MAIN MENUPREVIOUS MENU---------------------------------Search CD-ROMSearch ResultsPrintProceedings of the 2000 Winter Simulation Conference J. A. Joines, R. R. Barton, K. Kang, and P. A. Fishwick, eds. UNDERSTANDING THE IMPACT OF EQUIPMENT AND PROCESS CHANGES WITH A HETEROGENEOUS SEMICONDUCTOR MANUFACTURING SIMULATION ENVIRONMENT Jeffrey W. Herrmann Brian F. Conaghan Laurent Henn-Lecordier Praveen Mellacheruvu Manh-Quan Nguyen Gary W. Rubloff Rock Z. Shi Institute for Systems Research University of Maryland College Park, MD 20742, U.S.A. ABSTRACT Simulation models are useful to predict and understand the impact of changes to a manufacturing system. Typical factory simulation models include the parts being manufactured in the factory and the people and resources processing and handling the parts. However, these models do not include equipment or process details, which can affect operational performance such as cycle time and inventory. Separate models are used to evaluate processes and equipment. Thus, it is difficult to evaluate the operational impact of equipment or process changes. However, this information could help factory managers and manufacturing process engineers make better decisions when changing processes or selecting equipment configurations. This paper describes a heterogeneous simulation environment for understanding how equipment and process changes affect the performance of a wafer fabrication facility. This integrated tool incorporates response surface models that describe process behavior, operational and optimization models of equipment behavior, and a discrete-event simulation model of factory operations. Thus, the tool can measure how process changes and equipment configuration changes change the system performance. We have applied this tool to a specific wafer fab problem. 1 INTRODUCTION Understanding the operational impact of equipment and process changes typically requires the expertise of multiple engineers and analysts. Each person uses different models to evaluate some segment of the entire manufacturing system. Thus, significant time and effort is needed to gather information to help factory managers and manufacturing process engineers make better decisions when changing processes or selecting equipment configurations. For instance, experimental approaches to optimizing and controlling manufacturing processes by changing the process parameters has been very successful (see, for example, Stefani et al. 1996). However, process engineers often focus on the process itself and may find it difficult to consider how process parameter changes affect the overall manufacturing system performance. One significant impact of changing process parameters is a change to the time that a process requires, which affects the total lot processing time (the time needed to process all of the wafers). If the process (or a sequence of processes) is performed by a cluster tool, which can process multiple wafers simultaneously, then the impact of the process change on the total lot processing time may be very complex (see, for instance, Herrmann et al. 1999). However, process engineers usually develop response surface models (RSMs) for process rate (like etch rate or deposition rate). Although a higher rate should reduce the total lot processing time, the quantitative relationship is often complex, involving the consequences not only of the process, but also the overhead associated with startup and ending of the process cycle in the tool. Thus, a small change to the process time may change the total lot processing time drastically, or it may not. Process “improvements” that significantly increase the total lot process time and reduce a tool’s capacity (especially if that tool is a bottleneck tool) can seriously degrade 1491Herrmann, Conaghan, Henn-Lecordier, Mellacheruvu, Nguyen, Rubloff, and Shi manufacturing system performance by increasing cycle time and decreasing maximum throughput. Consider the following, somewhat exaggerated, scenario. A process engineer wants to modify a particular semiconductor manufacturing process in order to improve process performance. Changing the process parameter values (or recipe) will change the process performance in various ways. If the change affects the time needed to perform the process, the process engineer must determine if the change will affect the equipment’s ability to satisfy its throughput requirements (wafers processed per day). Thus, the process engineer calculates the modified process time and gives that value to an industrial engineer, who then determines whether the process change is acceptable. The industrial engineer uses a capacity planning model to determine, if the proposed process change did occur, whether the equipment’s utilization would remain at an acceptable level, since a very high utilization can cause excessive delays. Such a utilization constraint, although practical, is a myopic way to avoid potential problems, since it does not consider benefits that could occur in other parts of the factory. Also, the process engineer faces a delay while the industrial engineer does the capacity analysis. Furthermore, the declaration that a process change is unacceptable does not give the process engineer feedback needed to find a change that satisfies everyone’s requirements. Consider also the problem of selecting equipment configurations for a semiconductor wafer fab. A cluster tool has integrated processing modules linked mechanically. Typical cluster tools include load locks, process modules, and a wafer handler. A cluster tool can process multiple wafers simultaneously. Sequential cluster tools integrate a sequence of processes, while other tools have two or more identical modules that are used in parallel. Hybrid configuration are also possible. Unlike single-process tools, the complex behavior of a cluster tool makes analyzing the throughput of different configurations a difficult task. Adding a second chamber to a tool does not automatically halve the total lot processing time (or double the tool capacity). Understanding the impact of different tool configurations requires an integrated model that can describe both the tool behavior and the factory behavior. This paper describes a heterogeneous simulation environment (HSE) that integrates a variety of simulation and analytical models of the manufacturing system and its components. Specifically, the HSE