Proceedings of the 1st Workshop on Multidisciplinary and Its Applications Part 1, WMA-01 2018, 19-20 January 2018, Aceh, Indonesia

Research Article

Genetic Algorithm with Baker’s SUS Selection for Shift Scheduling of the Security Officers at Rawa Buntu Train Station, Indonesia

Download443 downloads
  • @INPROCEEDINGS{10.4108/eai.20-1-2018.2281891,
        author={Aghistina  Kartikadewi and Achmad  Solichin},
        title={Genetic Algorithm with Baker’s SUS Selection for Shift Scheduling of the Security Officers at Rawa Buntu Train Station, Indonesia},
        proceedings={Proceedings of the 1st Workshop on Multidisciplinary and Its Applications Part 1, WMA-01 2018, 19-20 January 2018, Aceh, Indonesia},
        publisher={EAI},
        proceedings_a={WMA-1},
        year={2019},
        month={9},
        keywords={transportation shift scheduling ga stochastic universal sampling},
        doi={10.4108/eai.20-1-2018.2281891}
    }
    
  • Aghistina Kartikadewi
    Achmad Solichin
    Year: 2019
    Genetic Algorithm with Baker’s SUS Selection for Shift Scheduling of the Security Officers at Rawa Buntu Train Station, Indonesia
    WMA-1
    EAI
    DOI: 10.4108/eai.20-1-2018.2281891
Aghistina Kartikadewi1,*, Achmad Solichin2
  • 1: Post Graduate of Computer Science, Universitas Budi Luhur, Jakarta, Indonesia
  • 2: Faculty of Information Technology, Universitas Budi Luhur, Jakarta, Indonesia
*Contact email: aghistinakd@gmail.com

Abstract

The increase of railway passengers in Indonesia should be balanced with security and convenience, especially at departure and train arrival stations. In ensuring passenger safety, PT Kereta Api Indonesia (KAI) provides some security officers at each railway station, including the Rawa Buntu Station. The Station Head and the Security Commander are responsible for drafting a guard schedule for 30 security officers at the station. The security schedule is divided into three shifts and eight guard locations. Manually arranged schedules often result in unbalanced work hours, clashing schedules and allowing emptiness of security officers. It is getting harder with additional rules that whenever a security guard gets a night shift, the next day must be off. In this research, we proposed a shift scheduling method based on a genetic algorithm using Stochastic Universal Sampling (SUS) selection method, double point crossover, random mutation and using generational model (elitism) for its generation update process. This research produces an application that can assist in the preparation of automatic shift scheduling of the security officers. Based on the test results, the application can arrange the schedule with the suitability level of 70% and the average time needed to generate schedule is 72.48 seconds. With this research, the process of arranging the schedule of security officers at the Rawa Buntu Station to be faster.