(313) 593-5274

Applications:
Behavioral Science, Transportation Research
Methodologies:
Causal Inference, Data Visualization, Human-Computer Interaction, Machine Learning, Predictive Modeling, Statistical Inference
Relevant Projects:

An examination of drivers overtaking bicyclists using naturalistic driving data. A naturalistic bicycling study in the Ann Arbor area


Connections:

Human Factors and Ergonomics Society

Fred Feng

Assistant Professor

Industrial and Manufacturing Systems Engineering, University of Michigan-Dearborn

Dr. Feng’s research involves conducting and using naturalistic observational studies to better understand the interactions between motorists and other road users including bicyclists and pedestrians. The goal is to use an evidence-based, data-driven approach that improves bicycling and walking safety and ultimately makes them viable mobility options. A naturalistic study is a valuable and unique research method that provides continuous, high-time-resolution, rich, and objective data about how people drive/ride/walk for their everyday trips in the real world. It also faces challenges from the sheer volume of the data, and as with all observational studies, there are potential confounding factors compared to a randomized laboratory experiment. Data analytic methods can be developed to interpret the behavioral data, make meaningful inferences, and get actionable insights.

9.9.2020 MIDAS Faculty Research Pitch Video.

Using naturalistic driving data to examine the interactions between motorists and bicyclists