Event graph modeling of a homogeneous job shop with bi-inline cells

doi:10.1016/j.simpat.2011.08.002

Simulation Modelling Practice and Theory

Volume 20, Issue 1, January 2012, Pages 1-11

https://doi.org/10.1016/j.simpat.2011.08.002 Get rights and content

Abstract

In modern electronics Fabs, most production lines consist of inline cells that have processing machines along the conveyor line inside. In this type of Fab, a large number of different products are loaded concurrently, which requires a production simulator to analyze the planning and scheduling of the Fab operations. In order to build such a simulator, (1) a typical TFT-LCD Fab is simplified into a job shop consisting only of bi-inline cells and (2) the job shop is represented as an event graph model. This type of job shop consisting only of machines of the same type is referred to as homogeneous job shop. The event graph model was verified using SIGMA and sample input data. However, if a production line contains uni-inline cells as well as bi-inline cells, we need a heterogeneous job shop model and this requires further study.

Introduction

In modern electronics Fabs such as TFT-LCD Fabs, most production systems consist of inline cells, which we call an inline production system [1]. Unlike a simple table-type machine which processes a part at a time, an inline cell consists of an inner conveyor which carries the parts through the equipment and multiple subordinate processing machines along the conveyor belt [2]. Inline cells are connected by an automated material handling system (AMHS) such as inline stockers and conveyors [3]. In TFT-LCD Fabs, the glasses are stored in a container called cassette while they travel among the inline cells. Once a cassette full of glasses arrives at the ports in front of an inline cell, a robot moves the glasses one by one onto the conveyor belt. Similarly, the processed glasses are unloaded to an empty cassette by another robot. In this type of Fab, a large number of different parts are processed simultaneously in the same line. When the processing requirements are different from one part type to another, the inline cell needs a set-up before it starts to process a part of different type. The more this kind of part-type change occurs, the longer the idle time of the inline cell becomes. As a result, there is an increasing demand for a simulator that can analyze the efficiency of planning and scheduling of inline production system [4].

Many commercial simulation packages have been introduced to simulate an inline production system: ASAP (AutoSched AP™) is one of the most popular tools for production simulation [5], [6], [7]. However, a simulator implemented with a commercial package generally has less flexibility than the one implemented based on a well-defined formal model. This paper presents a systematic method of building a formal event graph model that can be implemented into a simulator of an inline production system.

The inline cells in TFT-LCD Fabs are largely classified into two groups: bi-inline cells and uni-inline cells. This classification is based on whether the loading ports and the unloading ports of the cell are separated or not. A bi-inline cell has two types of separated ports: in-ports for loading new glasses from a cassette and out-ports for unloading finished glasses into a cassette. It has two glass-handling robots: a track-in robot at the in-port section and a track-out robot at the out-port section.

In this paper, an event graph model of a homogeneous job shop with bi-inline cells is proposed. To verify the model, the commercial simulation package SIGMA® is used for the implementation and testing of the model [8]. SIGMA is the most widely used simulation tool to build, test, animate and experiment event graph models. The formal model provided in this paper may be considered as a starting point for building a production simulator for performance analysis of a target TFT-LCD Fab. In uni-inline cells, on the other hand, in-ports and out-ports are not separated and the loading and unloading of the glasses are handled by one robot. A feature of uni-inline cell is that the glasses are unloaded to the cassette where they were contained at the arrival to the cell. (In bi-inline cell, processed glasses are unloaded to a new cassette.) An event graph model of homogeneous job shop with uni-inline cells has already been suggested in a previous work [2]. In order to make the paper self-contained, the result of our previous work is summarized in Appendix A.

A hypothetical example of modern TFT-LCD Fab layout is depicted in Fig. 1. As shown in the figure, a uni-inline cell is connected to one inline stocker through in/out-ports, whereas a bi-inline cell has separate in-ports and out-ports. In-port and out-port of a bi-inline cell are often connected to different inline stockers. If the material handling processes are simplified, this Fab can be viewed as a typical job shop consisting of inline cells.

