How Freeways Make or Break Active Transportation Networks

Freeways are the pinnacle of car infrastructure, allowing motorists to travel long distances safely and conveniently. But when freeways run through urban areas, they have major impacts on people’s ability to walk and cycle.

Picture a modern freeway interchange. A wide, swooping roadway overpasses the freeway with many lanes, large signalized intersections handle the thousands of vehicles existing the freeway, and smooth, free-flowing on-ramps allow you to comfortably accelerate to highway speeds well before you reach the merge point.

Of course, this is the perspective of a motorist. What about the freeway crossing experience for pedestrians and cyclists? At the interchange described, the experience would be downright awful – crossing high speed traffic at ramps like the one pictured below, while crossing a barren, noisy, polluted landscape that might stretch for upwards of a kilometre without shade, amenities, or places to rest.

Family crossing the free-flow ramp on Jeanne d’Arc Boulevard where a 13-year-old boy was killed in 2019 (CBC News)

But not all freeways are created equal. While freeways are car infrastructure, they can be designed to be less destructive to walking and cycling. This is done by making them “permeable”, so that someone travelling on foot or by bicycle is not cut off from communities on the other side of the freeway. A highly permeable freeway depends on two key ingredients:

  • Safe crossings: Freeway interchanges, and especially those with freeflow ramps, are terrible for pedestrian and cycling safety. They are often designed for speeds of 60 km/h (40mph) or more and require people crossing to wait for a gap in often-heavy traffic – a very difficult task, especially for children.
  • Frequent crossings: Much more so than motorists, pedestrians and cyclists rely on a dense network of sidewalks and bike paths that provides direct access to destinations. Most of the time, freeways carry crossings for cars every two kilometres (1.2 miles); people walking and cycling need much more frequent crossing opportunities.

A previous post on this blog presented ways that a single freeway interchange can be built differently to support walking and cycling. This post will focus on how a segment of freeway affects urban walkability and bikeability – in other words, its “permeability”. Through a self-developed scoring process, I’m going to draw attention to some exceptional and some poor examples of permeability.

Crossing Score

The scoring process is simple: choose a given segment of freeway (aim for ~5km), identify each crossing, and score each using the methodology below. At the end, calculate the freeway’s “permeability score” as the average crossing score per kilometre.

Here’s my general rationale for how points are earned, with more explanation below:

  • First, if there is a crossing for pedestrians or cyclists that is separated from traffic in its own right of way (i.e. a pedestrian bridge or an underpass), the maximum possible five points are earned for the crossing.
  • On the other extreme, if a freeway crossing does not even provide a continuous, dedicated place to walk (in most cases, an accessible sidewalk) across the full crossing, it is worth zero points.
  • A maximum of one point is earned for a crossing with freeflow ramps, and a maximum of two points are earned for a crossing that interacts with highway ramps exclusively at signalized intersections.
  • Finally, if the crossing is a roadway that doesn’t provide access to the highway, either two, three, or four points are earned depending on the number of vehicle lanes and the presence of separated bike lanes.

To make it easy to follow, I’ve organized this scoring into a flow chart below, followed by some examples of crossings and their respective scores.

Flow chart to determine the safety score of a freeway crossing for pedestrians and cyclists


Dedicated active transportation crossing (no interaction with traffic): 5 points
Non-interchange, single-lane roadway crossing with sidewalks and physically-separated cycling facilities (in this case a multi-use pathway): 4 points
Non-interchange, single-lane crossing with sidewalks but no physically-separated bike lanes: 3 points
Non-interchange, multi-lane crossing with sidewalks and cycle tracks: 3 points
Non-interchange, multi-lane crossing with sidewalks but no physically separated bike lanes: 2 points
Interchange crossing with freeflow ramps and sidewalks: 1 point
Crossing with inadequate walking facilities (in this case the sidewalks are 1.2m [4ft] and are not AODA or ADA compliant as a passing area is not provided every 200ft ): 0 points

Now You May Be Wondering…

Why don’t painted bike lanes get any points?

There is an emerging body of evidence that in most situations, painted bike lanes are not comfortable enough to be appealing to the average person interested in cycling, often categorized as the “interested but concerned” user. A recent study from Edmonton, Canada found that, for the majority of the population, painted bike lanes do not increase comfort compared to mixed traffic.

Does a multi-use path crossing a freeflow ramp get more points compared to no cycling facility?

