Faced with ever increasing demand and congestion on its highways and arterials, Miami-Dade County and state transportation agencies are assessing establishing a major intermodal hub near the Miami International Airport an idea that started while considering options to improve the existing airport. PB has been and continues to be involved in providing travel demand forecasting and traffic simulation services, among others, for the agencies involved in this new intermodal facility and other transportation projects in the area.

Evolution of an Idea

Traffic congestion in Miami-Dade County has been a concern for citizens and transportation agencies for several years. It continues to worsen because of the county's continuous and rapid increase in population, particularly in its suburban areas.

Concerns about airport congestion go hand-in-hand with those related to roadway congestion. Located at the southeastern tip of the U.S, Miami has been called the "gateway to South America." Flights to and from South America and the Caribbean have increased tremendously in the past two decades, taking a toll on the existing facilities at Miami International Airport.

The Miami-Dade Aviation Department has developed a $6 billion capital improvement program for the next twenty years to address the airport congestion concerns. This program focuses on two major goals:

• Improve access to the airport
  
  
• Increase airport's capacity to meet the ever-increasing demand.

Meeting these two goals was soon found to be a challenge, however, because increased airport capacity would also increase travel to and from the airport, taxing the already deficient roadway system. In addition, a physical expansion of the airport was quickly found to be cost prohibitive because the airport is surrounded by major freeways, arterials, office buildings and airport related facilities.

A solution was to create a remote facility where some of the airport's functions could take place, including ticketing counters, domestic baggage claim area, parking, and passenger drop-off and pick up. State and county transportation agencies soon realized, however, that such a facility also offered a tremendous opportunity to serve as a hub for rail and bus service, making it a major intermodal center (MIC).

Overview of the Major Intermodal Center

Under the auspices of the Florida Department of Transportation (FDOT), the MIC is being developed as two major facilities:

• Rental Car Facility. This facility will consolidate all of the major rental car agencies scattered throughout the area into a 7,500-space parking facility that provides full rental car services. It will include customer service counters, administrative offices, return car lot, employee parking, and service maintenance areas.
  
  
• The MIC Core. Adjacent to the rental car facility, the MIC core will house limited passenger ticketing and baggage claim, as well as other non-airport land uses, such has hotel, shopping, and employment. Rail and bus stations as well as taxi and cruise bus staging areas are also incorporated into the MIC in order to provide connections to the local and regional transit system.

The MIC core, rental car facility and Miami International Airport will be linked to each other by a people mover, and to the rest of the region by:

• Tri-Rail, a heavy rail line between Miami-Dade, Broward, and Palm Beach Counties
  
  
• Metrorail, a light rail line that provides service in selected areas of Miami-Dade County
  
  
• Metrobus, which provides bus service throughout Miami-Dade County
  
  
• Private bus service to the seaport for cruise ship passengers.

Both the rental car facility and the MIC core are still under development. Their specific uses, densities, access modes and locations are being refined.

Figure 1: Multimodal Projects

With the proposed improvements at the airport and the implementation of the MIC core and rental car facilities, vehicular access to the area has become crucial. The two main highways that provide access to the area were transferred recently to the Miami-Dade Expressway Authority and were programmed for improvements. Included in the program is construction of a new six-lane highway that will link the two roadways and provide direct access to the airport and the two new facilities. The ultimate roadway layout under consideration is shown on Figure 1. This figure also shows the location of several other projects in the area that PB is involved in. Our role in these projects is described briefly below.

PB's Role in Several Area Projects

• A proposed new highway that will link SR 836 to SR 112 between Miami International Airport and the MIC
  
  
• Improvements to SR 112.

These services included developing intelligent transportation system (ITS) measures, travel demand forecasts and traffic simulation of the proposed highway and adjacent and intersecting roadways.

People Mover Design. As a subconsultant, we are participating in the design of the people mover line, the connector that will link the airport to the future rental car facility/MIC core.

Travel Demand Forecast: Daily and Peak Hour

Faced with a complex roadway network, land-uses with unique characteristics, an already congested area, and presence of a multitude of travel modes, we used our basic knowledge of traffic operations to face the technical challenges associated with simulating traffic operations for these assignments.

We use a nested logit TRANPLAN-based model developed by FDOT to develop the projected future highway and transit trips for the opening (2005) and design (2020) years of the proposed roadway and rail improvements. Rail and bus services to be provided at the MIC were coded as part of the transit network. The daily volumes generated by the model needed to be evaluated carefully for reasonableness because of the regional nature of the model and the complexity of the area's proposed roadway networks.

Comparison of Model Results with Available Data. The first reasonableness check consisted of comparing the model results with available data. A master plan developed for the airport as part the capital improvement plan contained projected increases in passenger and cargo activities and an assessment of existing conditions. The total number of trips produced by the model was compared to the projections based on information contained in the master plan report.

The Institute of Transportation Engineers Trip Generation Manual (6th Edition) was used to compare total daily trips generated by the MIC core based on land-use types and densities provided by FDOT. Trips generated by the rental car facility were compared to standard parking facility generation rates and results of a limited survey of traffic patterns at the major rental car facilities. Both the airport and the MIC core/rental car facility were coded as special generators in the regional model.

