Solution
Defect Analytics
Classify, correlate, and rank the defect signatures driving yield loss — across every lot, layer, and inspection step. Turn thousands of KLARF records into one ranked cause list.
Signature confidence
3,412 defects · 42 lots · classified in 1m 48s
Defect data lives in a dozen disconnected tools.
Every inspection and review tool speaks its own dialect; engineers export KLARF, hand-stitch sort data, and eyeball maps in spreadsheets — losing hours per excursion and missing systematic patterns. By the time a signature is confirmed, the lot has moved three steps downstream.
How it works
01
Ingest
Pull every source
02
Classify
Recognize the signature
03
Correlate
Rank the cause
Signature Explorer
Every failure has a shape.
Pick a signature to see how it maps across the wafer. Yieldform learns these shapes — so it names the cause the moment the pattern appears.
Edge ring
— die failingLikely cause
Capabilities
Automatic classification
Every defect is classified against a signature library the moment inspection data lands — no manual review queue.
Wafer-to-wafer stacking
Overlay maps across every wafer in a lot to separate systematic patterns from random noise.
Spatial signature library
A growing set of known failure shapes — edge ring, center cluster, scratch, and more — matched automatically.
Step commonality
Automatically finds the tool, chamber, or recipe step shared by every failing die.
Exportable evidence
Every ranked cause ships with the underlying die-level data, ready for a review meeting or a KDB entry.
Specs
Formats
KLARF 1.2/1.8, STDF v4, SECS/GEM
Deployment
SaaS, On-prem, Air-gapped
Standards
SEMI E10/E58, EDA/Interface A
Security
SOC 2 Type II, SSO/SAML, RBAC
Name the defect. Defend the yield.
Bring a lot of KLARF files to a 30-minute session and watch Yieldform rank the cause live.