AI & ML Paradigm Challenge

Adding more variety to human behavior actually makes traffic jams harder to predict and solve.

April 16, 2026

Original Paper

Departure Time Choice with Parametric Heterogeneity: Equilibrium and Instability

Hillel Bar-Gera, Stephen D. Boyles, Liron Ravner

arXiv · 2604.13831

The Takeaway

Traffic engineers have long assumed that if they could just account for the 'messiness' of human variety—different work schedules and preferences—their models would stabilize. This paper proves the mathematical opposite: adding more 'human heterogeneity' actually makes the system more unstable. The learning dynamics of how we choose when to leave for work are fundamentally chaotic, meaning we never actually reach a 'stable' traffic equilibrium. This suggests that the quest for a perfectly smooth commute is a pipe dream because the more we adapt to each other, the more the system wobbles. For you, it means that new lanes or smarter apps might never fix your commute, because the math of human choice is rigged toward instability.

From the abstract

Vickrey's classic single-bottleneck departure time choice equilibrium model exhibits instability under many plausible day-to-day learning dynamics. Such instability is not observed in reality -- does this difference stem from the day-to-day dynamics or from one of the simplifying assumptions of the basic model? This paper explores a variant of the basic model with a continuous distribution of schedule delay parameters which we intuitively expect to have more favorable stability properties. To at