Part 2: How Multimodal Accessibility is Measured
This is part two of a multi-part series on multimodal accessibility, an integrated planning framework and analytical tool that all planners should know about. Part One of this series focused on what multimodal accessibility is, and why it matters. Part Two focuses on how accessibility is quantified and measured. Once you understand the nuances of measuring accessibility, you’ll see that there are many ways to measure accessibility, all of which can be useful. In future posts, we’ll start applying these accessibility metrics to tackle a range of planning questions, from forecasting travel behavior to informing placemaking decisions, and many areas in between. But first, we need to know how to measure accessibility in the first place.
A Simple Formula
Like the definition of accessibility, the measurement of accessibility can be very simple, but its power is in its flexibility and potential for great complexity. As such, there are many different ways to get a quantified accessibility “score”. They all stem from the same simple definition, though, of “the number of destinations reachable in a given amount of time”.
In Part One of our series, we mentioned that accessibility is inherently an integrated concept because it required knowledge about land use (i.e. destinations) and transportation (i.e. the ability to reach the destinations). Any measurement of accessibility, therefore, requires some information about the spatial representation of destinations, and the means by which the destinations are reached. The simplest version of this is something we’ve all seen countless times – the map of local activities within a simple radius, like the example below (Map A).
From some origin, in this case a mixed use office development, relevant destinations are highlighted. A band of circles approximate a set of travel times. That right there is an accessibility map – and a fairly useful one at that, at least for understanding local travel destinations from this particular development.
But there’s some room for improvement here. There’s no way that a circle is a true measure of travel time, unless you’re a bird. So let’s substitute that crow-flies version of travel and use an actual travel network. Using tools like Network Analyst in ArcGIS, we can now apply rings of network-based travel times and come up with a new, more accurate accessibility map that looks more like the one below (Map B). In this example, the perfect circle is flattened, as the network more easily facilitates east-west travel than north-south.
At this point, it’s time to start getting more complex. The maps shown so far have focused on access from one point in space. But what if you want to know accessibility at a larger scale, like for a neighborhood, city, county, or even country? Well, that’s what accessibility practitioners like WalkScore, the Accessibility Observatory, or, pardon the self-promotion, Renaissance Planning are doing. By summing the destinations reachable from each point in space over some larger area (for example, all block groups in a study area), accessibility maps like the ones here can be generated. Instead of using buffers of travel time, you need to create origin destination matrices that provide discrete travel times from each origin to all its destinations, which requires some serious computational muscle. But as long as you know 1) the location of your origins, 2) the location of your destinations, and 3) the transportation networks connecting them, then you can make yourself an accessibility map. Whether it’s a map of the number of jobs accessible by car in 45 minutes, or of health care services accessible by transit in 60 minutes, or customers who can reach a new coffee shop in a 15 minute walk, you’ve got yourself a an easy-to-understand visual representation of accessibility.
But there’s one last step that is crucial for transforming accessibility analysis into predictive powerhouses.
The Value of Weighing Opportunities and Cost
When it comes to making travel decisions(like “where should we eat?”, or “how should we get there?”) and location decisions (like “where should we buy a house?” or “where should we open a restaurant?”), not all destinations or transportation networks are equal. For instance, you will go to the grocery store 50 times a year, and the dentist twice a year. Are these places then equally valuable to a potential user, or equally likely to be used on a given day? Of course not. Similarly, are you more likely to go to the dentist that’s 15 minutes away and can be reached by car or transit, or the one that’s 45 minutes away and only reachable by car? Odds are that closer dentist is going to get your business.
Factoring in the “value” of different opportunities (grocery stores is generally more likely to generate a trip than a dentist offices) and travel costs (15 minutes away is better than 45 minutes away) into an accessibility score begins to mimic travel and location decisions, as well as the financial decisions that accompany them. There are lots of different ways to weigh opportunities and travel costs, and there’s no consensus among practitioners. However, it’s clear that being thoughtful about weighing the values of individual destinations and the means for reaching them adds a jolt of analytical power to accessibility analyses. Places like Walk Score have been weighing destinations to great benefit, and have shown a strong correlation between increased walk access and increased home prices. We have found similar things to be true for regional job access, both by car and by fixed guideway transit (not so much for bus service, however). In work for NCHRP and several state and local DOTs, we have been finding surprisingly consistent connections between mode choice and the ratio of jobs reachable by transit to jobs reachable by car, which may help communities make predictions about the impacts of transportation investments without needed costly travel surveys or larger travel demand models.
We’ll get into details about these types of projects in subsequent posts. For now, the thing worth knowing about measuring accessibility is that there are many ways to measure it, from simple to complex, but all of them have value. Because in the end, knowing more about accessibility - the number of destinations reachable in a given amount of time – is clearly a major factor in how and where people travel. And the fact that “how” and “where” can be answered with a simple little metric is a fundamental part of why accessibility explains the world.