Disseminating Traveler Information on Travel Time Reliability

CHAPTER 2. COMMUNICATING TRAVEL TIME RELIABILITY

Cognitive science has demonstrated that most people are not good at understanding complex statistical concepts, which is the foundation of Travel Time Reliability (TTR) information. As with other forms of traveler information communication like traffic signs, a Lexicon of phrases to convey TTR information should accomplish the following:

Communicate a useful message.
Improve on-time performance.
Encourage trust in the message.
Communicate the "riskiness" of a route.
Distinguish TTR from real-time traveler information.⁽²⁾

Based on the results of the initial human factors studies conducted as part of the Strategic Highway Research Program 2 (SHRP2) L14 project and on current traffic engineering practices regarding communication to drivers, recommendations were presented regarding the use of various terms related to TTR concepts. This document is based on that original research, along with additional research funded by the Federal Highway Administration (FHWA) that validated the initial findings of the L14 project and fine-tuned the Lexicon of TTR terms to better represent motorists' understanding and preference for various terms and access method.

The report, A Lexicon for Conveying Travel Time Reliability Information, developed as part of the SHRP2 Reliability Project L14—Effectiveness of Different Approaches to Disseminating Traveler Information on Travel Time Reliability—established a preliminary set of recommended terminology and guidelines for conveying TTR information to road users and stakeholders to reflect reliability conditions.^(2,³⁾ Specifically, a Lexicon of phrases was developed for each of the eight TTR terms:

95^th percentile.
Arrival time.
Average travel time.
Buffer time.
Departure time.
Recommended departure time.
Recommended route.
Reliability.⁽²⁾

Each Lexicon contained detailed guidelines for TTR information that would most likely be understood and used by the travelers. The Lexicon terminology and guidelines developed in L14 were based on a series of laboratory studies, and none of the terms were tested in a field environment.

To help agencies and other transportation-related entities better deploy and use the recommended Lexicon terminology, a field study was conducted to test the phrases to demonstrate the technical and institutional feasibility of their use and determine the potential costs and benefits of using these products of L14. The overall project objectives were to:

Convey TTR information from theory to reality.
Better understand travelers' perceived value of TTR information.
Better understand the current and future dimensions of the TTR information marketplace.
Better understand what network travel and TTR information travelers require.
Develop and implement a plan to test and evaluate the preliminary design guidelines and Lexicon phrases for disseminating TTR information.
Develop guidelines based on the outcome of the test and evaluation.
Lay out the barriers to communicating TTR information to travelers and steps to overcome barriers.
Outline how different travelers will use TTR information differently (e.g., one-time visitor vs. regular commuter).
Recommend and develop outreach activities in order to encourage use and adoption of the recommended Lexicon phrases and guidelines.
Outline clear steps that agencies need to take to start getting the TTR information into travelers' decision processes.

The field study was conducted in three different metropolitan areas—Houston, Texas; Columbus, Ohio; and Raleigh-Durham, North Carolina. At each of the three study locations, two separate assemblies of Lexicon terms for TTR information were delivered using three distinct methods: agency website, mobile smartphone application, and a traditional keypad response 511 system. The study was intended to assess the effectiveness and utility of TTR information to travelers in these three different areas of the United States and to help develop guidelines for agencies to disseminate this information in their jurisdictions.

FIELD STUDY RESULTS

The results of the field study are briefly summarized in this section. A complete discussion and presentation of the study results can be found in the final report from the study.⁽⁴⁾

Local partners from each study location provided historical traffic datasets as a source for determining the TTR information that was provided to study participants in the field study. Using a custom Smartphone application, study participants collected Global Positioning System (GPS) and travel characteristic data for all trips made over a four-week period along specific travel corridors in each study location. The travel period was divided into two equal phases. Only travel data was collected during the first phase. In addition to collecting travel data during the second phase, TTR information was provided to participants according to their randomly assigned Lexicon assembly (A or B) and dissemination platform (i.e., App, Web, 511).

Baseline and exit surveys were administered to study participants to obtain their typical travel patterns and perceptions of the provided TTR information. The data collection for the field study occurred during multiple rounds between April 2015 and April 2016. Depending on the round, participant recruitment was conducted in various manners, including targeted postcard mailings to randomly selected households in zip code areas near the study corridors, flyers and newsletter advertisements at major university campuses within the study corridor, and postings on social media and local partner websites.

A total of 762 participants across the three metropolitan areas completed the field study. Their aggregate information for the Baseline Survey, Phase 2 trip diaries, and Exit Survey were utilized to conduct an analysis for determining if TTR information dissemination Lexicon (A, B) or channel (App, Web, 511) generated a significant impact on the utility or satisfaction of trip planning and execution. The Baseline Survey results were analyzed to assess any preexisting differences between treatment groups, while the Exit Survey results were analyzed to establish the response probabilities associated with the TTR information dissemination channel and lexicon as a function of demographic and travel characteristic data.

Analysis of the baseline survey data found that there were no statistically significant associations between treatment group and demographic and travel characteristics variables. Given this result, subsequent differences between the treatment groups could be interpreted to be associated with the testing and not possibly reflective of an a priori bias in the panel composition.

Only one survey question resulted in statistically significant differences between the Lexicon assemblies. Those that were provided Assembly A were more likely to change their trip plans for a familiar trip than those provided Assembly B. However, it should be noted that across all the questions evaluated, no multiple comparison adjustments were made to the statistical test results. Therefore, each "statistically significant" result has a potential to have been reached in error. Across a large number of such results, it becomes likely that one or more of the significant results really represent only random variability in the responses. As such, this one result should be evaluated with caution.

Throughout the survey responses, there were several instances showing lower utility or satisfaction for the 511 information channel than either the Web or App access channels. In a few instances, the App access demonstrated superior responses than 511 or the Web.

Statistical models accounted for several exogenous factors, and this clearly seemed important to properly interpreting the results of interest for the Lexicon assemblies and information channels. The city of the participants only rarely appeared as a significant factor in the models, suggesting that location was not an important differentiator in the outcomes. Among other exogenous variables, education level and the number of Phase 2 trips taken with the TTR information were found to significantly correlate with responses for multiple questions. Specifically, participants with college degrees and those taking more trips in the TTR phase were less likely to report that they had made changes to either a familiar or unfamiliar trip as a result of the information. This finding was reinforced by the fact that in subsequent survey questions, these two groups were identified as finding TTR information less useful and reliable, and disagreeing that it reduced their travel time or their travel stress.

Previous | Next