Installed capacity is the maximum number of wafers a fab can process per month when running all equipment at full utilization with optimal scheduling. It represents the fab's theoretical production ceiling.

How Capacity Is Determined

Capacity is set by the bottleneck tool group—the process step with the least throughput relative to demand. Even if all other steps have excess capacity, the bottleneck limits total fab output. Common bottlenecks include lithography (most expensive tools, longest process times at advanced nodes) and etch/deposition for complex multi-patterning flows.

Capacity Metrics

• Nameplate capacity: Theoretical maximum based on equipment count and throughput specs • Effective capacity: Realistic maximum accounting for PM downtime, qualification, and engineering holds (~85-90% of nameplate) • Demonstrated capacity: Highest monthly output actually achieved

Expanding Capacity

Add tools at bottleneck (quickest method—buy more scanners, etchers, etc.). Increase tool throughput (shorter process times, reduced PM frequency, faster wafer handling). Improve utilization (better scheduling, faster PM recovery, reduced engineering holds). Build new fab (takes 2-3 years and $5-20+ billion—last resort for major expansions).

Industry Capacity

Global installed capacity in 2024 exceeded 30 million 300mm-equivalent wafers per month. TSMC alone represents approximately 15-17 million WSPM. The industry added significant capacity after the 2020-2022 chip shortage, with new fabs from TSMC, Samsung, Intel, and others coming online through 2025-2027.