Simulation study of order release policies for double bottleneck systems | |
| Author | Kanyanat Arunvipas |
| Call Number | AIT Thesis no.ISE-06-05 |
| Subject(s) | Computer simulation Theory of constraints (Management) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. ISE-06-05 |
| Abstract | This study demonstrates the significance of release algorithms and different location of two bottlenecks on the production efficiency of a jobshop operating under TOC policy. A procedure is proposed for setting parameters via the Drum Buffer Rope (DBR) system. The simulation model is developed based on SSJ library in Java Language for jobshop operating under TOC policy. Various situations are created to evaluate the performance of release algorithms compare with the current system performance under the push system via simulation analysis. The bottleneck identification starts by selecting bottleneck candidates which have the two highest utilizations and the two longest queues. Drums are called bottleneck stations in TOC. Experiments are performed to identify the real bottlenecks of the system. The bottlenecks can be located in any parts of the system. Four possible locations are studied. The first case is where one bottleneck is located at the beginning and the other is at the end. The second and the third cases are where one bottleneck is at the beginning or at the end and the other is in the middle of the process. The last case is where both bottlenecks are adjacent. Buffer is represented as the waiting line in front ofthe bottleneck station. The queuing formula is used to estimate the size of time buffer. Rope is used to protect the excess work in process in the system. Two published release algorithms, Pull from Botteneck1 (PFB1) and Pull from Bottleneck2 (PFB2) are adopted as a rope for this study. During the simulation runs, statistics are collected and the performances of the system are analyzed. The throughput data can be obtained from simulation experiment is used as stopping criteria and the model stops simulating when the target demand is met. The concept of mean confidence interval comparison is applied to compare the performance of the based case with the system under various release rules. Simulation experiments are conducted with the single and multiple product systems with different location of bottlenecks. The makespan and the total WIP are represented as the key performance measures. The results demonstrate the advantage of PFB2 over PFB1 on the WIP measure. On the makespan measure, PFB2 did not give better performance than PFB1 in the single product case. For the multiple product system, based on this simulation experiment, there is no significance evidence to indicate which is better between PFB1 and PFB2 |
| Year | 2006 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. ISE-06-05 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Industrial Systems Engineering (ISE) |
| Chairperson(s) | Voratas Kachitvichyanukul; |
| Examination Committee(s) | Huynh Trung Luong;Pisut Koomsap; |
| Scholarship Donor(s) | Royal Thai Government Fellowship; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2006 |