Blog post

First Mile-Last Mile, Intermodalism, and Making Public Transit More Attractive

As planners seek to leverage public transit investments with enhanced first mile-last mile connections, it is critical that market analysis guide those initiatives and that impacts and cost effectiveness are part of the performance assessment.
Steven Polzin | July 25, 2017, 8am PDT
Share Tweet LinkedIn Email Comments
Michael Warwick

In the ever-trendy world of transportation planning people seem to be infatuated with discussions of first mile-last mile public transportation connections and intermodalism. Given all the attention, one would think that the traveling public is anxiously awaiting their next opportunity to transfer vehicles to complete their trip. Nothing can be further from the truth. People don't aspire to transfer; they don’t aspire to experience an intermodal terminal. They almost always want to get door to door in the fastest, simplest, and most reliable fashion. Transferring between vehicles is a necessary inconvenience, not a virtue.

The concept of using multiple means of travel to complete a given trip is an outgrowth of the reality that different services and technologies offer the optimal means of travel for different contexts, which can result in trips that require transfers for the overall optimal means of travel. The most obvious example is traveling from, say Chicago to New York. Air travel is the time and cost superior means of carrying out the line-haul component of the trip. U.S. airlines, for example, routinely extract less than $.20 per passenger mile from travelers to transport them between airports while also saving them time and perhaps lodging and meal expenses. But jet aircraft will not pick you up at the door or delivered you to the entrance to your destination. Thus, transferring between modes at airports is a necessary and logical interface between air and surface modes. The opportunity to take advantage of the premium performance of air travel more than offsets the onerousness of navigating through airports and transferring between access and egress modes.

On other kinds of trips, the onerousness of transferring might not be as easily offset by the travel benefits of the line-haul or primary mode of travel. For many shorter urban trips, it becomes very challenging for the onerousness of a transfer to be offset by the benefits of using a combination of modes or vehicles to complete a trip. Travel modeling has long recognized the onerousness of transferring, thus quantitatively penalizing the need to transfer by calculating time spent transferring as two or more times more onerous than in-vehicle travel time. From a practical perspective, transferring introduces uncertainty into a trip. Your arrival at the transfer point is captive to the system schedules and you cannot necessarily minimize the transfer wait. The second vehicle introduces an additional chance to be impacted by unreliable service. For first-time trips, you need to figure out both the location of the destination and how to get to it. You may lose your seat or place and interrupt whatever you are doing during your travel. You might be exposed to weather or other risks, and you can’t use the time as productively as you might have had a transfer not been required.

If you do have to transfer, you want it to be as quick and convenient as possible. While basic amenities such as restrooms and convenience retail might be appreciated, the local traveler is most often interested in getting quickly to their destination and not turning the transfer experience into a retail opportunity or recreational outing. For longer distance intercity trips where the traveler may be captive to more lengthy waits between travel segments, additional retail and personal service accommodations might be appreciated to the extent that they don’t disadvantage other passengers by excessively increasing walk distances or causing other delays.

The vehicle travel to and from the transfer location should deviate from the optimal origin-destination travel path as little as possible. If one does have to suffer a transfer, they would much preferred that the point of transfer not dramatically impact the circuity of their travel.

Given these basic principles, let’s look a little more closely at urban public transit trips as it relates to first mile-last mile connections. For local transit travel, according to the American Public Transportation Association, about 50% of trips from an origin to a destination involve a transit to transit transfer in addition to any non-transit connections to or from the transit trip [1]. Often times these are simple transfers to a perpendicular route at an intersection in order to use a grid route network. Many basic public transit networks require a transfer for travelers to complete their trip. Limited walk distance, safe, protected environments, and frequent service minimizing the transfer wait can ameliorate the negative impacts of having to transfer.