The most vulnerable of vulnerable road users are people walking, and other than some minor signage and visibility improvements, a MUP freeflow ramp crossing is still a freeflow ramp crossing. Freeflow ramp crossings are not all-ages-and-abilities friendly, and are overrepresented in pedestrian and cyclist serious injuries and deaths.

Multi-use pathways crossing freeflow ramps do not improve crossing conditions for pedestrians, who are the most vulnerable road user.

Why does a single-lane roadway get more points than a multi-lane roadway?

Multi-lane roadways carry more traffic, generate more noise, have worse safety performance, and have higher rates of speeding compared to single lane roadways. These are all factors that directly affect the comfort and safety of people walking and cycling.

Why does a freeway interchange crossing automatically get less points?

Freeway interchanges are points where very high volumes of vehicle and truck traffic are concentrated. Even at interchanges where freeflow ramps are not used and traffic turns at intersections, vehicle turning volumes can be very high, and corners are typically built very large to accommodate big trucks.

Permeability Score in Action

Let’s put this new methodology to work! Below, I’ve reviewed three examples, starting with a “gold standard” example from the Netherlands, followed by what could be considered a typical suburban design in Canada, then finally an urban freeway in Montreal that was retrofitted into an existing community.

Example #1: Amsterdam, Netherlands

For an exceptional example of freeway permeability, there are few places better than Amsterdam. I’ve chosen the segment of the A10 Motorway between the A1 and A2 motorways. This area is about 4.5km from the City Centre and thus represents a more “outer urban / inner suburban” context, where there are more wide roads and more people rely on driving to get around.

The freeway runs between Amsterdam’s Zuidoost (south-east) neighbourhood and its core, making the permeability of this freeway highly important for facilitating cross-town trips by bicycle. There are also two rail stations and four metro stations within 1km of this freeway segment, stressing the importance of good walking connections to feed transit ridership.

The image says it all – this is a highly permeable freeway, no matter how you look at it:

  • There is a high quality (score of 3+) crossing every 500m, on average
  • The average crossing score is 4.1 out of 5
  • The “permeability score” is 8.8 points per km

Despite this freeway ranging between 8 and 12 lanes wide, It’s likely that anyone wishing to cross this freeway on foot or by bicycle can do so easily, comfortably, and with minimal detours.

Scoring assessment of the A10 motorway in Amsterdam (click to enlarge)

Example #2: Ajax, Canada

For the other end of the spectrum I’ve selected Ajax, a suburban municipality in the Greater Toronto Area (thanks @michellekearnel for the recommendation). This 120,000 population city is bisected by the 10-lane-wide Highway 401, forming a major challenge for the City. Adding to its challenges, there is both a library and a major regional rail (GO Train) station within 100m of the highway, both of which are major potential places to bike or walk to.

I’ve selected a 3.6km segment of Highway 401, spanning from Church Street to Salem Road. Along this stretch there are five crossings, two of which form highway interchanges with freeflow ramps.

Although the permeability of this highway is arguably just as important as the Amsterdam example, the results are very different:

  • There is only one high quality (score of 3+) crossing, and it’s at the periphery rather than near one of the major destinations
  • The average crossing score is 1.4 out of 5 (with only one crossing scoring higher than 1 point)
  • One of the most important crossings, a non-interchange crossing connecting to the public library, receives a score of 0 because although it has sidewalks on both sides, neither is accessible (width of less than 1.5m with no passing spaces)
  • The “permeability score” is 1.9 points per km

For those without a choice, this freeway presents an uncomfortable, inconvenient barrier between two communities. For those with a choice, it would be difficult to convince someone that driving is not the best way to cross this freeway.

Side note: Michelle tells me that amazingly enough, the crossing that scored zero with an asterisk below actually used to be better for people on foot. When the former interchange here was removed and the bridge replaced, the barrier separating pedestrians from traffic was not reinstated, and the sidewalks were narrowed.

Scoring assessment of Highway 401 through Ajax (click to enlarge)

Example #3: Montreal, Canada

For my final example I’ve chosen an example of an urban freeway that was retrofitted into an existing city neighbourhood, the A15, also known as the Autoroute Décarie (thanks to @brandonlind14 for the suggestion via Twitter).

It’s immediately apparent that this sunken freeway design has a lot of crossings. From a purely numbers perspective there are 20 crossings in the 4.9km stretch I’ve selected – an average of four crossings per kilometre. This is a huge expense, both in terms of construction costs and ongoing maintenance, but must have been seen as necessary at the time of construction, given how the new freeway would split up an existing neighbourhood. Here’s some of the highlights of the results:

  • Similar to the Amsterdam example, there is a high quality (score of 3+) crossing every 500m, on average
  • The average crossing score is 2.7 out of 5
  • The “permeability score” is 10.8 points per km

Especially for someone on foot, this is an incredibly permeable freeway, requiring minimal detours, though the lower average score compared to Amsterdam likely means that most crossings offer less-than-ideal comfort and may be unsuitable for people of all ages and abilities.

