TY - JOUR
T1 - Hybrid cyclicity
T2 - Combining the benefits of cyclic and non-cyclic timetables
AU - Robenek, Tomáš
AU - Sharif Azadeh, Shadi
AU - Maknoon, Yousef
AU - Bierlaire, Michel
PY - 2017
Y1 - 2017
N2 - We propose a new type of timetable that would combine both the regularity of the cyclic timetables and the flexibility of the non-cyclic ones. In order to do so, several combinations of the two timetables are considered. The regularity is incorporated in their design and the flexibility is evaluated using the passenger satisfaction (in monetary units). Each of the tested timetables is constructed using the Passenger Centric Train Timetabling Problem (PCTTP), that is solved using a simulated annealing heuristic. Note that the PCTTP, unlike the traditional Train Timetabling Problem (TTP), does not take into account the conflicts among trains. The aim of the PCTTP is to design such timetables that the passengers’ satisfaction is maximized and it remains the aim of the TTP to remove any potential conflicts. The performance of each of the considered timetables is assessed on the real network of Israeli Railways. The results of the case study show that our proposed hybrid cyclic timetable can provide the benefits of the cyclic and the non-cyclic timetable simultaneously. This timetable consists of 75% of cyclic trains (securing the regularity of the service) and of 25% of non-cyclic trains (deployed as supplementary trains during the peak hours and capturing the demand fluctuation). The level of the passenger satisfaction of the hybrid cyclic timetable is similar to the level of the non-cyclic one, which has about 18.5% of improvement as compared to the purely cyclic one.
AB - We propose a new type of timetable that would combine both the regularity of the cyclic timetables and the flexibility of the non-cyclic ones. In order to do so, several combinations of the two timetables are considered. The regularity is incorporated in their design and the flexibility is evaluated using the passenger satisfaction (in monetary units). Each of the tested timetables is constructed using the Passenger Centric Train Timetabling Problem (PCTTP), that is solved using a simulated annealing heuristic. Note that the PCTTP, unlike the traditional Train Timetabling Problem (TTP), does not take into account the conflicts among trains. The aim of the PCTTP is to design such timetables that the passengers’ satisfaction is maximized and it remains the aim of the TTP to remove any potential conflicts. The performance of each of the considered timetables is assessed on the real network of Israeli Railways. The results of the case study show that our proposed hybrid cyclic timetable can provide the benefits of the cyclic and the non-cyclic timetable simultaneously. This timetable consists of 75% of cyclic trains (securing the regularity of the service) and of 25% of non-cyclic trains (deployed as supplementary trains during the peak hours and capturing the demand fluctuation). The level of the passenger satisfaction of the hybrid cyclic timetable is similar to the level of the non-cyclic one, which has about 18.5% of improvement as compared to the purely cyclic one.
KW - Cyclicity
KW - Passenger centric train timetabling problem
KW - Passenger satisfaction
KW - Railway timetable
KW - Simulated annealing
UR - http://www.scopus.com/inward/record.url?scp=85007610457&partnerID=8YFLogxK
U2 - 10.1016/j.trc.2016.12.015
DO - 10.1016/j.trc.2016.12.015
M3 - Article
AN - SCOPUS:85007610457
SN - 0968-090X
VL - 75
SP - 228
EP - 253
JO - Transportation Research. Part C: Emerging Technologies
JF - Transportation Research. Part C: Emerging Technologies
ER -