The growing motivation for providing first mile-last mile connections derives from the logical desire to increase the accessibility to public transportation for more homes and destinations. A multitude of efforts in recent years have been carried out to quantify accessibility of residents and activities to public transit. Early work carried out by CUTR indicated that about half the homes in the America were within a half a mile of a transit route. A slightly higher share of employment locations were similarly within a half a mile of transit. More recently, sophisticated software tools have been developed to evaluate accessibility via transit, such as initiatives by the Brookings Institute and the University of Minnesota Accessibility Observatory, as well as tools such as Transit Score. The collective message of these analyses indicate that, in general, access to transit both geographically and temporally is, on average, limited. Hence, folks are interested in improving first mile-last mile connections with the hopes of making transit more attractive and productive. 

Historically, line-haul premium transit services provided feeder bus, park-and-ride, and kiss and ride (drop off) opportunities so that travelers could access these premium modes, most typically for longer-distance commute travel. More recently, additional means of access, including bikeshare, carshare, and transportation network company (TNC) connections (i.e., Uber, Lyft, etc.), are being deployed. Automated shuttles are being evaluated as yet another means of enhancing the appeal of line-haul premium travel modes. These concepts make sense in contexts where the line-haul mode is sufficiently attractive by virtue of its speed or cost advantages that the traveler is willing to incur the inconvenience, time cost, trip circuity, or other potential negative characteristics of incurring one or more transfers to complete a trip.

Better first mile-last mile connections work where they work. But where is that and what planning and service investments makes sense to enhance first mile-last mile connections? Individuals who use intermodal connections do it either because there is no viable alternative or because the disutility of transferring is more than made up for by being able to take advantage of the line-haul mode of travel. This is most possible in situations where the line-haul mode is superior to other travel options, typically meaning it is faster by virtue of fewer stops, exclusive guideway, signal priority, utilization of a higher performance travel path (freeway versus arterial), and that the transfer penalty is minimized most typically by having high-frequency service on the line-haul. Faster travel speed is typically only virtuous in instances where the distance of the trip is sufficient to accumulate enough marginal travel time advantage to offset the transfer induced delays. Thus, enhancing first mile-last mile connections has the greatest leverage for longer distance trips and premium services. 

Over 60% of person trips according to the last National Household Travel Survey, are less than 5 miles in length, over 75% less than 10 miles in length. Many of the shorter trips are unlikely to be appealing as trips requiring first mile-last mile connections to travelers who have choices. Absent extremely high quality first mile-last mile connections, the circuity and delays likely to be introduced by a first mile-last mile connection(s), as opposed to a direct door-to-door single vehicle trip, are unlikely to make this arrangement attractive for travelers with choices. Such services could incentivize more trips or increase convenience by shortening walk access for travelers without personal vehicle options.

So what does this have to do with anything? Numerous communities are striving to leverage their transit investments and increase mobility for their populations by exploring additional first mile-last mile connections. Though well intentioned, first mile-last mile programs will be most successful if fully informed by an understanding of traveler behavior in general and market conditions in particular. Context has implications in terms of the magnitude of ridership response as a result of improved connections based on the geography of deployment and the trip pattern emanating to and from that geography. First mile-last mile connections are most likely to attract new travelers if they offer high-quality connections, support high performance modes, and serve sufficiently long trips such that the circuity and transfer disutility can be amortized over a longer line-haul premium service segments.

In addition, equity considerations may become an issue. Additional investments in first mile-last mile connections will have to be evaluated in the context of alternative investments in service and facility improvements. Additionally, attention needs to be paid to the question of who will benefit, both geographically and demographically, from various first mile-last mile connections. How much should be spent to coax travelers with personal or private sector mobility options to use public transportation, or should resources be directed to basic service improvements for those dependent on transit?

Experimentation and a learning curve are to be expected as new technologies, business models, and deployment strategies are deployed and experience accumulates. But it will be important to glean a well-informed sense of the public and user costs, travel impacts, and environmental, safety, and other impacts. The role of new technologies and service models in enhancing connections to public transportation is important, but like everything about public transit, it’s not so easy to make it work.

The comments are those of the author and are intended to provoke reflection. They do not reflect the policy positions of any associated entities or clients. Polzin@cutr.usf.edu.


[1] WHO RIDES PUBLIC TRANSPORTATION, American Public Transportation Association, page 74, January 2017.

Share Tweet LinkedIn Email