Hence, this scenario is not viable under HOV-2+ free operations.

Under the HOV-3+ free case the scenario does operate, with an estimated 3,100 vehicles per hour in this section of the reversible roadway in the a.m. peak southbound, and about 2,800 vehicles per hour in the p.m. peak northbound. It is important to note that operating conditions in the general purpose lanes actually deteriorate in the minor traffic direction by virtue of the elimination of the HOV lane in the minor direction under this Alternative.

About one-fourth of total SOV traffic in the southbound morning peak and northbound evening peak would be assumed to buy into the two available lanes. A similar proportion of vehicles with two or more occupants would buy in, while virtually 100 percent of vehicles with three or more occupants would be accommodated toll-free. In total, the managed lanes would accommodate about one-fourth of the total demand, with three-fourths still using the four toll-free general purpose lanes.

Table 5-6
Peak Hour Corridor Traffic Distribution - Alternative 2

Southbound
Northbound
Operating
Scenario
HOV-2+ Free
HOV-2+ Free
HOV-3+ Free
HOV-3+ Free
Peak
Hour
AM
PM
AM
PM
Vehicle
Type
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
SOV
HOV2
HOV3+
HOV
Trucks
 
Total
Vol./Cap. Ratio
General
Purpose
8.0
0.1
0.0
0.1
0.9

9.0
1.12
6.2
2.4
0.1
2.5
0.7

9.4
1.12
6.7
1.6
0.0
1.6
0.9

9.2
1.14
6.3
2.3
0.1
2.4
0.7

9.4
1.12
HOT/Toll
Lanes
0.0
3.0
0.2
3.2
0.0

3.2
---
0.0
0.0
0.0
0.0
0.0

0.0
---
2.3
0.6
0.2
0.8
0.0

3.1
---
0.0
0.0
0.0
0.0
0.0

0.0
---
HOV
Lane
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
Total
8.0
3.1
0.2
3.3
0.9

12.2
1.45
6.2
2.4
0.1
2.5
0.7

9.4
1.12
9.0
2.2
0.2
2.4
0.9

12.3
1.46
6.3
2.3
0.1
2.4
0.7

9.4
1.12
Percent
HOT/Toll
0.0%
96.8%
100.0%
97.0%
---
26.2%
0.0%
0.0%
0.0%
0.0%
---
0.0%
25.6%
27.3%
100.0%
33.3%
---
 
25.2%
0.0%
0.0%
0.0%
0.0%
---
0.0%
General
Purpose
7.0
1.7
0.1
1.8
0.8

9.6
1.16
7.5
0.1
0.0
0.1
0.8

8.4
0.93
7.0
1.6
0.1
1.7
0.8

9.5
1.15
6.0
2.0
0.0
2.0
0.8

8.8
1.08
HOT/Toll
Lanes
0.0
0.0
0.0
0.0
0.0

0.0
1.04
0.0
3.3
0.1
3.4
0.0

3.4
0.81
0.0
0.0
0.0
0.0
0.0

0.0
---
2.0
0.7
0.1
0.8
0.0

2.8
0.67
HOV
Lane
0.0
0.0
0.0
0.0
0.0

0.0
1.07
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
0.0
0.0
0.0
0.0
0.0

0.0
---
Total
7.0
1.7
0.1
1.8
0.8

9.6
0.81
7.5
3.4
0.1
3.5
0.8

11.8
0.97
7.0
1.6
0.1
1.7
0.8

9.5
0.80
8.0
2.7
0.1
2.8
0.8

11.6
0.95
Percent
HOT/Toll
0.0%
0.0%
0.0%
0.0%
---
0.0%
---
0.0%
97.1%
100.0%
97.1%
---
28.8%
0.0%
0.0%
0.0%
0.0%
---
 
0.0%
---
25.0%
25.9%
100.0%
28.6%
---
24.1%

figure 5-3 average weekday traffic volumes alternative 2
Text Box: Figure 5-3  Average Weekday Traffic Volumes Alternative 2 – HOV3+ Free

Figure 5-3 shows estimated weekday traffic for Alternative 2 under the HOV-3+ condition. Again, volumes are shown at 2010 and 2030 levels, in blue and red, respectively. Total weekday volumes range from 12,200 vehicles per day at the south end of the project to a peak of just under 20,000 vehicles per day in the vicinity of 95th Street in 2010. Peak volumes reached more than 26,000 by 2030, although revenue increased significantly because of the need for higher toll rates.

It should be kept in mind that south of the Golden Glades Interchange (Florida’s Turnpike), two reversible lanes are provided. North of this location, only a single HOT lane in each travel direction would be operated.

Table 5-7 shows estimated daily toll transactions and annual toll revenue for Alternative 2. Estimates are shown for 2010, 2020 and 2030 levels. Under Alternative 2, assuming a continuation of the HOV-2+ definition of carpools, the HOT lanes would have only very limited capacity to sell during peak hours in 2010, and would be available for non-HOV traffic only in off-peak and shoulder hours in years 2020 and 2030. This significantly reduced traffic and revenue potential under the HOV-2+ free scenario as shown in Table 5-7.

