A step toward capacity planning at finite capacity in semiconductor manufacturing
Emna Mhiri, Mireille Jacomino, Fabien Mangione, Philippe Vialletelle, Guillaume Lepelletier
Abstract
Emna Mhiri, Mireille Jacomino, Fabien Mangione, Philippe Vialletelle, Guillaume Lepelletier
Abstract
Production planning in the Semiconductor Industry (SI) has emerged as the most complex process due to its process complexity, technological constraints and high-mix low-volume production characteristics. In this paper, we present two different production planning approaches, developed by STMicroelectronics and G-SCOP research laboratory, to better control the production in 300mm production line at Crolles. At first, a mixed integer program (MIP) is proposed that projects the production lot trajectories (start and end dates) for the remaining subsequent steps, taking into account finite production capacities. A heuristic is then proposed to simplify the problem of finite capacity by neglecting equipment capacity. This approach results in the development of an infinite capacity WIP projection engine that complies with lots due dates and takes into account cycle time variability.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Production planning in the Semiconductor Industry (SI) has emerged as the most complex process due to its process complexity, technological constraints and high-mix low-volume production characteristics. In this paper, we present two different production planning approaches, developed by STMicroelectronics and G-SCOP research laboratory, to better control the production in 300mm production line at Crolles. At first, a mixed integer program (MIP) is proposed that projects the production lot trajectories (start and end dates) for the remaining subsequent steps, taking into account finite production capacities. A heuristic is then proposed to simplify the problem of finite capacity by neglecting equipment capacity. This approach results in the development of an infinite capacity WIP projection engine that complies with lots due dates and takes into account cycle time variability.
Key concepts: Capacity planning, Semiconductor device fabrication, Production (economics), Production line, Heuristic, Production planning, Process (computing), Computer science