Bellevue Downtown PRT Planning Study

Scope, Objectives and Context

During the Winter Quarter of 1994, a class project was conducted at the University of Washington that was focused on learning how to design a Personal Rapid Transit network for a suburban downtown area. The instructor for the class was Professor J. Schneider of the Transportation Systems Program in the UW's Department of Civil Engineering. It was the third such class project - the previous two were focused on designing PRT networks for the SeaTac airport area (1990) and the City of Tukwila (1992).

Downtown Bellevue, Washington, is one of several major activity centers located in the greater Central Puget Sound Region in the State of Washington. It currently contains about 28,000 jobs and is forecast to grow to about 33,000 by 2005. It is also expected to contain about 6,500 residents in 2005. This downtown area is the site of the region's most successful shopping center, called Bellevue Square, which attracts about 50,000 shopping trips each day and provides about 12,000 free parking spaces in adjacent multistory parking structures.

The downtown is laid out in a superblock form, has wide, multilane streets and copious off-street parking, most of which is no longer free. Its traffic congestion is very bad at times and its air quality is substandard fairly often. There is considerable room for new development as many sites are now used by low-rise or one-story buildings. Several high-rise office towers have been constructed in the past decade and the skyline is now visible from many parts of the region. The 1993 assessed value of this downtown was approximately $1.27 billion. A new convention center has recently been opened that served about 175,000 visitors in 1994.

A vicinity map is provided that shows the geography of the area used in this study. Downtown Bellevue is located about 6 miles east of downtown Seattle on the east side of Lake Washington. It is in the center of a large suburban area that will soon contain about 500,000 residents and has several large employers, including Microsoft and several other software companies. Household incomes are the highest in the state and auto ownership and use rates are also very high. The study area is bounded by three major freeways on the north (SR 520), the east (Interstate 405) and the south (Interstate 90). I-405 has been suggested as the most likely route for a high speed train route that would link Vancouver, B.C. with Portland, Oregon at some time in the future. If such an alignment were selected, a major station would probably be sited adjacent to downtown Bellevue. It is also possible that an urban rail station might provide 2-3 urban rail stations in the downtown area at some future time.

Given this context, the students were asked to design a PRT circulator system for the downtown that would serve two transportation functions. One would be to provide a means to circulate within the downtown without using an auto (as is commonly done now). The other would be to allow people to commute to the downtown on a bus or rail system and transfer to the PRT to reach their destination in the downtown. In both cases, the overall objective is to reduce the use of the auto, thereby reducing congestion and air pollution.

Other objectives considered were in the land use category. The widely accepted goal of making existing activity centers more dense, thus reducing the tendency toward urban sprawl in the region, is one of these. Currently, the growth of downtown Bellevue is limited by its traffic congestion and air pollution levels. If it is to become more densely developed, then some non-auto means of dealing with increased traffic (goods and people) will have to be found. Increased density is desired by both the local landowners and the City which would like to received increased tax revenues from the downtown. It is also desired by persons who wish to reduce the negative effects of urban sprawl in the region and to encourage the evolution of more pedestrian-friendly downtown areas.

The students were given a budget of $200 million dollars and were constrained to designing a PRT network no longer than 12 miles. A system cost estimate of about $16 million per one-way mile was used that includes all stations, vehicles and a storage/maintenance facility.

PRT Network Design Process

A computer program, called Circulator-CAD was provided to assist the conduct of this design process. C-CAD requires a digital database of the downtown an input. The students initially digitized maps of the downtown to obtain the necessary building outlines, street boundaries and walk paths that describe the physical properties of the downtown. Forecasts of the locations of both people and jobs in 2005 were also prepared, using data provided by the City and the Puget Sound Regional Council. A map showing this demand data , by location and size is provided. Two maps of portions of the digital database that show the level of detail used in this work are also provided. It is centered on the Bellevue Square area which is the largest traffic generator in the downtown at the present time. It shows a portion of a PRT route, a PRT station and various demand locations in the surrounding buildings.