For example, in 2010, annual toll transactions are estimated at about 4.7 million, expected to produce about $8.2 million in revenue with an HOV-2 definition of carpools. If free vehicles are limited to those with three or more occupants, toll transactions would increase to an estimated 10.5 million and revenue to just under $25 million in the same year.
 
By 2030, revenue potential under an HOV-2+ scenario is reduced to just under $4.2 million, since the lanes would primarily be available during midday hours only for non-HOV traffic. However, if only vehicles with three or more occupants are granted toll-free access, revenue potential would increase to over $63 million, a 12-fold increase in revenue potential between the two operating scenarios.

Under the HOV-3+ condition, average peak hour tolls in 2010 were generally in the range of $3.28 to $4.52 depending on hour and direction of travel. Lower average toll rates were found in shoulder and off-peak hours, of course. By 2030, with increasing levels of demand and increasing congestion levels in the general purpose lanes, a.m. peak hour toll rates reached as high as $10.00 for the average trip, although this was heavily driven by severe capacity limitations on the single HOT lane sections of the project, mostly in Broward County. The p.m. peak hour tolls were somewhat more reasonable, generally in the range of $5.00 - $6.00.

Table 5-7
Estimated Toll Transactions and Revenue - Alternative 2

Year 2010 - HOV2+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled Trips
550
560
330
4,500
1,080
790
840

8,650
2,431,000
6,660
12,857
Average Toll
$0.91
4.11
3.12
0.80
2.30
2.68
1.73
Revenue
$500
2,300
1,030
3,580
2,480
2,120
1,200

$13,210
Tolled Trips
1,160
880
1,240
3,370
490
460
450

8,050
2,262,000
6,197
Average Toll
$0
5.78
3.87
0.57
1.39
1.93
1.38
Total
Daily
Revenue
$2,070
5,090
4800
1910
680
890
620

$16,060
Annual
Average
Total Tolled Trips
1,710
1,440
1,570
7,870
1,570
1,250
1,290

16,700
4,693,000
12,858
Revenue
$2,570
7,390
5830
5490
3160
3010
1820

29,270
8,225,000
22,534

Year 2010 - HOV3+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled Trips
1,030
1,750
980
10,240
3,010
3,110
1,460

21,580
6,064,000
16,614
28,715
Average Toll
$1.26
4.07
3.12
0.85
3.26
3.28
3.26
31,310
Revenue
$1,300
7,120
3,060
8,680
9,820
10,200
4,760

$44,940
Tolled Trips
1,770
3,220
2,560
4,640
1,420
1,230
880

15,720
4,417,000
12,101
Average Toll
$2.63
4.52
3.65
1.03
2.64
3.67
2.68
Total
Daily
Revenue
$4,650
14,540
9,350
4,790
3,750
4,520
2,360

$43,960
Annual
Average
Total Tolled Trips
2,800
4,970
3,540
14,880
4,430
4,340
2,340

37,300
10,482,000
28,718
Revenue
$5,950
21,660
12,410
13,470
13,570
14,720
7,120

$88,900
$24,982,000
$68,444

 

Year 2020 - HOV2+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled Trips
360
0
590
3,200
0
0
0

4,150
1,166,000
3,195
7,274
Average Toll
$2.86
--
3.75
1.19
--
--
--
Revenue
$1,030
0
2,210
3,810
0
0
0

$7,050
Tolled Trips
1,050
0
900
2,580
430
0
340

5,300
1,489,000
4,079
Average Toll
$3.50
--
6.21
1.85
2.74
--
2.79
Total
Daily
Revenue
$3,670
0
5,590
4,770
1180
0
950

$16,160
Annual
Average
Total Tolled Trips
1,410
0
1,490
5,780
430
0
340

9,450
2,656,000
7,277
Revenue
$4,700
0
7800
8580
1180
0
950

$23,210
$6,522,000
$17,868

Year 2020 - HOV3+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions
Tolled Trips
1,170
1,970
1,810
9,400
3,520
4,140
2,470

24,480
6,879,000
18,847
35,077
Average Toll
$2.62
7.69
3.73
2.79
4.87
4.31
3.76
Revenue
$3,070
15,140
6,750
26,180
17,140
17,830
9,280

$95,390
Tolled Trips
2,410
3,820
3,210
7,150
1,540
1,890
1,060

21,080
5,924,000
16,230
Average Toll
$3.16
5.44
4.15
2.09
4.51
4.43
4.11
Total
Daily
Revenue
$7,610
20,770
13,330
14,970
6,950
8,380
4,360

$76,370
Annual
Average
Total Tolled Trips
3,580
5,790
5,020
16,550
5,060
6,030
3,530

45,560
12,803,000
35,077
Revenue
$10,680
35,910
20,080
41,150
24,090
26,210
13,640

$171,760
$48,266,000
$132,236

 

Year 2030 - HOV2+Free
Northbound
Southbound
Both Directions
Time Period
AM1 (6-7)
AM2 (7-8)
AM3 (8-9)
MD (9-15)
PM1 (15-16:30)
PM2 (16:30-18)
PM3 (18-19)
Weekday Total
Total annual
Daily average
Both directions