Skip to content

Latest commit

 

History

History
43 lines (39 loc) · 2.23 KB

README.md

File metadata and controls

43 lines (39 loc) · 2.23 KB
Raw

SFMTA TSP

Motivation

  • San Francisco has invested in a Transit Signal Priority (TSP) system, which helps buses get green lights more often.
  • After implementing TSP on several lines, the city now has data to be able to answer the key question:
    • Is Transit Signal Priority resulting in improvements to travel time for MUNI bus riders?

Goal

  • Build a web tool that allows SFMTA staff (and other interested parties) to:
    • Visualize travel time improvement over time at different levels of data, including:
      • Different stops / intersections
      • Various times of day
      • Different kinds of intersection (near vs. far side stop, types of signals)
    • Visualize “dwelltime,” or how long buses stay at different stops
    • Filter and otherwise explore data on their own

Design Strategy

  • Visualization Tool
  • Key Interaction: Comparisons of Travel Time
    • Stop Type:
      • Outbound vs. Inbound
      • Near side vs. Far side stops
    • Time:
      • Day of Week
      • Time of Day
    • Individual Condition:
      • Traffic Lights between Stops
  • Key Metric:
    • Distributions of Bus Speed Before vs. After TSP

Evaluation Strategy

  • Analytic Evaluation by the SFMTA staff:
    • A heuristic evaluation by the SFMTA staff providing a rating on the usability of the visualization in answering the key question, “Is Transit Signal Priority resulting in improvements to travel time for MUNI bus riders?”
    • Cognitive walk-throughs of the prototype for specific tasks using available features of the visualization
      • Will the user use a given feature?
      • If so, is the control for the action visible?
      • Is there a perceptible link between control and action?
      • Is the feedback from the feature appropriate?
  • Empirical Evaluation by the project team:
    • A qualitative evaluation of the important features of the visualization by:
      • sending questionnaires/surveys to the users asking them specific to rate features of visualizations
      • conducting interviews with the users-- both individually and as part of focus groups, asking specific questions to prompt user’s judgement of the visualization
      • observing the users while using and interpreting the visualizations with and without any directions given by the project team