Potential large-scale transportation investments in the Pittsburgh area have made headlines with the announcement of Joe Biden’s infrastructure plan as well as the publication of the Port Authority’s NEXTransit plan. One of the investments being considered is extending the East Busway.
The existing East Busway is the Port Authority’s best asset. On the segment with a dedicated right of way, the buses average 27 mph1The P1 Bus Schedule lists a 20 minute travel time for the 9.1 miles between the start of the busway in Swissvale and the Penn Station Stop downtown.$\frac{9.1\ \textrm{miles}}{20\ \textrm{minutes}}\cdot\frac{60\ \textrm{minutes}}{1\ \textrm{hour}}=27.3 \frac{\textrm{miles}}{\textrm{hour}}$ including stops. This is faster than many light rail systems, and faster than driving in the busway’s vicinity. It also has incredibly low operating costs. The P1/P2 bus which operates on the East Busway has an operating cost of $\$$2.41 per passenger, compared with $\$$5.64 for buses overall and $\$$8.11 per passenger for the light rail system.2Operating costs per passenger are from the 2018 annual service report.
There are a number of ways a busway expansion can be implemented. Getting the details right is important to be cost effective and serve the most people.
The Extension should be Built to McKeesport through East Pittsburgh and Duquesne
The busway extension alignments being considered in the NEXTransit plan 3See page 59. are shown in Figure 3. The existing busway ends in Swissvale. The plan calls for phased extensions of the East Busway; first the 2.9 miles to East Pittsburgh, and then either from East Pittsburgh to McKeesport through Duquesne and/or to Monroeville through Turtle Creek. The extensions to McKeesport and Monroeville from East Pittsburgh would be approximately 4 miles each, so a total of 7 miles of new busway would be needed to reach either one or 11 miles to serve both.
An East Busway extension that continues to McKeesport would generate more additional riders per mile of extension than an extension to East Pittsburgh alone. It would also represent a huge improvement in transportation infrastructure to residents of the Mon valley.
The Extension Would Provide Huge Time Savings
During rush hour, it takes 78 minutes to get from McKeesport to Downtown by bus4The 78 minute travel time is based on an 8 AM weekday departure on the 61C published schedule. The P7 and 56 buses provide a faster travel time of 55 minutes at rush hour, but those buses run much less frequently than the 61C. The Downtown to Oakland BRT development should reduce the 61C travel time., and up to 45 minutes by car 5Based on setting the departure time from McKeesport to Monday at 8 AM in Google Maps. If speeds on the extended busway are the same as the existing busway, the time would be reduced to 44 minutes6$\frac{20\ \textrm{miles}}{27.3\ \textrm{mph}}\cdot\frac{60\ \textrm{minutes}}{1\ \textrm{hour}}=44\ \textrm{minutes}$.
The busway would provide a larger improvement to other destination pairs. For example, it would be 36 minutes from McKeesport to East Liberty on the busway compared with up to 55 minutes by car at rush hour.
A Longer Extension Would Serve More Riders
An extension to McKeesport is superior because it would serve significantly more people than an extension just to East Pittsburgh or one to Monroeville.
The East Pittsburgh extension would serve Braddock, North Braddock and East Pittsburgh which collectively have a total population of 8,186. As can be seen in Table 1, continuing the extension through Duquesne to McKeesport or through Turtle Creek to Monroeville would serve a much larger number of people than an extension that ends in East Pittsburgh.
Table 1: Population of Municipalities Served by Possible Busway Extension
Municipality | 2019 Population | Population Density $\left(\frac{\textrm{People}}{\textrm{mi}^2}\right)$ | Walk Score7Obtained here. A higher number means a more walkable place with 100 being the best. The methodology is explained here. |
Braddock | 1,869 | 2,875 | 66 |
North Braddock | 4,741 | 3,059 | 37 |
East Pittsburgh | 1,576 | 4,041 | 39 |
Duquesne | 5,543 | 2,717 | 40 |
McKeesport | 19,225 | 3,554 | 44 |
Turtle Creek | 5,197 | 5,358 | 53 |
Monroeville | 27,687 | 1,404 | 19 |
McKeesport is Easier to Serve with Transit Since it is More Walkable
Despite having a lower population than Monroeville, McKeesport is a much better target for busway expansion. McKeesport was largely developed before the widespread adoption of the automobile. Its population in 1910 (42,694) was more than twice what it is today. By contrast, Monroeville (then known as Patton Township) was largely rural in 1910 with a population of only 3,210. As a result, McKeesport retains a development pattern that is much more compatible with transit than Monroeville. Some examples include:
- McKeesport’s road network still has a street grid. Jarrett Walker has a great post describing why street grids are easier to serve with transit than dendritic urban sprawl.
