EAP connectivity
Connect FUJI, Panasonic and ASM SMT equipment for unified status and production-data acquisition.
VSKYSOFT · BEST PRACTICESBest practices.
Best practices.
Built for real manufacturing.
From SMT equipment connectivity and rapid changeover to wafer genealogy, AI defect recognition and process prediction, these customer practices show how VSKYSOFT turns manufacturing data into an operational loop.
View case directory ↓01 · CASE DIRECTORY
Select a case title to jump directly to its details. Each case is based on the supplied project material; no unverified outcomes have been added.
CASE 01 · SMT BEST PRACTICE
Connecting multi-site, multi-vendor SMT operations for rapid changeover.
CUSTOMER PROFILE
The vivo smartphone brand of Guangdong Vivo Communication Co., Ltd. entered the mobile-phone market in 2011, with the X, Xplay and Y product families. Beyond mainland China, vivo also expanded into markets including India, Thailand and Myanmar.
PROJECT SCOPE
The project covered multiple equipment groups and business units, including FUJI M8, FUJI M2, Panasonic NPM and ASM. VSKYSOFT delivered software capabilities for equipment automation, production error prevention, traceability, material preparation and line changeover.
38FUJI M8 LINES~900 NXT MODULES25PANASONIC NPM MACHINES2ASM LINES
Error-proof traceability
Link equipment, materials and production records to support poka-yoke controls and end-to-end genealogy.
Remote Skip
Provide a centralized remote-access path for equipment software support and multi-site collaboration.
Offline preparation
Move material preparation ahead of line loading to reduce waiting and improve preparation accuracy.
Rapid changeover
Coordinate equipment, programs, materials and work orders to shorten line-change preparation.
IMPLEMENTATION SITES01vivo Chang'an Industrial Park02Chongqing Factory03India Factory
CASE 02 · PRODUCT LIFECYCLE TRACEABILITY
Build wafer-level traceability.
Starting at wafer release, bind production, equipment, tooling, process, quality, packaging and inventory data to one identity—supporting forward, backward, vertical, horizontal and exception traceability.
Equipment
Machine identity, status, run parameters, production time and lot
Tools & fixtures
Tools, fixtures, inspection assets, calibration and usage history
Quality
IQC, IPQC, OQC / FQC and a wafer-level quality record
Process
Route, version, recipe, critical parameters and change history
Packaging & inventory
Package lot, pallet / carton, inventory receipt and movement
ONE WAFER · ONE RECORDLocate anomaly sources faster, support quality improvement, complaints and recall, and create a data foundation for process optimization and the digital factory.
CASE 03 · AI DEFECT CLASSIFICATION
Turn every inspection image into actionable quality data.
Deep learning and machine vision automatically recognize, classify and mark defects in EL, PL and appearance images—including location, type and severity.
DEFECT TAXONOMY
Micro crackBroken fingerBlack spot / black coreContaminationEdge damageScratchWeak solder / soldering anomalyPrinting anomalyAbnormal EL textureOther process and electrical defects
Classification goes beyond the image and connects to its complete manufacturing context.
- Wafer ID
- Production equipment
- Recipe and process settings
- Production lot
- Operator records
- Historical quality data
01RecognizeLocate and classify automatically02CorrelateBind wafer, equipment and process03AnalyzeMeasure distribution, frequency and trends04ImproveFind equipment / process-related causes
CASE 04 · PROCESS PREDICTION & OPTIMIZATION
Move from post-event analysis to prediction and closed-loop action.
Fuse equipment history, real-time parameters, inspection, lot, environment and machine state to predict next-lot risk and generate recommendations that can be executed and verified.
Quality prediction
Fuse historical and real-time data to predict risk for the next lot.
→Intelligent analysis
Identify parameter impact, trends and the optimal process window.
→Optimization guidance
Recommend changes to temperature zones, gas flow, process time and machine settings.
→Closed-loop control
Validate results after parameter dispatch and continuously improve the model.
↻
GOLDEN LINE ANALYTICS
Use wafer efficiency and line data to identify and replicate best practices.
Link efficiency grade, route, equipment, critical settings and inspection results by Wafer ID; classify high-efficiency, low-efficiency and failed cells, then benchmark production lines.
Average conversion efficiencyYieldEquipment availabilityProduction taktProcess stabilityException rateRework and scrap rate
VSKYSOFT · CASE VALUE