Scoring assessment of a segment of A15 through Montreal (click to enlarge)

So What?

The three examples above show just how wildly different freeway design can be from a permeability perspective. While the Montreal Autoroute Décarie was retrofitted into an existing built-up area, the Ajax and Amsterdam examples were built around the freeways. While a retrofitted freeway comes with significant public pressure and expectations to maintain pedestrian connectivity, a new freeway relies on careful planning to provide it.

Amsterdam’s A15 (permeability score of 8.8) should be celebrated as a successful example of a highly permeable planned freeway. There are frequent underpasses exclusively for bikes and pedestrians, and the roads that do cross contain walking and cycling facilities that are physically separated from traffic. In fact, only one of the crossings is at an interchange – this is a great example of how the Dutch plan cycling and walking networks to be different from the driving networks (as opposed to a focus to adding facilities on all driving routes). There’s even two motor vehicle only overpasses, demonstrating that people driving also receive an exceptional level of service here.

Contrast that with Ajax’s Highway 401 (permeability score of 1.9), which forms a major barrier for this community. Despite having some key walking and cycling destinations – a public library and a regional rail station – within 100m of the highway, most of the crossings reviewed involve using sub-standard sidewalks or finding a gap in high-speed traffic to cross a freeflow ramp. The one crossing dedicated for walking and cycling is much more oriented to recreational travel, passing through green spaces at the edge of the City rather than serving helpful destinations.

Finally, while Montreal’s A15 (permeability score of 10.8) has by far the highest frequency of crossings, over half are “low-quality” crossings with a score of less than 3, due to many bridges carrying multiple lanes of traffic without any separated cycling infrastructure. Though it’s possible to cross these bridges on foot, the presence of traffic at almost every crossing would add some stress and discomfort. While Montreal scored better than Amsterdam in terms of points per kilometre, its average crossing score was notably lower (2.7 vs. 4.1), indicating that Amsterdam’s A15 prioritizes high-quality crossings for pedestrians and cyclists, likely a major contributor to the Amsterdam’s success as a cycling city.

The Bentway is a project to turn the unused space under Toronto’s Gardiner Expressway into an active public realm, making this freeway less of a barrier for the community (source: The High Line Network)

As cities commit more and more to embracing active transportation, we can no longer afford to turn a blind eye to freeways. Even in the suburban context, it’s essential that we focus on increasing freeway “permeability” by adding frequent, safe crossings for people walking and cycling. This must be done from the planning level as part of all freeway projects by the highway authorities themselves, and cannot simply be left to municipalities to retrofit their own improvements into existing high-speed interchanges.


  1. Great post Matt. One other element to consider: over or under pass. An underpass allows a rider to enter going downhill and build up speed for the climb on the other side so near zero extra energy required. An overpass requires the rider to put forth extra effort climbing (typically from a flat start) and then often wastes the downhill portion because they have to stop at the bottom for a junction.

    Underpasses require less height difference.

    Underpasses can provide shelter from elements.


    1. Thanks for the feedback! I find in my context the over or underpass decision is usually determined by the terrain and not the design user (which usually leads to an overpass being built), compared to the Dutch example that is mostly underpasses.


  2. This is a good discussion of a big problem, and a helpful way to evaluate and compare permeability in different settings – thanks!

    In my town of Bowmanville, Ontario, we had a long stretch of Highway 401 with no safe and pleasant way for cyclists or walkers to get across. There were two crossings, one an overpass and one an underpass, but both had heavy traffic including a lot of big truck traffic.

    As of a few months ago, we have a beautiful paved multi-purpose path right along a creek bank under the 401 – there’s no traffic, it’s mostly quiet, a joy to use. There is a catch, though. The path doesn’t directly connect to any services – no schools, stores, restaurants, workplaces – so most of the many users are recreational. There are many more hurdles to cross before we have a flourishing active transportation scene.


    1. A big problem indeed! But because freeways are generally built and managed by one agency (at the state or province level) it should actually be easier to do this properly, with the right leadership. Thanks for sharing your experience; too often the crossings are added simply where it’s convenient because of a watercourse crossing, rather than what’s useful for cycling, as you’ve said.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s