DEVELOPMENT OF REACTIVE SCHEDULING MODEL WITH ROBUSTNESS PERFORMANCE CRITERIA TO MINIMIZE MACHINE BREAKDOWN
The pre-planned initial schedule when executed to the production floor had poor performance. This is because the initial schedule does not take into account unexpected disturbance such as machine breakdown so that schedule improvements are needed to make it more feasible to execute. This study tries...
Saved in:
Main Author: | |
---|---|
Format: | Theses |
Language: | Indonesia |
Subjects: | |
Online Access: | https://digilib.itb.ac.id/gdl/view/56928 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Institut Teknologi Bandung |
Language: | Indonesia |
Summary: | The pre-planned initial schedule when executed to the production floor had poor performance. This is because the initial schedule does not take into account unexpected disturbance such as machine breakdown so that schedule improvements are needed to make it more feasible to execute. This study tries to develop a scheduling model on a single machine that is robust against machine breakdowns using a predictive-reactive rescheduling approach. The purpose of this study is to obtain a job sequence that has a minimal total tardiness value and is robust against machine breakdowns. Robustness performance criteria can be obtained by considering the performance criteria of efficiency and stability simultaneously.
The predictive stage in the study produces an initial schedule that only considers the objective function of total tardiness while the reactive stage produces an actual schedule as a response to machine breakdowns that occur by considering the performance criteria of total tardiness and expected delay which are combined linearly. Event-driven policies and right shift scheduling are used as a method of improving the schedule in the reactive stage. event-driven policy changes the job sequences whenever a machine breakdown occurs and right shift scheduling maintains the job sequences on the initial schedule by shifting unfinished jobs to the right in amount of repair durations. Multi-Offspring Genetic Algorithm is used to generate the initial schedule and the actual schedule with the smallest performance criteria in the solution space. The simulation results show that if the frequency of engine failure increases, then the Multi-Offspring Genetic Algorithm with event-driven policy produces a schedule with lower robustness performance criteria, while if the frequency of engine failure is low, even driven and right shift scheduling policies produce robustness performance criteria almost the same. |
---|