The PRT network design activities began with the identification of all feasible and desirable PRT station locations in the downtown. Some 45 such locations were identified. Then, the C-CAD program was used to find a "best" subset of these candidate station locations. "Best" means providing a high level of walking access for the spatially distributed demand (i.e. people and jobs). This search program will find those station locations that are as close as possible to the distributed demand, thereby reducing the access walk time/distance for the system users to a minimum. This is a consumer-oriented approach that is aimed at making the PRT system as accessible to the potential users as possible. The result of these computations was a set of 29 stations. Both larger and smaller numbers of stations were also examined but the set of 29 were determined to be preferred. The next task was to connect these stations with one-way PRT guideway segments, using a series of connected loops and trying to avoid interchanges (overpasses and underpasses) to the extent possible. A map of the PRT network designed by one of the three teams is provided. It shows the PRT guideway alignment (routes), direction of travel and the locations of stations devised by the team.

Analysis and Evaluation

Once an initial network design has been developed, it is necessary to analyze and evaluate its strengths and weaknesses. C-CAD provides several analytical and mapping capabilities that allow such an evaluation process to be conducted interactively, quickly and effectively. A sample of some of the results obtained in this analysis is provided. An overview display provides an assessment of overall performance and is generally examined first. It will tell the planner if the system performance is within the "ballpark" or not and indicate how much work remains to be done before a satisfactory result is achieved. Another display shows the service areas of each station. It provides a means to examine the likely level of patronage that might be experienced by each station so that weak stations can be deleted or moved and relief can be provided to overloaded stations in subsequent network design iterations.

Another type of display shows the locations that can be reached from a particular station in a given number of minutes of walk + ride time. The first map shows the locations that can be reached from the Bellevue City Hall within 5 minutes . A second map shows the locations that can be reached within 10 minutes . It also shows those demand locations that are more than 10 minutes away from the City Hall. Several other maps displays can be obtained with the C-CAD program that show travel times, covered and uncovered demand locations within variable time intervals and travel time contour maps. Examples are not included at this time, due to space limitations. These same data are also available in printed or digital form.

Taken together, these maps and data files provide the planner with the information (clues) needed to modify the initial design in ways that will improve its performance. Design modifications can include adding, deleting or moving stations or adding, deleting, moving or changing the direction of PRT guideway segments. After two or three iterations, this interactive graphic design process will generally produce satisfactory results. Such results can then be used as an input to a PRT operations simulation program. Such a simulation program will provide results needed to be certain that the system can be operated under expected loading conditions without major problems.


The results from this PRT planning effort do suggest that a PRT circulator system could provide a high level of mobility to the persons living and working in the downtown area. While the cost of the circulator would be approximately $200 million dollars, this amount is only about 5% of the $3.7 billion regional transit plan that will be voted on in November of 1996. An investment of this scale would add considerable value to the land in the downtown area and permit it to be developed to much higher densities without significantly increasing auto congestion and its associated noise and air pollution impacts. Such in-filling of existing auto-oriented activity centers is a key element in the Vision 2020 growth management strategy that has been adopted for the region. In fact, it is difficult to imagine how existing auto-oriented activity centers can become more dense unless significant non-auto circulation facilities are provided that can make further development both feasible and desirable.

There are expected to be about 250,000 trips per day to/from and within downtown Bellevue in 2005. If the PRT circulator could attract about ten percent of these trips, its patronage would be abut 25,000 trips per day. At this level, it would surpass Portland's MAX light rail system which carries only about 20,000 trips per day. PRT riders would include commuters making work trips, workers making work, shopping and other non-work trips, residents making shopping and other non-work trips, visitors (business and social), recreational riders and the young/old segment of the population who do not have access to an auto.

Further information about this project and the C-CAD computer program can be obtained from Professor J. Schneider. A paper by him that describes C-CAD's capabilities in more detail can be found in the Proceedings from APM IV held in Los Colinas, Texas, in 1993. It is entitled Designing APM Circulator Systems for Major Activity Centers: An Interactive graphic Approach, pp 535-546. The publication citation is Automated People Movers IV: Enhancing Values in Major Activity Centers, edited by Sproule, Neumann and Bondada, American Society of Civil Engineers, New York, NY, 1993, 838 pp.

  home2.gif (1492 bytes)

Last modified: December 03, 2003