Leveraging mobility data analytics to inform mobility hubs

Mobility hub is a platform where people can connect to multiple modes of transportation to make their trip safe, convenient and reliable.

The objectives of mobility hubs include the followings:

1.      Provide transit supply and serve multimodal travel needs.

2.       Enhance first-/last-mile connectivity and facilitate facilitate seamless transfers.

3.       Promote equitable accessibility.

This projects develops a GIS-based analytical framework for Florida agencies to decide the optimal locations of mobility hubs based on five criterias. we apply our analytical framework to Gainesville as the case study region in this study. In the future, we will expand this framework to other cities in Florida.

Step 1: define the spatial unit for locating mobility hubs.

We believe that mobility hubs should be sited within about 1 mile buffer from groups of adjacent transit stops. Then we define this as spatial unit in our analysis.

Step 2: define criteria and weights.

Based on the stakeholder objectives, the siting of mobility hubs considers five criteria: transit supply, first-/last-mile connectivity, accessibility, infrastructure, and socioeconomic equity. Each of the criteria is assigned with different weights and has several sub-criteria, which are weighted sums of scaled continuous variables. The weight scheme and chosen variables were decided according to the literature and discussion with stakeholders.  

Step 3: overall analysis and ranking.

Five index scores of different criteria are calculated and aggregated into the mobility hub index score for a given spatial unit. These scores are ranked to select the top amount as most potential sites for identifying the neighborhood level mobility hubs.

As mentioned, we consider five criteria in deciding the mobility hubs. Each criteria have several sub criteria, which involve different continuous variables to be calculated.

The goal of mobility hubs is to provide transit ridership and supply to satisfy multimodal travel needs. The related data is mainly collected from Gainesville Regional Transit System (RTS), the local transit bus service in Gainesville. The bus routes and stops information are collected from the General Transit Feed Specification (GTFS) dataset, a public dataset for public transportation schedules and associated geographic information. The bus ridership data (including passenger counts, on-board wheelchair and bicycle amount at stops) are collected from city of Gainesville.

Mobility hubs should also solve the first mile/last mile gaps and enhance transit connectivity. The FM/LM connectivity data includes two aspects: micromobility trip and census block level FMLM gap score. The micromobility trip data is collected from city of Gainesville, including the spatiotemporal information of e-scooter and micro transit trips. The data source of census block level FMLM gap score includes American Community Survey (ACS) and LEHD Origin-Destination Employment Statistics (LODES). ACS is a survey conducted by the U.S. Census Bureau and includes detailed demographic, economic, social, and housing data. LEHD provides employment and workplace characteristic data. To calculate the census block level FMLM gap, we need latest block level population data from ACS and job data from LODES.

Mobility hubs should also contain some infrastructure for pedestrian and cyclists, which includes two aspects of data: intersection density and road infrastructure for pedestrians and cyclists. The intersection density data is collected from Smart Location Database, a nationwide geographic data resource for measuring location efficiency that includes neighborhood design, destination accessibility and transit service. We collected intersection density at which multi-modal facilities or pedestrian-oriented facilities met and where the number of legs was greater than 4. The road infrastructure data is collected from OpenStreetMap (OSM), which provides detailed information about road networks.

Besides, mobility hubs should solve social equity issues, making sure that even disadvantageous groups can access the mobility hubs. For socioeconomic considerations, we take into account five variables: Household without vehicle, Black population, . These socioeconomic data is also collected from ACS.

Finally, mobility hub aims at enhancing the accessibility to destinations. To measure the accessibility, we collected data regarding destination accessibility via auto or transit from Smart Location Database. To evaluate the walkability around bus stops as another measurement of accessibility, we also collected walk score from WalkscoreAPI. 

The first criteria, transit ridership and supply, is decided by two sub criteria: ridership and service frequency, which involves some variables to be calculated as table above shows (e.g., passenger count, number of bus stops). These variables are all stop-level and need to be aggregated to the spatial unit. Then they are scaled to 0-100 and the weighted sum are calculated to derive the index score of transit ridership.

FMLM problems refer to the gap between transit stops and travelers’ origin/destination. Micromobility can solve the FMLM problems by enhancing the connectivity to transit stops. If more micromobility trips are around bus stops, then such bus stops have more needs for solving the FMLM gap. We measured the FMLM connectivity with the micromobility (scooter and micro transit) trip origin/destination trip amount within 100ft buffer zones at grouped bus stops.

Additionally, we calculated the census block level FMLM gap score as another measurement of the FMLM connectivity index score. This score is evaluated based on the distance between the centroid of each census block and the nearest bus stops, weighted the number of jobs/ the total population of the block. This involves the following steps as illustrated.

Beside micromobility trip amount and block level FMLM gap score, the number of bikes passenger carry onboard at each stops is another measurement to evaluate the FMLM connectivity.

The infrastructure index score is measured by two dimensions:

1.        The sidewalk and bicycle lane length, the ratio between sidewalk/bicycle lane length and overall road network length within the spatial unit.

2.       The intersection density at which multi-modal facilities or pedestrian-oriented facilities met.

The original data should be clipped and assigned to the spatial unit.

The socioeconomic variables are collected at census block group level. To aggregate the socioeconomic factors to the spatial units, we selected the census block groups intersected with the spatial unit and then calculated the indicators (e.g. percentage of non-Hispanic white people).

The accessibility to destinations is measured by the following two aspects: (1). destination accessibility via auto or transit; (2) workability score.

The mobility hub index is the weighted sum of index score of each criteria. We assign the same weights (20%) for each criteria, and also emphasize each criteria by assigning 50% weights, while others remained 12.5%.

By weighting each criterion, we can compute the mobility hub index score of each spatial unit. To identify multiple mobility hubs from the spatial units, we implemented an algorithm to choose from the spatial unit following four steps:

1. Select the existing (or planned) mobility hubs.

2. Exclude all potential hubs within 1-mile of the selected hubs from considerations

3. Select the hub with the highest mobility hub index as the next hub

4. Repeat steps 2 and 3 until the service coverage is >60% or the total number of hubs reaches N

There are three hubs planned to be sited in Gainesville as figure 10 shows: Butler Plaza Transit Center; Eastside hub; a downtown hub. These are considered at the initial stage of siting the mobility hubs.

For each of the five criteria above, the calculation result is shows as follows.

We also adjusted the weights of different criteria and visualized the outcomes of planned mobility hubs given different scenarios.

We acknowledge the Florida Department of Transportation (FDOT) as our funder and the city of Gainesville as collaborator.