Evaluating equitable mobility impacts with the Mobility Energy Productivity metric 

Abstract
The Mobility Energy Productivity (MEP) metric [1] offers a new paradigm for evaluating the quality of mobility within an urban area. The metric is grounded in traditional accessibility theory with the novel addition of weighting energy expenditures of mobility options in addition to cost and time. Thus, MEP is well-positioned as an evaluation metric to understand current and future mobility impacts of emerging transportation technologies and enable exploring the geospatial and socioeconomic variations of the quality of mobility in communities. For a given location, the metric (default resolution of km2) quantifies the number of opportunities that can be accessed within set travel time thresholds (10, 20, 30, and 40 minutes) by various modes (walk, bike, transit, and car). The opportunities measure is first proportioned by activity engagement frequencies and then weighted by the time, energy, and cost efficiency of each mode available at a location. The metric can be viewed at any location as disparate mode-level scores or be aggregated using reported mode shares. The location level scores are also aggregated to the city level using geospatial population weighting.

While the initial iterations of the metric were developed using average traveler characteristics (i.e., activity frequencies, mode shares, and population weighting derived from the whole population), it was quickly realized that an average MEP score for a location (encompassing all modes, activities, and travel times) might not resonate with specific sociodemographic cohorts. For example, individuals who are car-constrained or belonging to the lower income cohort can have different travel preferences and temporal/modal constraints compared to an ‘average’ traveler. Thus, the MEP metric is being customized to reflect the characteristics of various cohorts. The representation of the MEP scores for specific cohorts involves the customization of a variety of parameters used in MEP calculations (e.g., trip frequency, travel time tolerances, modal preferences, and population distributions). While initial explorations of sociodemographic MEP calculations are being carried out on unidimensional cohorts (for example: age, education, or income independently), the goal is to be able to carry out multidimensional cohort analyses (e.g., age + education + income) to quantify the quality of mobility for more specific cohorts to help examine impacts to disadvantaged communities.

The presentation will focus on: 1) the motivation, requirements, and framework for the metric in its current state; 2) how the metric can be used to better understand how the quality of mobility differs across socioeconomic cohorts; and 3) future work plans to expand the framework and highlight collaborations with industry partners, cities, and potential to be a standard metric for Smart City assessment. It is important to note here that additional factors (beyond the ones already being considered in the calculation) are likely needed to fully understand impacts to disadvantaged communities in an environmental justice context. The MEP research team is currently exploring additional methods and data sources to expand the framework to include emissions, safety, and other mobility-related externalities. This will be included in the presentation based on the progress made in the coming months.