Manufacturing labor content in Intel's 200mm factories has come under increased scrutiny. Newer processes require more direct headcount per wafer start. This results in increased cost and complexity of hiring, training and employing factory labor. Intel's use of various forms of factory automation has increased concurrently. To some observers, there is the perception that the labor required for interactions with the automation systems is excessive. The interpretation of an earlier labor study supported this assertion, though the granularity of the data did not allow further analysis. Additionally, no model existed for exploring the labor cost imposed by the automation system.
This paper presents a study of the automation components of manufacturing labor content for Intel's 0.35 μm semiconductor process technology (known as P854). The intent of the study was to generate accurate information in sufficient detail to provide a basis for continuous improvement for the current generation of automation, and to guide the definition of future automation systems. We present details of the design, execution and analysis methods, including a discussion of the known deficiencies. We briefly compare our approach to an earlier labor study. We also present a reusable model we developed to evaluate automation labor content.
We conclude the paper with a discussion of key findings and recommendations for reducing labor content for 0.35 μm and beyond. We have integrated information from other sources in the recommendations for the future automation system. The relationship of manufacturing philosophy to automation system usage is discussed briefly in terms of the proposed future automation system.