Section snippets

Formal representation of job shop

General job-shop models are used to represent manufacturing systems as well as service systems in various formalisms. A formal model may be described either using an algebraic syntax or in a well defined graph form. An event graph is a formal model which is quite simple yet extremely powerful and natural to represent a discrete-event system [9]. In event graphs, the influence of events on the state variables is represented by vertices and the relationships between events are represented as

Reference model of bi-inline cell

In the job shop model, the objects included in the system can be divided into two categories: resident entities and transient entities [14]. A resident entity such as a server or machine resides in the system, whereas a transient entity such as a customer or part constantly enters and leaves the system after being served by resident entities [8]. In the job shop model presented in this paper, bi-inline cells are resident entities and the cassettes arriving in the queue of a cell are transient

Cassette loading section modeling

Fig. 3 provides a detailed illustration of events occurring at the cassette-loading section of the bi-inline cell presented in Fig. 2. There are four events defined in this section:

1.
Queue: A cassette arrives in the queue of the bi-inline cell.
2.
X2PI: The cassette moves to the in-port of the bi-inline cell.
3.
FGL: start of First-Glass-Loading (including the set-up time).
4.
LGL: end of Last-Glass-Loading (the robot finishes loading the last glass in a cassette).

The state variables used in this section are

Implementation with SIGMA

In this section, the event graph model of the bi-inline cell implemented with SIGMA is presented and the results of virtual TFT-LCD Fab with sample input data are illustrated in Fig. 8 for verification. This virtual line manufactures two types of products: 4-mask and 5-mask (i.e., gate, active, source-drain, passivation and pixel layers) with different processing steps (i.e., cleaning, deposition, photolithography, and etching) with eight bi-inline cells, and two identical machines for each

Conclusions

Presented in the paper is a systematic procedure for constructing an event graph model of a homogeneous job shop consisting of bi-inline cells. First, the bi-inline cell is divided into two sections, cassette loading section and cassette unloading section, to construct an event graph model for each section. Secondly, these two event graph models are combined to form an event graph model of the entire bi-inline cell. Finally, the event graph model of the bi-inline cell is parameterized to

Acknowledgements

The authors would like to thank Joo-Young Lim for her precious help on building a SIGMA model of the parameterized event graph model of Fig. 7b.

References (14)

J.R. Morrison et al.
Closed reenterant queueing networks under affine index policies: throughput bounds, examples and asymptotic loss
R. Tavakkoli-Moghaddam et al.
A computer simulation model for job shop scheduling problems minimizing makespan
Computers & Industrial Engineering
(2005)
J.H. Lee, A Study on Equipment Modeling Method for Simulation Based Semiconductor Fab Scheduling, M.S. Thesis,...
E. Song, S. Gu, B.K. Choi, Event graph modeling of electronics Fab with uni-inline cells, in: Proceedings of the 2010...
Y.J. Jang et al.
Introduction to automated material handling system in LCD panel production lines
B.C. Park et al.
Simulation based planning and scheduling system for TFT-LCD Fab
M. Pool et al.
Productivity modeling of semiconductor manufacturing equipment

There are more references available in the full text version of this article.

Cited by (9)

