Capacity model with sustainability scope to predict the semiconductor manufacturing's energy consumption and carbon dioxide emissions
MetadataShow full metadata
Semiconductor industries are not only technology-intensive, but also highly energy-intensive. A wafer fab consumes about 300-400 kWh/day. To supply this amount of electricity, the amount of carbon dioxide released is approximately 180–360 metric tons per day. This emission causes climate change and also rise in energy costs. Thus, it is important to take measures to reduce energy consumption.
The scientific merit of this research is to develop two methods designed to estimate the electric energy consumption and production of a semiconductor wafer fab. The broader impact of the research is that the methods can be extended to other manufacturing industries, even though they are framed in this study in the context of the semiconductor manufacturing industry. The first method, referred to as kWh-WIP, is capacity model that estimates electricity consumption by using WIP lot information. The second method referred here as kWh-Tool is a capacity model that estimates electricity consumption by using tool-level utilization. A relation is established between the two levels of detail models by comparing them with respect to annual electricity consumption and carbon emissions. Power consumption of tools and tool-process mapping are obtained by intensive research and surveying with the professionals from the semiconductor manufacturing industry. The Measurement and Improvement of Manufacturing Capacity data set is used as the basis for the capacity simulations of this thesis work. The dataset represents 200mm wafer fab processes. The average Electrical Utilization Index (EUI) and Production Efficiency Index (PEI) computed with kWh-Tool methodology across all nine 200 mm fab considered in this study was 0.145kWh/UOP and 1.987 kWh/cm2 respectively. The average annual power consumption is 44,745,707 kWh. Annual Power Consumption values calculated in kWh-WIP and kWh-Tool methodologies are closer with a variation of less than 3%. The PEI values for different fab type computed from the kWh-Tool methodology is similar to the Optimal theoretical value of 1.231 (kWh/cm2).