Proper Assignment of Trips. The main purpose of the proposed improvements at the airport was to provide separate roadways for service vehicles and passenger-related traffic. Although the proposed roadway network was coded carefully in the travel demand forecast model with proper facility types, area types, number of lanes, and turn penalty/prohibition, the model assigned most of the trips to the shortest path to and from the traffic analysis zone representing the airport and the MIC core/rental car facility. Due to its regional nature, the model was not sensitive enough to assign the proper trips generated by specific groups to the various available paths. The travel demand model combined all of the trips generated by the airport into two groups: low occupancy vehicles and high occupancy vehicles. Some of the trips related to the intermodal aspect of the MIC, such as bus trips related to cruise ship passengers. Furthermore, although the trips were reported by purpose (home-based, non-home based, taxis, etc.), the resulting assignment did not reflect the expected trip distribution.

Refinements to the model were not included in our scope of work, so we developed a post-process methodology that was based on the observation that each of the various modes (taxis, cruise buses, hotel shuttles, etc.) had specific daily trip, distribution and assignment characteristics. The total daily trips produced by the model were, therefore, divided into ten different modes to represent the various groups using the intermodal facilities:

Based on daily vehicle classification data, the percentages of total traffic were developed for categories 1 through 9. The percentage of passenger vehicles reported from the classification counts included airport passengers, employees, and through trips. Due to the level of congestion in the area and the lack of a direct connection between SR 836 south of the airport and SR 112 north and east of the airport, local drivers used the airport roadways as an alternate route. The number of through trips was, therefore, estimated by comparing the number of entering and exiting vehicles following the path used by vehicles that were avoiding congestion on the adjacent roadways. The remaining trips, air passengers and employees were determined by using the proportion of work versus non-work trips generated by the travel demand forecast model.

A separate trip distribution model was then developed for each of the groups identified previously. It was based on the proposed function for each of the roadways identified in the proposed improvement plans. Total morning and peak hour trips on each of the roadways within the airport were assigned based on that distribution and using the daily counts and passenger enplaning and deplaning daily distribution. Trip distribution and assignment for the MIC core/rental car facility were refined based on cardinal distribution of existing traffic.

Traffic Simulation

The shear magnitude of the combined roadway network proposed for the MIC core/rental car facility and Central Boulevard at the airport tested the limits of the CORSIM model used to analyze existing and future traffic operations in the area. This model allows for analysis of integrated roadway networks, including arterials, highways, signalized intersections and, with some minor modifications, toll plazas and at-grade rail crossings. Basic input into the simulation model included roadway geometry, design speeds, signal phasing and timings, traffic counts and existing operating speeds from actual field data.

Setting up Roadway Files. The first challenge came when setting up the roadway files for developing a calibrated model based on existing traffic patterns. Although certain roads, such as Central Boulevard, operated as arterials with a posted speed of 20 km (35 miles) per hour, the actual geometry resembled that of a highway with on and off-ramps. The arterial simulation module of CORSIM model (NETSIM) would have been the correct tool to use for these roadways; however, the on and off-ramps would have been modeled as left and right turns with the delay associated with these movements. The highway module (FRESIM) was used, therefore, to properly replicate existing traffic operations on certain arterials within the study area.

Coding Toll Plaza Operations. Another challenge was developing a methodology to accurately code toll plaza operations along SR 112, the highway that provided access from the north and east of the county to the airport and the MIC core/rental car facility. One of the limitations of the CORSIM model is its inability to simulate traffic operations at toll plazas. We developed a methodology that:

• Treated the toll plaza as a signalized intersection without cross streets
  
  
• Coded the express lanes with electronic toll collection system as highway links
  
  
• Calculated separate signal timings to reflect the amount of time spent by each vehicle when doing a manual transaction or dropping exact change in the hopper.

Figure 2: Toll Simulation

The times were set to replicate the processing time for each lane type. The total cycle length was developed in order to produce the field observed throughput per type of lanes at this particular plaza. Instead of relying on general values, this approach allowed us to take into account physical features of the plaza that contribute to its overall operation. The toll plaza animation file is shown on Figure 2.

Conclusion

The traffic simulation models developed for each of the projects were eventually merged into a single model that enabled our clients to look at the impact of the various modes of transportation on the mobility of the area as a whole and to develop appropriate solutions. This model was being used to evaluate various scenarios, including changes in ridership, geometry, and travel patterns of various transportation modes that impact the immediate study area.

As a result of the events that occurred on September 11, 2001, the Miami-Dade Aviation Department has reduced the funds allocated for construction of the MIC-MIA connector due to the temporary reduction in air travel. Another rail connection between the airport and the MIC is being evaluated, however.

For more information on the MIC, visit its Web site at www.micdot.com. To find out more about the MIC-MIA connector or the Central Boulevard Widening projects, contact Rich Lear (lear@pbworld.com), project manager or John Wyatt (wyatt@pbworld.com), lead engineer at 1-305-261-4785. To find out more about the Earlington Heights EIS, contact Larry Foutz (foutz@pbworld.com), project manager at 305-261-4785.