A method for the spatiotemporal correlation prediction of the quality of multiple operational processes based on S-GGRU
2023, Advanced Engineering Informatics
Predicting quality indicators is essential in process production, process production often involves multiple processes. However, modeling the prediction model for quality indicators faces challenges due to frequent fluctuations in raw material composition and the control system's automatic adjustment of process parameters. Additionally, due to the mutual influence between processes, the problem of predicting quality indicators in process production is a systematic prediction problem. In order to deal with the challenges, firstly, the physical control relationships are transformed into graph data, which is input into GNN to calculate the spatial characteristics of process parameters on quality indicators under physical information constraints. GRU captures the temporal features of quality indicator data. Based on above study, a combination network GGRU integrating quality indicators' spatial and temporal characteristics was proposed. Finally, each GGRU is then connected based on the material flow between each process, and path information between processes is converted to the calculation flow between GGRU models to form the S-GGRU prediction model. Experimental results show that the proposed S-GGRU model, a multi-process prediction model based on GRU and GNN, outperforms the recurrent neural network model and the graphical neural network model in solving the quality prediction problem for multi-process and multi-metrics in-process production. Therefore, both the time-series information of quality metrics and the influencing factors among variables should be considered to achieve better predictions, and the S-GGRU performance better in predicting quality indicators for a 5-process process line.
Object-oriented event-graph modeling formalism to simulate manufacturing systems in the Industry 4.0 era
2020, Simulation Modelling Practice and Theory
Citation Excerpt :
To fulfill this task they developed a DES library, SharpSim, using C# (CSharp) programming language. Event Graphs have been used to model manufacturing systems, such as flat panel display (FPD) production lines [14, 15], applications on Internet, such as World Wide Web (WWW) and File Transfer Protocol (FTP)[16], and to model the business processes of a credit card application [17]. Other EGs templates and examples related to various manufacturing systems can be found in Choi and Kang [1].
In the event-based modeling formalism a system is modeled by defining the changes that occur at event times and the system dynamics can be described using an ‘Event Graph’. An event graph model is a network of event nodes describing the events that take place in the system and the relationships among these events. In the paper we extend the Event Graph framework by proposing an event-graph modeling formalism suitable to represent discrete event simulation models developed through the object-oriented approach, named Object-Oriented Event Graph (OOEG). The importance of Object–Oriented simulation is recently growing under the Industry 4.0 paradigm, in which optimization via simulation, real-time simulation, automatic decisions systems based on simulation, on line scenario analysis play a relevant role. The OOEG representation provides the same compactness of the EG representation. The advantage of the OOEG representation is that allows supporting a modelling methodology where systems are described by linking components analogously as the system components are linked, and the dynamic of the systems can be visualized in terms of interactions between objects, which have their physical correspondence in the real world.
Simulation for manufacturing system design and operation: Literature review and analysis
2014, Journal of Manufacturing Systems
Citation Excerpt :
In their work, they also illustrate the use of CMSD to integrate real-world manufacturing applications through case studies. By simplifying a typical TFT-LCD fab into a homogeneous job shop consisting of only bi-inline cells, Song et al. [253] propose and verify an event graph model of the fab that can be considered as a starting point for developing simulators for performance analysis in such environments. Garg et al. [254] propose generic interface specifications to enable interoperability among distributed discrete-event simulation models.
This paper provides a comprehensive review of discrete event simulation publications published between 2002 and 2013 with a particular focus on applications in manufacturing. The literature is classified into three general classes of manufacturing system design, manufacturing system operation, and simulation language/package development. The paper further categorizes the literature into 11 subclasses based on the application area. The current review contributes to the literature in three significant ways: (1) it provides a wide coverage by reviewing 290 papers; (2) it provides a detailed analysis of different aspects of the literature to identify research trends through innovative data mining approaches as well as insights derived from the review process; and (3) it updates and extends the existing classification schemes through identification and inclusion of recently emerged application areas and exclusion of obsolete categories. The results of the literature analysis are then used to make suggestions for future research.
Production system-based simulation for backward on-line job change scheduling
2014, Simulation Modelling Practice and Theory
Citation Excerpt :
There are three simulation modeling approaches for discrete event simulations: the event, activity and state approaches [13]. Among these approaches, event-based modeling [21] and simulation are largely used for production scheduling. In previous research on backward scheduling using event-based simulation, the release plan was initially created by a backward simulation.
Backward on-line job change scheduling, referring to the on-line job change scheduling of a current processing step to satisfy the job change schedule of the subsequent processing step, is a common problem in modern Fabs. In this research, the production system-based simulation methodology is proposed to solve the backward on-line job change scheduling problem. This simulation is processed by the state change that is caused by an execution of the operator, and it finds the schedule with the best handle values considering the current status. Several simulation runs with diverse handle values were required to find the best values because the status of the shop floor can change dynamically. To validate the simulation, this production system-based simulation is applied to the on-line job change scheduling of a tire belt processing step as part of the tire manufacturing process.
A comparative study on virtual machine model for testing semi-conductor production equipment
2019, International Journal of Innovative Technology and Exploring Engineering
Event graph modeling of a heterogeneous job shop with inline cells
2015, Proceedings - Winter Simulation Conference

View all citing articles on Scopus

View full text