Penerapan Constraint-Based Scheduling Menggunakan Algoritma Hibrida GA-DPSO untuk Penjadwalan Jaga Dokter IGD di RSI Siti Hajar Sidoarjo

Bakhtiar, Lathiifah Nabiila (2026) Penerapan Constraint-Based Scheduling Menggunakan Algoritma Hibrida GA-DPSO untuk Penjadwalan Jaga Dokter IGD di RSI Siti Hajar Sidoarjo. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025221130-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Penjadwalan jaga dokter merupakan permasalahan optimasi yang kompleks karena harus memenuhi berbagai batasan operasional serta mempertimbangkan pemerataan beban kerja dan preferensi individu. Penelitian ini bertujuan mengembangkan sistem penjadwalan dokter otomatis di RSI Siti Hajar Sidoarjo menggunakan algoritma hibrida Genetic AlgorithmDiscrete Particle Swarm Optimization (GA-DPSO). Model penjadwalan menerapkan tiga hard constraint dan lima soft constraint dengan fungsi objektif berbasis penalti berbobot. Selain itu, diterapkan mekanisme repair dan formation untuk menghasilkan jadwal yang layak. Hasil pengujian menunjukkan bahwa seluruh solusi yang dihasilkan tidak mengandung pelanggaran hard constraint sehingga dinyatakan layak (feasible). Pada pengujian variasi jumlah iterasi, algoritma Hibrida GA-DPSO menghasilkan performa terbaik pada 1.000 iterasi dengan ratarata nilai fitness 123,90, rata-rata pelanggaran soft constraint 21, rata-rata nilai SSR 94,43%, dan rata-rata waktu komputasi 162,39 detik, lebih baik dibandingkan algoritma GA maupun DPSO. Hasil penelitian menunjukkan bahwa pendekatan Hibrida GA-DPSO efektif menghasilkan jadwal dokter yang layak dan mampu meningkatkan pemenuhan soft constraint, meskipun memerlukan waktu komputasi yang lebih tinggi dibandingkan penggunaan algoritma tunggal.
================================================================================================================================
Doctor shift scheduling is a complex optimization problem because it must satisfy various operational constraints while ensuring a balanced workload distribution and accommodating individual preferences. This study aims to develop an automated doctor scheduling system for RSI Siti Hajar Sidoarjo using a hybrid Genetic Algorithm–Discrete Particle Swarm Optimization (GA-DPSO) approach. The scheduling model incorporates three hard constraints and five soft constraints with a weighted penalty-based objective function. In addition, repair and formation mechanisms are implemented to generate feasible schedules. The experimental results show that all generated solutions satisfy all hard constraints and are therefore considered feasible. In the iteration variation experiments, the hybrid GA-DPSO algorithm achieved the best performance at 1,000 iterations, with an average fitness value of 123.90, an average of 21 soft constraint violations, an average Soft Constraint Satisfaction Rate (SSR) of 94.43%, and an average computational time of 162.39 seconds, outperforming both the standalone GA and DPSO algorithms. The results demonstrate that the hybrid GA-DPSO approach is effective in generating feasible doctor schedules and improving soft constraint satisfaction, although it requires higher computational time than the use of a single algorithm.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Penjadwalan Dokter IGD, Genetic Algorithm, Discrete Particle Swarm Optimization, GA-DPSO, Optimasi Hibrida, Emergency Department Doctor Scheduling, Genetic Algorithm, Discrete Particle Swarm Optimization, GA-DPSO, Hybrid Optimization.
Subjects:	T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming T Technology > T Technology (General) > T57.84 Heuristic algorithms.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Lathiifah Nabiila Bakhtiar
Date Deposited:	21 Jul 2026 02:48
Last Modified:	21 Jul 2026 02:48
URI:	http://repository.its.ac.id/id/eprint/135989

Actions (login required)

View Item