- McKeesport’s population density is greater than Monroeville’s.
- While McKeesport’s pedestrian network could use improvement, it has better pedestrian infrastructure than Monroeville.
- McKeesport’s zoning code allows for higher residential density in much of the city and allows residential and commercial uses to be built closer to one another than Monroeville’s zoning map and code. The land use patterns permitted by McKeesport’s zoning code allow it to be more easily served by transit.
This development pattern allows more origins and destinations to be sited within walking distance of McKeesport’s bus stops than Monroeville’s. This can be seen by reviewing the 15 minute walksheds8The walksheds were generated with this tool which assumes a walking speed of $1.4 \frac{\textrm{m}}{\textrm{s}}$ or 3.1 mph. McKeesport and Monroeville have a hilly topography that may result in slower walk speeds for some travelers. of the Monroeville Mall compared with Lysle Boulevard in McKeesport illustrates the advantages of McKeesport’s development pattern. Pedestrians in Monroeville have to walk through a sea of parking and along streets that prioritize motorists over pedestrian safety to reach destinations from a bus stop. A bus stop sited in McKeesport is in walking distance of far more residences, jobs, and destinations than a bus stop sited in Monroeville.
How Much Money the Port Authority Should Invest
To determine how much the Port Authority should invest, it is helpful to consider likely ridership. While a rigorous study to model ridership for a McKeesport extension has not been performed, a rough estimate may be made.
The East Busway Extension Feasibility Study estimates that the 2.9 mile east busway extension to East Pittsburgh would result in 1,629 additional daily boardings9 See pages 40-44.. If the number of riders scale proportional to the population of the municipalities served, extending to McKeesport would result in 6,558 additional daily boardings10There are 8,186 people in the municipalities served by an extension to East Pittsburgh, and 32,954 people in the municipalities served by an extension to McKeesport. $1,629\ \textrm{riders}\cdot \frac{8,186}{32,954}=6,558\ \textrm{riders}$.
In addition to increasing the number of people who ride transit every day, a busway extension would provide significant time saving to existing riders from the municipalities served, and result in long term operating cost savings for the Port Authority. Even passengers at existing stations would benefit. 117 articulated buses would be needed to carry the additional passengers11There are 56 seats on an articulated bus. $\frac{6,558}{56}=117$. For comparison, the average number of passengers per service hour for buses in FY17 was 33 and traveling the extended busway would take less than an hour. and those additional buses would reduce headways at existing busway stations. This would save existing passengers time as well.
To determine what construction cost would be justified from these benefits, it is worth comparing with other recent projects. The Port Authority is currently spending approximately $\$$37,000 per parking spot to build a garage at the Carnegie Park and Ride. While a parking spot would result in two boardings for a round trip if each spot was used by a new rider, a majority of new park and ride spots are used by people who previously walked to the stop or parked further away. A busway extension has a number of benefits over a park and ride:
- It improves service for existing riders as well as facilitating new ridership.
- It enables people to reduce their car ownership and dependence.
- It has network effects that can be leveraged for future transit expansions and improvements. For example, any other transit improvement on a line that has a transfer connection with the busway would be made more valuable because passengers could reach the destinations served by the busway extension more rapidly.
Therefore spending $\$$37,000 per additional boarding on a busway expansion would be a significantly better investment than developing park and ride facilities. At this level of investment per additional boarding, $\$$243 million would be justified on the 7 mile expansion to McKeesport. The next section explains that recent interstate projects in Southwestern PA have been completed at lower per mile costs than this.
The FY2020 Budget lists $23.9 million12See page 38. for park and ride expansions. The Port Authority should end park and ride expenditures or charge enough for parking to cover both capital and operating costs of the parking itself. This would save enough money to pay for the extension to McKeesport within a decade.
Cost Estimate from Previous Extension Study is too High
In 2017, Port Authority published a study evaluating extending the East Busway by 2.9 miles on an alignment continuing from Swissvale along the border of Braddock and North Braddock to East Pittsburgh. The study estimated this extension would cost $\$$549 million. This is substantially higher than freeway projects in the Pittsburgh metro area.
