Probabilistic post-earthquake restoration process with repair prioritization of highway network system for disaster resilience enhancement

Comprehensive realization of post-earthquake restoration process for highway network considering repair prioritization are necessary for preparing effective countermeasures to restore transportation service and social activity in seismically active and automobile dependent regions like broader Los Angeles area quickly. Since the progress and the time to achieve full-recovery realistically change depending on repair orders and reconstruction works, restoration strategies without taking into account actual restoration phenomena negatively affect decision-making, efficient recovery, damage mitigation effort and network resilience enhancement.For dealing with these concerns, a probabilistic model to simulate post-earthquake restoration process of highway network is developed with consideration of repair prioritization and reconstruction constraint in this research. Analytic Hierarchy Process, prioritizing algorithm, is used to decide bridge repair orders based on initial damage state due to earthquake and priority weight of four criteria; difficulty, importance, urgency and cost. Numbers of available construction labor around the target region is applied for constraint of repair work progress. The recovery passage of bridge, the critical component of highway restoration process, is modeled probabilistically based on Markov Chain process, Uniform distribution and Normal distribution. As performance of highway network, drivers' delay and trip opportunity loss over entire recovery period are estimated by conducting traffic network analysis through origin-destination matrix, gravity model and user equilibrium model considering the models of trip reduction and traffic demand recovery. The adequacy of developed model is then verified by the documented recovery records and loss estimations of Northridge earthquake.As an application study, a regional possible scenario earthquake is applied to the established methodology implemented in the highway network system of Los Angeles and Orange counties. A number of simulations through Monte Carlo technique to express restoration processes corresponding to several repair prioritizations are presented by restoration curves and loss estimations. The analyzed results show that the developed procedure can simulate numerous thinkable recovery scenarios according to repair orders and contribute to decision-making for choosing the best suited repair prioritization for minimizing loss and maximizing resilience. Moreover, the basic scheme of this innovative technique can be applied to evaluation of restoration process of other infrastructure network systems and other disasters