Solving late for relief event in bus crew rescheduling using multi agent system

Ahmad Fadzli Nizam Abdul Rahman, Abdul Samad Shibghatullah*, Zuraida Abal Abas, Tillal Eldabi, Chit Su Mon, Amir Aatieff Amir Hussin

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


Unpredictable event is an event which happens anytime without notice that will disrupt bus services. Bus crew is one of the causes of the unpredictable event as if a bus crew comes late - s/he will cause certain bus to depart late. In this paper, three types of bus crew lateness are defined which is Late For Sign-On (LFSO), Late For Relief (LFR), and Late For Second Work (LFSW). However, this paper will only discuss the solution for LFR type. When LFR happens, the schedule needs to be rescheduled. Currently, there is no automated mechanism to handle LFR issue especially in Internet of Thing (IoT) environment. Most real time rescheduling approaches are not supported due to static schedules constraint. Mathematical approaches require extensive computational power, therefore delaying real-time results. Manual rescheduling by supervisor is likely to have an errors and not an optimize solution. This paper presents a new approach for rescheduling the bus crew's timetable in the event of LFR. The multi agent system will adapt quickly to the dynamic environments to find the best and optimize solutions. The experiment of LFR is conducted by using the AgentPower simulation tool. The result concluded that the proposed technique can produce quick rescheduling the for bus crew schedule in the event of LFR.

Original languageEnglish
Pages (from-to)1972-1983
Number of pages12
JournalJournal of Engineering Science and Technology
Issue number3
Publication statusPublished - Jun 2020


  • Bus crew scheduling
  • Late for relief
  • Multi-agent system
  • Unpredictable event

ASJC Scopus subject areas

  • Engineering(all)


Dive into the research topics of 'Solving late for relief event in bus crew rescheduling using multi agent system'. Together they form a unique fingerprint.

Cite this