Capacitated Refuge Assignment for Speedy and Reliable Evacuation

When a large-scale disaster occurs, each evacuee should move to an appropriate refuge in a speedy and safe manner. Most of the existing studies on the refuge assignment consider the speediness of evacuation and refuge capacity while the safety of evacuation is not taken into account. In this paper, we propose a refuge assignment scheme that considers both the speediness and safety of evacuation under the refuge capacity constraint. We first formulate the refuge assignment problem as a two-step integer linear program (ILP). Since the two-step ILP requires route candidates between evacuees and their possible refuges, we further propose a speedy and reliable route selection scheme as an extension of the existing route selection scheme. Through numerical results using the actual data of Arako district of Nagoya city in Japan, we show that the proposed scheme can improve the average route reliability among evacuees by 13.6% while suppressing the increase of the average route length among evacuees by 7.3%, compared with the distance-based route selection and refuge assignment. In addition, we also reveal that the current refuge capacity is not enough to support speedy and reliable evacuation for the residents.

En-route headway-based bus reliability with real-time data at network scale

Bus reliability has long attracted attention and been extensively studied to enhance service quality. However, existing research generally evaluates bus reliability of specific routes or stops. To this end, this study explores en-route bus reliability with real-time data at network scale. Drawing on data of bus automatic vehicle location and smart card usage in Ningbo, China, this study calculates headway-based reliability with the difference between actual and scheduled headway at each stop. To demonstrate the trend of stop-level reliability along a bus route, reliability is graded and visualized on a map with ridership at each stop, which is then weighted with passenger-boarding volume. Route-level reliability is then quantified and mapped, where unreliable service basically concentrates in or extends through the centre area. With respect to network-level reliability, temporal changes are demonstrated with ridership on weekdays and at the weekend. It is observed that on weekdays, the reliability trend is similar to that of ridership, implying a causal relationship between bus travel-time variation and bus waiting-time at stops. Furthermore, a reliability comparison between weekdays in December and October shows the necessity of evaluating periodically and around important events to avoid negative riding experiences that discourage public transport usage. This research provides insights for bus agencies to systematically evaluate service reliability both spatially and temporarily, in order to identify and prioritize the routes and stops where the scope for reliability improvement and the expected benefit are greatest.