
A Threshold Policy for Dispatching Vehicles in Demand-responsive Transit Systems
Author(s) -
Nikola Marković,
Myungseob Kim,
Eungcheol Kim,
Sanjin Milinković
Publication year - 2019
Publication title -
promet
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.315
H-Index - 19
eISSN - 1848-4069
pISSN - 0353-5320
DOI - 10.7307/ptt.v31i4.3027
Subject(s) - mode (computer interface) , transit (satellite) , computer science , function (biology) , work (physics) , operations research , transit system , real time computing , mathematical optimization , transport engineering , simulation , engineering , public transport , mathematics , mechanical engineering , evolutionary biology , biology , operating system
This paper considers vehicle dispatching for a flexible transit system providing doorstep services from a terminal. The problem is tackled with an easy-to-implement threshold policy, where an available vehicle is dispatched when the number of boarded passengers reaches or exceeds a certain threshold. A simulation-based approach is applied to find the threshold that minimizes the expected system-wide cost. Results show that the optimal threshold is a function of demand, which is commonly stochastic and time-varying. Consequently, the dispatching threshold should be adjusted for different times of the day. In addition, the simulation-based approach is used to simultaneously adjust dispatching threshold and fleet size. The proposed approach is the first work to analyse threshold dispatching policy. It could be used to help improve efficiency of flexible transit systems, and thereby make this sustainable travel mode more economical and appealing to users.