Restoration of damaged highway-bridge networks is critical to the recovery of the networks’ functionality in support of post-disaster recovery actions. Restoration-scheduling problems have traditionally assumed that parallel restoration actions to different bridges on a highway segment can be undertaken simultaneously. However, such assumption may have unfavorable consequences insofar as, in real-world scenarios, downtime may result from impassability of a segment due to blockage arising from the preceding restoration actions to their subsequent ones on the same segment. Consequently, downtime due to interactions of parallel restoration actions may hinder effective recovery of networks’ functionality. Accordingly, this paper proposes an integer program to investigate the restoration interactions in post-earthquake highway–bridge networks and such interactions’ impacts on optimal restoration scheduling and networks’ resilience recovery processes. A hybrid genetic algorithm that combines a genetic algorithm with a specifically designed heuristic approach is developed to improve the proposed integer program’s computational efficiency. The results of a case study using the proposed method and data from the 2008 Wenchuan Earthquake show that the downtime due to restoration interactions can delay the recovery of the networks’ functionality, and thus that neglecting such interactions can lead to the overestimation of the networks’ resilience.

Co-worker safety support has been given prominence in manufacturing and transportation field for its positive effect on individual workers’ safety; however, there is little evidence to show if such supporting role of co-workers is significant in improving project-level safety performance in construction workplace. This study adopts agent-based modeling (ABM) to understand the effectiveness of two distinct co-worker-safety-support actions on the safety performance of a construction project. Based on the risk theory, the ABM model simulates a construction site where worker agents reinforce steel bars with the likelihood of suffering crane-related incidents. The results indicate that both co-worker-support actions can significantly reduce the occurrence of nonfatal incidents but shows little influence in fatal incidents, and in reducing high-severity incidents, the action of warning peers to leave the hazardous area has the same effectiveness as reminding peers to wear Personal Protective Equipment. The present study provides a fresh insight into the safety-related role of co-workers: not only reveals how the local-level effects of co-workers’ safety assistance emerge the system-level consequences, but demonstrates the effectiveness of specific peer-support actions on three levels of construction safety performance, and thereby extends our existing body of knowledge on co-worker safety support in the construction field.