STUDYING THE BOTTLENECK ISSUE AT WORK ZONES AND ASSESSING THE EFFECTIVENESS OF A PORTABLE DYNAMIC LANE MERGING SYSTEM IN PROMOTING ZIP MERGE BEHAVIOR

Abstract

In this research study, drivers’ preferences of and responses to text and graphic road sign messages at work zones were analyzed in an attempt to reduce the bottleneck conditions at lane-reduced work zones. A particular emphasis was placed on zip merging, an application of alternate vehicle merging already successfully employed in the Czech Republic and other nations, that eliminates the perceived right-of-way held by drivers in the open lane(s) at merge points. Both experimental and currently existing advisory messages were evaluated and compared for effectiveness in this study. These advisory messages, associated with three driving advisory conditions (DACs), “Merge to the Right Lane,” “Zip Merge” (vehicles take turns), and “Continue Travel Normally,” were assessed through a questionnaire survey, driving simulation and traffic study to seek the best messages in advising drivers in different traffic conditions when approaching work zones. A questionnaire survey was first deployed to identify participants’ preferences towards a series of messages posted on variable message signs (VMSs). Participants rated each message from one to five as to their effectiveness in advising drivers in different conditions. By comparing the highest rated text and graphic messages under each DAC, participants then gave their preference toward either text or graphic messages. A total of 81 subjects participated in the survey. Survey results indicated that text messages were preferred at a 4:1 ratio over graphic messages, where the Zip Merge text sign messages were the least preferred of any combination of DAC and message type. The effectiveness of several top rated messages identified in the survey was further assessed through a driving simulation. Various text and graphic messages were posted on portable VMSs along a straight freeway in a fixed-base driving simulator. Subjects were asked to verbally respond with a number when they identified a message, denoting the DAC associated with that message. It was found that graphic messages were most effective in all three DACs in terms of response time and accuracy, while the Zip Merge graphic messages elicited the fastest and most accurate responses. Through regression and analysis of variance models, the questionnaire and driving simulation results show a bias of the public towards text sign messages, especially those which are currently being used in lane-reduced work zone setups, that is not supported through drivers’ responses to messages. While the survey offered the conclusion that drivers prefer to be advised by text sign messages using wording familiar to them,