Rigorous Systems Research Group (RSRG) Seminar

I will discuss two problems in transportation network resilience. The
first problem considers network flow routing under link disruption
attacks. We formulate a security game where the defender routes flow
through a network to maximize her value of effective flow while facing
transportation costs. The attacker simultaneously disrupts one or more
links to maximize her value of lost flow but also faces cost of
disrupting links. We characterize the Nash equilibria of this game in
terms of the player valuations of effective flow and the incurred
costs, and relate their structure to min-cost max-flow and min-cut of
the network. Our results provide a measure of network vulnerability
under strategic attacks.

The second problem studies selfish routing when travelers (players)
have asymmetric access to information about incidents on network
links. We propose a Bayesian congestion game to model the
heterogeneity in the players' access to information. We characterize
the Bayesian Wardrop Equilibrium of the game, and study how the cost
to individual players and the social cost change with the fraction of
highly informed players. Our results suggest that access to
information reduces individual costs, but the relative value of
information reduces to zero if many players are highly informed.
Moreover, there exists an optimal fraction of informed players beyond
which the aggregate social costs may increase.