Home > News > News > 正文


Prof. Ronghui Liu at Leeds University Shared Researching Experience with SEM Faculties and Students

Publish Date: 2018/04/16 11:18:40    Hits:

Under the support of the Beihang Short-Term Project for Overseas Scholars, Prof. Ronghui Liu at Leeds University was invited to visit School of Economics and Management(SEM), Beihang University, from March 16th to April 11th, 2018. During hervisiting, Prof. Liu connected, shared and exchanged ideas with SEM facultymembers and students in transportation management, and gave two talks on thetopic,Bus Bunching,from the approach of real-time bus control (talk 1) and real-time passenger information (talk 2).

Busholding control is a real-time control strategy used to reduce bus bunching.The holding control works by keeping buses adhere to a scheduled timetableand/or to a regular headway. In her talks, Prof. Liu considers the influence of bus driving behaviour and passenger boarding behaviour on the bus holdingcontrol and on bus bunching. She proposes: (a) a distributed passenger boarding(DPB) behaviour to model the dynamic distribution of passengers, taking accountof bus bunching at stops and bus capacity constraints; and (b) afirst-depart-first-hold strategy for bus holding control. The results show thatwhen the combined overtaking and queue-swapping behaviour are considered, theproposed control strategies allowing for bus overtaking and DPB behaviour leadto better headway regularity, less passenger waiting time and less on-boardtravel time than when overtaking and DPB are not considered. Furthermore, thebenefit is even greater when travel time variability is higher and headway issmaller, suggesting that the control strategies are preferably deployed inhigh-frequency service.

Realtime information allows passengers to time their arrival to bus stop so as toreduce their wait time. The aggregate result is a peaked passenger arrivaldistribution at bus stops, which has been shown to potentially lead to moreserious bus bunching. Prof. Liu examines the trade-off between these twoeffects of real time information, and formulates an analytical model ofinformation content (in terms of a predicted bus departure times) andinformation dissemination (in terms of its update frequency and reliability). Shemodels passenger’s perception and response to information, and deriveinstantaneous arrival flows based on passengers’ risk-averse disutility to waittime and a penalty for missing the last attractive bus. The results show thatfrequent information updates and moderately fuzzy (none precise) informationcan be helpful in spreading the passenger arrival times and reducing busbunching. Crisp more accurate information may not always be the best foroverall system performance. The results have implications on the design of realtime information dissemination strategies. Finally, Prof. Liu eagerly answeredthe questions that interest SEM faculty members and students.

Dr.Ronghui Liu is an Associate Professor and the Director of InternationalActivities at the Institute for Transport Studies (ITS), University of Leeds,UK. She received her BSc from Peking University and PhD from CambridgeUniversity. Before joining ITS Leeds, she was a Research Fellow at UniversityCollege London, and while at ITS, she was seconded to head the TransportModelling Division at TRL, UK in 2005. She served as an Associate Editor forjournal IEEE Transaction on Intelligent Transportation Systems, and Member ofEditorial Board of IET Journal of Intelligent Transport. Her research areasspan a number of themes in the field of transport studies: in vehicle dynamicsand traffic microsimulaton model developments; in traffic control theory andalgorithms; in travel behaviour and Intelligent mobility; in stochastic modelsand reliability analysis; in public transport operations and controls; intimetabling and schedule coordination; and in train control and railway trafficmanagement systems. She is particularly interested in the interrelationshipsbetween these themes, and methodologies for estimation, design and evaluationof their network-wide effects.