Page 48 - Conference Book The 3rd IConVET "Future of TVET Graduates: Developing Talent for Industry 4.0 and The New Normal"

The 3 International Conference
rd
on Vocational Education and Technology (IConVET)
th
Universitas Pendidikan Ganesha - Bali | 7 November, 2020
MODIFIED GENETIC ALGORITHM FOR EMPLOYEE
WORK SHIFTS SCHEDULING OPTIMIZATION

2
1
N W S Saraswati , I D M D Artakusuma and I G A A D Indradewi 3

1,2,3 Study Program of Informatics Engineering, STMIK STIKOM
Indonesia, Denpasar-Bali 80225, Indonesia

2
1
sumartini.saraswati@stiki-indonesia.ac.id , dewadwi@hotmail.com ,
didyrakata@gmail.com 3

Abstract. Arranging an employee shift work’s schedule requires high
accuracy. It is because we have to pay attention to several constraints
simultaneously. The genetic algorithm presents as a method which can
automatize the process of arranging the schedule as well as optimizing the
result of the schedule. The Shala Bali is a hospitality business which has
employed dozens of employees, thus scheduling the shift work of the
employee is something complex. This research aimed to produce a shift
work schedule of the employees in a week and to know the optimum genetic
algorithm parameter in this case. The constraints that are taken into account
in the arrangement of the schedule include the schedule conflict of the
employees in one shift, schedule conflict of employees in 1 day, the same
composition of employees per shift, employees may not get morning shifts
after previously getting a night shift, each shift has at least 1 employee in the
front office, and each employee is required to get 1 day off within 1 schedule
period. This study was able to produce an optimal work schedule of
employees with crossover probability (Pc) of 0.6, and mutation probability
(Pm) of 0.3. The modification algorithm in chromosome generation and
chromosome structure in this study results that changes in gene length
(additional number of employees) do not have to be followed by an increase
in the number of chromosome populations to get optimum results.

| IConVET 2020 | 48

https://conference.undiksha.ac.id/iconvet/ | 48 |