Optimizing an agent-based traffic evacuation model using genetic algorithms

Author(s):  
Matthew Durak ◽  
Nicholas Durak ◽  
Erik D. Goodman ◽  
Robert Till
2018 ◽  
Vol 97 ◽  
pp. 82-95 ◽  
Author(s):  
Alan Poulos ◽  
Felipe Tocornal ◽  
Juan Carlos de la Llera ◽  
Judith Mitrani-Reiser

2018 ◽  
Vol 2018 ◽  
pp. 1-14 ◽  
Author(s):  
Xia-zhong Zheng ◽  
Xue-ling Xie ◽  
Dan Tian ◽  
Jian-lan Zhou ◽  
Ming Zhang

In order to analyze the evacuation capacity of parallel double running stairs, a dozen stairs merging forms are set by investigation and statistics, and the improved agent-based evacuation model that considers the merging behavior is used to simulate the process of merging and evacuation in the stairs. The stairs evacuation capacity is related to the evacuation time and the robustness of stairs, and the evacuation time can be calculated by using the improved agent-based model based on computer simulation. The robustness of each merging form can be obtained according to the fluctuation degree of evacuation time under the different pedestrian flow. The evaluation model of stairs evacuation capacity is established by fusing the evacuation time and the robustness of stairs. Combined with the specific example to calculate the evacuation capacity of each stairs form, it is found that every merging form has different evacuation time and different robustness, and the evacuation time has not positive correlation with the robustness for the same form stairs. Meanwhile, the evacuation capacity of stairs is not related to the number of the floor entrances. Finally, the results show that the evacuation capacity of stairs is optimal when the floor entrances are close to out stairs in parallel double running stairs and suitable to the case where pedestrian flow and the change of pedestrian flow are large.


Author(s):  
Sheng-Uei Guan

Agent-based system has great potential in the area of m-commerce and a lot of research has been done on making the system intelligent enough to personalize its service for users. In most systems, user-supplied keywords are normally used to generate a profile for each user. In this chapter, a design for an evolutionary ontology-based product-brokering agent for m-commerce applications has been proposed. It uses an evaluation function to represent the user’s preference instead of the usual keyword-based profile. By using genetic algorithms, the agent tries to track the user’s preferences for a particular product by tuning some of the parameters inside this function. A Java-based prototype has been implemented and the results obtained from our experiments look promising.


2020 ◽  
Vol 37 (5) ◽  
pp. 1757-1786 ◽  
Author(s):  
Jian-Ping Wang ◽  
Mei-Ru Wang ◽  
Jian-Lan Zhou ◽  
Qing-Jun Zuo ◽  
Xun-Xian Shi

Purpose The purpose of this study is to develop optimal evacuation plan to provide valuable theoretical and practical insight in the fire evacuation work of similar structures, by proposing a systematic simulation-based guided-evacuation agent-based model (GAM) and a three-stage mathematical evacuation model to investigate how to simulate, assess and improve the performance efficiency of the evacuation plan. Design/methodology/approach The authors first present the self-evacuation and guided-evacuation models to determine the optimal evacuation plan in ship chamber. Three key performance indicators are put forward to quantitatively assess the evacuation performance within the two fire scenarios. The evacuation model in tower is built to obtain the dividing points of the three different fire evacuation plans. Findings The study shows that the optimal evacuation plan determined by the GAM considering social relationships effectively relieves the congestion or collision of evacuees and improves the evacuation uniformity. The optimal evacuation plan not only solves the crush caused by congestion or collision of evacuees but also can greatly shorten the evacuation time for passenger ship fire. Originality/value This study establishes the GAM considering the interactive evacuee characteristics and the proportion of evacuees guided by the crew members to make the optimal evacuation plan more time-efficient. The self-evacuation process is simulated to assess the performance of the guided-evacuation strategies, which are used to verify the effectiveness and feasibility of the optimal evacuation plan in this research.


2012 ◽  
Vol 50 (8) ◽  
pp. 1685-1694 ◽  
Author(s):  
Nitish Chooramun ◽  
Peter J. Lawrence ◽  
Edwin R. Galea

Sign in / Sign up

Export Citation Format

Share Document