A comparison is shown in Table 2. The Southern Beltway and Mon-Fayette expressway costs each include the construction of six new interchanges. The Mon-Fayette expressway cost is an estimate as construction has not yet started.
Table 2: Comparison of Roadway Construction Costs
Roadway | Cost ($\$$m) | Length (mi) | Lanes | Cost per mile ($\$$m) | Cost per lane mile ($\$$m) |
Mon-Fayette Expressway | $\$$2,100 | 14 | 4 | $\$$150 | $\$$37.5 |
I-76 Lane Expansion | $\$$117 | 7.5 | 2 | $\$$15.6 | $\$$7.8 |
Southern Beltway | $\$$900 | 13 | 4 | $\$$69 | $\$$17.3 |
East Busway Expansion | $\$$549 | 2.9 | 2 | $\$$189 | $\$$94.7 |
A report from the Federal Highway Administration (FHA) gives typical road construction costs based on data from 2004 to 2014. A subset of this data is provided in Table 3 and it shows lower construction costs than the Pittsburgh area projects in Table 2.
Table 3: FHA Typical Construction Cost Data
Road Typology | Typical Cost per Lane Mile ($\$$m) |
Rural, Flat Interstate | $\$$3.551 |
Rural Mountainous Interstate | $\$$10.121 |
Small Urban Interstate | $\$$19.373 |
Large Urbanized Interstate | $\$$38.303 |
Large Urbanized Principal Arterial | $\$$25.864 |
A 7 mile busway extension would cost $\$$109.2 million at I-76 lane addition costs and $\$$ 242 million at Southern Beltway costs. Given that a busway extension would require fewer interchanges than the interstates, a lower cost may be achievable. An additional expense for the segment between East Pittsburgh and McKeesport would be two bridges crossing the Monongahela River. A new bridge for a four lane highway across the Monongahela River in Fayette County constructed in 2012 cost $\$$96 million. While a two lane bridge would cost less, this would still represent a significant cost that may make it challenging to match the cost of projects at the lower end of the range.
The NEXTransit plan estimates $\$$121-151 million for this project, although it doesn’t state what alignments and roadway characteristics correspond to this budget.
The Port Authority should pursue an implementation that provides a grade separated right of way equivalent to the existing busway for the 7 mile extension to McKeesport. It should be a cost aggressive goal to do this within the $\$$151 million NEXTransit plan budget. However, the project would still be clearly justified at a $\$$243 million budget.
The Transit Improvements that Do Make Sense for Monroeville
A busway extension to Monroeville would only be justified if it was paired with transit oriented zoning reform, similar to changes I previously wrote about for Pittsburgh.
Pittsburghers for Public Transit funded a study entitled ‘Beyond the East Busway’ which evaluates transit improvements in the Mon Valley and Monroeville that are more cost effective than a busway extension. It recommends improvements from the East Busway to Monroeville that are similar to the Bus Rapid Transit (BRT) system being built between Oakland and Downtown. Recommendations include:
- Signal Priority at traffic lights. This should be pursued at every traffic light along a bus route.
- Dedicated Bus Lanes on various streets including South Avenue in Wilkinsburg. This is illustrated in Figure 6 taken from the Beyond the East Busway Study.
- A ramp connecting I-376 to the East Busway with dedicated bus lanes on the I-376 shoulder out to Churchill. This is illustrated in Figure 7 taken from the NEXTransit plan.
These are all good improvements that should be pursued.
Conclusion
An extension of the East Busway is an excellent investment if it is done with construction costs that are typical of other road projects in southwestern Pennsylvania. The alignment that would generate the most value would extend to McKeesport through East Pittsburgh and Duquesne. The cost of a grade separated busway to Monroeville is not justified, and there are more cost effective BRT style improvements that should be implemented.
The existing East Busway enables high quality transit service provided at low cost. Capital projects that leverage this asset are one of the most effective ways the Port Authority can improve the transit network.
I made this dedicated Monroeville Ramp from and to the Parkway east a suggestion for the past 15 years or so. This low cost high value improvement of bus ridership for the east suburbs (including Westmoreland) a home run for travel on PAT.
https://www.youtube.com/watch?v=iUji-2y84pM SPC’s SmartMoves Connections project included a simulation of shoulder-running bus lanes on I-376 that would connect via a bus-exclusive ramp to I-376. Simulation at the link.
[…] lanes for buses, including extending the East Busway and a bus lane on the Fern Hollow […]