• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.23 no.54
• /
• pp.77-85
• /
• 2000

One form of job shop scheduling problem in contemporary automated manufacturing such as flexible manufacturing systems (FMS's) is presented which we call the FMS dispatching priority problem. The FMS dispatching priority problem seeks the best dispatching priority of parts and operations, and is essentially a combinatorial optimization problem. Because of the complicated mechanism of the system, the performance of a given dispatching priority must be evaluated via simulation. Simulated annealing have been applied to the problem, and it is found that appropriate parameter setting will be desirable to get good, if not the optimal, solutions within a limited amount of time under the presence of heavy computational burden due to simulation. More specifically, experiments reveal that initial temperature is the single most important factor among other parameters and factors, and that the appropriate initial temperature depends on the allowable computer time in such a way that the less time one can afford to spend, the lower the appropriate initial temperature should be.

• IE interfaces
• /
• v.10 no.2
• /
• pp.91-97
• /
• 1997

A mass production system was implemented to reduce a manufacturing cost in a way of copying with a strong world market competition. However customer's demands are changing so rapidly and the mass production system is no longer competitive to meet the demands. FMS(Flexible Manufacturing System) has been introduced as a replacement for the mass production system, but it still does not meet system's requirements. A new manufacturing system, called a holonic manufacturing system (HMS), is emerging. This paper is giving a first approach of a HMS for a multirobot cooperative process.

• Journal of Korean Institute of Industrial Engineers
• /
• v.21 no.1
• /
• pp.137-151
• /
• 1995

A difficult problem in operating Flexible Manufacturing Systems (FMS) is to control the system in real-time by coordinating heterogeneous machines and integrating distributed information. The objective of the paper is to present the models and methodologies useful to resolve the difficult problem. The detailed objectives can be described in three folds. First, a hierarchical Colored and Timed Petri-Net (CTPN) is designed to control an FMS in real-time. The concerned FMS consists of a loading station, several machining cells, a material handling system, and an unloading station. Timed-transitions are used to represent the timed-events such as AGV movements between stations and cells, part machining activities in the cells. Signal places are also used to represent communication status between the host and the cell controllers. To resolve the event conflicts and scheduling problems, dispatching rules are introduced and applied. Second, an implementation methodology used to monitor and diagnose the errors occurring on the machines during system operation is proposed. Third, a Petri-Net simulator is developed to experiment with the designed control logic.

• Journal of Korean Institute of Industrial Engineers
• /
• v.21 no.2
• /
• pp.153-165
• /
• 1995

Many closed queueing network models have been used as an effective tool to make a preliminary design of FMS. Although a loading/unloading function is an important factor to effect the utilization and throughput of FMS as well as a transfer function and a processing function, the Solberg's model did not clarify the loading/unloading function. In this paper, we propose a closed queueing network model for analyzing a flexible manufacturing system that consists of a load/unload station, a material handling system and a group of workstations for processing jobs, and define the expected utilization and the expected throughput of the FMS. As applications of the proposed model, the cases of a material handling system consisting of a conveyor and the FMS including an inspection station are also formulated.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.22 no.50
• /
• pp.127-137
• /
• 1999

This study is concerned with the loading problems in flexible manufacturing system(FMS). The loading problem in FMS is a complex one, when the number of machine and job is increased. It may be time-consuming and even impossible to achieve an optimal solution about this problem mathematically. Thus, a heuristic method is recommended in order to gain near-optimal solutions in a practically acceptable time. A new loading algorithm is developed with a multi-criterion objective of considering the workload unbalance, and maximizing the machine utilization, throughput for critical resources such as the number of tool slots and the number of working hours in a scheduling period and so on. The results of SAS analysis indicated that true average throughput of proposed heuristic loading statistically exceeds that of Shanker and Srinivasulus loading algorithm at the significance level of 0.1.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.24 no.67
• /
• pp.19-26
• /
• 2001

A design issue for a flexible manufacturing system is to find the number of each resource type which assures a given production ratio with a minimal cost. The FMS is modeled as a closed queueing network with a local buffer of limited capacity. An efficient method to determine this optical configuration is presented. The proposed method consists of three steps : 1) determine a lower configuration, 2) derive a heuristic solution, 3) find the optimal solution. The derived heuristic solutions are close to the optical solution.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.18 no.33
• /
• pp.79-86
• /
• 1995

This paper intends to develope the multi-objective loading model in a flexible manufacturing system (FMS) to support decision maker under fuzzy environment. To obtain the optimal solution, this paper uses interactive fuzzy multi-objective linear programing(IFMOLP) and describes the process of optimal solution. As a case study, numerical examples are demonstrated to show the effectiveness of the proposed model.

• The Magazine of the IEIE
• /
• v.22 no.8
• /
• pp.86-97
• /
• 1995

Computer simulation is a powerful tool for the design and evaluation of manufacturing systems. This paper consists of two major parts. First, it presents a brief survey of simulation languages oriented towards the analysis of manufacturing systems, with their important properties. Second, it introduces a simple case of simulation application to a flexible manufacturing system(FMS).

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10a
• /
• pp.572-574
• /
• 1999

공장자동화는 최근 급격히 발전하고 있는 분야로서 부가가치가 매우 높은 산업에 속한다. 공장자동화는 자동화를 구성하는 기계장치뿐만 아니라 기계의 전반적인 운영을 담당하는 소프트웨어 또한 큰 비중을 차지하고 있다. 결국, 공장자동화를 통해 생산되는 제품의 품질은 기계 장치의 정밀도, 견고성 등의 측면과 함께 운영 소프트웨어의 품질로부터 받는 영향도 무시할 수 없다. 본 연구 과제에서는 공장자동화 소프트웨어의 한 유형인 FMS (Flexible Manufacturing System)나 Cell Controller를 중심으로 품질평가를 수행할 수 있는 평가 메트릭과 방법을 개발하고 이를 적용하여 효율적인 평가를 수행할 수 있는 툴킷의 프로토타입을 설계하고 구현하였다.

• Journal of the Korea Society for Simulation
• /
• v.10 no.4
• /
• pp.65-75
• /
• 2001

This study presents a heuristic algorithm for buffer allocation in FMS(Flexible Manufacturing System). The buffers, which are finite resources in FMS, are responsible for improvement of an overall system utilization. But, until now, the study for buffer allocation in FMS are rarely conducted because of the complexity in FMS. Most studies for buffer allocation had been addressed to the simple production line system. The presented algorithm uses a simulation for the description of system complexity and uses a genetic algorithm for finding better buffer allocation. Lastly, we compare performance of the presented algorithm with that of a simple heuristic, and analyze the experiment results.