Development of Real-Time Traffic Adaptive Crash Reduction Measures for the Westbound I-94/35W Commons Section

Principal Investigator:

Panos Michalopoulos, Professor, Civil, Environmental and Geo-Engineering

Project Summary:

Much research has been conducted in the development, implementation, and evaluation of innovative ITS technologies aiming to improve traffic operations and driving safety. An earlier project succeeded in supporting the hypothesis that certain traffic conditions are favorable to crashes and in developing real-time algorithms for the estimation of crash probability from detector measurements. This accomplishment naturally led to a question: how can this help prevent crashes? This project not only answered this question but also provided a multifaceted approach that offers various solutions to agencies aiming to reduce crashes in similar locations. The project's objectives including designing and visualizing different driver warning systems for the I-94 westbound high-crash location using the Minnesota Traffic Observatory's Digital Immersive Environment (DEN), and investigating the use of existing microsimulation models in the evaluation of safety improvements for the high-crash area. Although these objectives were not accomplished, the project produced a number of meaningful results and contributions. These are: 1) development of an I-94 westbound 3-D model and vehicle trajectory integration and model-building methodology; 2) development of an I-94 westbound microscopic model; 3) a historical review of car-following models, as well as identification of the research needs behind the use of simulators for safety assessment; and 4) an analysis of empirical safety-related data and an understanding of the I-94 high-crash location traffic patterns. This report describes the results of these investigations and the lessons learned during the research process, highlighting gaps of technology and knowledge that hampered this and other research projects with similar objectives.

Sponsor:

Project Details: