Units Tested Today ℹ Units Tested Total VFD units that have completed the full test cycle (thermal + vibration + load + HiPot) today. Includes both passed and failed units. Data Source: Test bench PLC — cycle complete signal + MES test records
47
↑ 12% vs yesterday
First Pass Yield ℹ First Pass Yield (FPY) Percentage of units that pass ALL tests on the first attempt without any rework. Higher FPY = better assembly quality. Target: >95%. Data Source: MES test records — first test result per serial number
94.2%
↑ 2.1% this week
In Retest ℹ Units in Retest Units that failed initial testing and are queued for rework and retest. High numbers indicate assembly issues that need investigation. Data Source: MES rework queue + test status flags
3
Thermal failures
Avg Test Duration ℹ Average Test Duration Average time to complete full test cycle per unit. Includes thermal soak, vibration test, load test, and HiPot. Lower is better for throughput. Data Source: Test bench PLC — cycle start/end timestamps
18:42
↑ 3 min faster
HiPot Failures ℹ High Potential Test Failures HiPot tests insulation integrity at high voltage. Failures indicate serious safety issues — units cannot ship. Zero is the target. Data Source: HiPot tester — pass/fail result + leakage current readings
0
Clean run
Recent Test Results — FRENIC Series ℹ Test Results Table Shows individual VFD test results with pass/fail status for each test type. AI Prediction column shows FluxAI's confidence that the unit will perform reliably in the field, based on test data patterns. Data Source: MES test database — serial, model, and all test parameters per unit Last 20 units
Serial Model Power Thermal ℹ Thermal Test Tests temperature rise under rated load. Verifies IGBT mounting and thermal interface. Failure indicates poor thermal pad contact or IGBT issues. Data Source: Thermal sensors on heatsink + IR camera readings during load test Vibration ℹ Vibration Test Measures vibration signature during operation. Detects loose components, bearing issues, or mounting problems. Marginal readings indicate future reliability risk. Data Source: Accelerometer on VFD chassis — FFT spectrum data at multiple frequencies Load ℹ Load Test Tests VFD performance under various load conditions (25%, 50%, 75%, 100%). Verifies power output, efficiency, and control accuracy. Data Source: Load bank controller + power analyzer — voltage, current, power factor, efficiency Status AI Prediction ℹ AI Prediction FluxAI's confidence score (0-100%) that this unit will perform reliably. Based on test data patterns compared to historical failures. Lower scores suggest investigation needed even if tests "passed". Data Source: ML model trained on all test parameters + historical field failure data
FRN-4521 FRENIC-Ace 22 kW Review Pass Pass Hold 82% — Check IGBT pad
FRN-4520 FRENIC-MEGA 45 kW Pass Pass Pass Shipped 98% — No issues
FRN-4519 FRENIC-Ace 15 kW Pass Pass Pass Shipped 99% — No issues
FRN-4518 FRENIC-Mini 7.5 kW Pass Marginal Pass Shipped 91% — Monitor trend
FRN-4517 FRENIC-Ace 30 kW Fail Pass Fail Rework 45% — Replace IGBT
AI Alerts ℹ AI-Generated Alerts Proactive alerts from FluxAI based on pattern recognition. These are predictions and recommendations, not just reactive alarms. Helps prevent issues before they cause downtime or quality escapes. Data Source: FluxAI anomaly detection engine — analyzes all connected data streams in real-time 3 active
⚠
Thermal anomaly — FRN-4521
Temperature rise 8% above baseline • 12 min ago
ℹ
Vibration trend — Station 2
Marginal readings increasing • 45 min ago
!
IGBT failure pattern detected
3 units from same batch • 2 hours ago
Thermal Test Trend (Last 50 Units) ℹ SPC Control Chart Statistical Process Control chart showing thermal test results over time. UCL/LCL are control limits. Points outside limits or trending toward limits indicate process drift that needs attention. Data Source: Thermal test readings from last 50 units — continuous time-series data
Anomaly UCL Target LCL
Failure Categories ℹ Failure Pareto Breakdown of test failures by category. Helps prioritize improvement efforts. Thermal failures are most common — often indicates IGBT mounting or thermal interface issues. Data Source: MES failure codes — aggregated from all test stations
Thermal Load Vibration HiPot
Station Utilization ℹ Test Station Utilization Percentage of time each test station is actively testing (vs idle or in changeover). Low utilization indicates bottlenecks or scheduling issues. Target: >85%. Data Source: Test bench PLCs — run/idle/fault state signals
Station 1 87%
Station 2 92%
Station 3 68%
Station 4 79%
Boards Produced ℹ Boards Produced Total PCBs that have completed the full SMT assembly process today — paste printing, component placement, reflow, and AOI inspection. Data Source: AOI machine — board count at end of line + barcode scan data
1,247
On target
First Pass Yield ℹ First Pass Yield Percentage of boards passing AOI inspection without any defects requiring rework. Target for SMT is typically >98%. Data Source: AOI inspection results — pass/fail per board barcode
98.7%
↑ 0.8%
Cycle Time ℹ Cycle Time Average time for one board to pass through the entire SMT line. Determined by slowest station (usually pick-and-place or reflow). Data Source: Conveyor sensors — board entry/exit timestamps at each station
12.4s
Optimized
Defects Predicted ℹ Pre-Reflow Defect Prediction FluxAI predicts defects BEFORE boards enter the reflow oven — using solder paste volume, placement accuracy, and environmental data. Allows intervention before defects become permanent. Data Source: SPI paste volume data + pick-and-place offset data + humidity/temp sensors
16
Before reflow
AOI Rejects ℹ AOI Rejects Boards flagged by Automated Optical Inspection after reflow. These require manual review — some are false positives, others need rework. Data Source: AOI camera system — defect images + coordinates + classification
3
Under investigation
Equipment Status ℹ SMT Line Equipment Real-time status of all machines in the SMT line. Green = running, Yellow = idle/changeover, Red = error/down. CPH = Components Per Hour for pick-and-place machines. Data Source: Machine PLCs via SECS/GEM or OPC-UA — state, counters, alarms
Paste Printer
98% efficiency
Pick & Place 1
34,521 CPH
Pick & Place 2
32,847 CPH
Reflow Oven
All zones nominal
AOI Station
2.1s/board
Rework Station
3 boards queued
SPI
100% coverage
Conveyor
Normal speed
Pre-Reflow Defect Prediction ℹ AI Defect Prediction FluxAI analyzes solder paste volume, component placement accuracy, and environmental conditions to predict which boards will likely have defects BEFORE they go through reflow. Allows intervention while defects are still fixable. Data Source: SPI (paste height/volume per pad) + pick-and-place (X/Y/θ offset per component) + environmental sensors AI
Insufficient solder 7
Bridging risk 5
Tombstone risk 3
Misalignment 1

Predictions based on paste volume, placement data, and environmental conditions.

Reflow Oven — Live Temperature Profile ℹ Reflow Temperature Profile Shows actual vs target temperature profile through reflow oven zones. Preheat → Soak → Reflow → Cooling. Deviations from target can cause solder defects. FluxAI can auto-adjust zones based on board characteristics. Data Source: Reflow oven — zone thermocouples + conveyor speed + profiler data All zones within spec
Preheat Soak Reflow Cooling Target Actual
Cpk — VFD Assembly ℹ Process Capability Index (Cpk) Measures how well the process stays within specification limits. Cpk >1.33 is acceptable, >1.67 is good, >2.0 is excellent. Lower values indicate process variation that needs improvement. Data Source: Test measurements — statistical analysis of all VFD output parameters
1.45
Above 1.33 target
Cpk — SMT Line ℹ SMT Process Capability Cpk for SMT placement accuracy and solder quality. Higher Cpk means more consistent assembly with fewer defects. Data Source: SPI paste measurements + pick-and-place placement accuracy data
1.52
Excellent
Cpk — Thermal Test ℹ Thermal Test Capability Below 1.33 target indicates too much variation in thermal test results. May indicate inconsistent IGBT mounting or thermal interface application. Data Source: Thermal test bench — temperature rise measurements per unit
1.18
Below target
In-Spec Rate ℹ In-Specification Rate Percentage of measurements falling within specification limits. Different from FPY — this tracks individual measurements, not whole units. Data Source: All test parameters — aggregated across test stations and product lines
99.2%
Stable
Escapes This Week ℹ Quality Escapes Defective units that passed all tests and shipped to customers. Zero is always the target. Escapes indicate gaps in test coverage or inspection. Data Source: Customer RMA/complaint data linked back to production serial numbers
0
Target met
Control Chart — VFD Output Voltage Accuracy ℹ X-bar Control Chart Statistical Process Control chart showing measured values over time. UCL/LCL = Upper/Lower Control Limits (±3 sigma). CL = Center Line (target). Points should stay within limits with random variation. Data Source: Load test bench — output voltage accuracy measurements per VFD USL: +2% | LSL: -2%
UCL CL LCL
Quality by Product Line ℹ Product Line Quality First Pass Yield breakdown by FRENIC product line. Helps identify which products have quality issues. Lower-power units (Mini) sometimes have different failure modes than higher-power units. Data Source: MES test records — FPY calculated per product model code
FRENIC-Ace 98.5%
FRENIC-MEGA 97.2%
FRENIC-Mini 94.8%
Control PCBs 99.1%
AI Quality Insights ℹ AI-Powered Insights FluxAI analyzes all production and test data to identify root causes, predict trends, and recommend optimizations. These are actionable insights generated by machine learning, not just reports. Data Source: All connected systems — ML correlation across machines, tests, environment, operators

Root Cause Correlation

78% of thermal test failures correlate with IGBT mounting torque variance. Recommend calibration check on Station 3.

Trend Prediction

Solder paste viscosity trending toward lower limit. Recommend replacement within 4 hours to maintain FPY.

Optimization

Reflow zone 4 dwell time can be reduced by 3 seconds without quality impact. Potential 2% throughput gain.

Overall OEE ℹ Overall Equipment Effectiveness OEE = Availability × Performance × Quality. Industry standard metric for manufacturing productivity. World-class manufacturing achieves 85%+. This is the single most important production metric. Data Source: Calculated from PLC run/stop signals, production counters, and quality pass/fail data
84.2%
↑ 3.1% this month
Availability ℹ Availability Percentage of scheduled time that equipment is available to run. Losses include breakdowns, changeovers, and unplanned stops. Target: >90%. Data Source: Machine PLCs — run/stop/fault state signals + downtime reason codes
92.1%
Performance ℹ Performance Actual production rate vs theoretical maximum rate. Losses include slow cycles, minor stops, and speed losses. Target: >95%. Data Source: Production counters ÷ (run time × ideal cycle rate)
94.5%
Quality ℹ Quality Rate Percentage of good units produced vs total units. Losses include defects and rework. Target: >99%. Data Source: MES pass/fail records — good units ÷ total units produced
96.8%
Production Line Status ℹ Line Status Overview Real-time status of all production lines showing target vs actual output, OEE, and next action items. Helps identify bottlenecks and prioritize improvements. Data Source: MES production schedule + real-time machine data + maintenance calendar
Line Product Target Actual OEE ℹ Line OEE OEE for this specific production line. Green >80%, Yellow 70-80%, Red <70%. Data Source: Line-specific availability, performance, and quality data Status Next Action
SMT Line 1 VFD Control PCB 1,200 1,247 87.2% Running Scheduled maintenance 18:00
SMT Line 2 Power Board 800 756 79.8% Running Paste printer calibration needed
VFD Assembly FRENIC-Ace 50 47 84.5% Running On track
VFD Test Station All FRENIC 55 52 82.1% Running 3 units in retest
OEE Trend — Last 30 Days ℹ OEE Improvement Trend Shows OEE improvement over time. The marked point shows when FluxAI was deployed — improvement should be visible after deployment as predictive maintenance and quality optimization take effect. Data Source: Historical OEE calculations — daily rollup from all production data
85% FluxAI deployed
Downtime Pareto ℹ Downtime Analysis Breakdown of downtime by cause. Pareto principle: 80% of downtime usually comes from 20% of causes. Focus improvement efforts on the largest bars first for maximum impact. Data Source: Machine fault codes + operator-entered downtime reason codes from MES
Setup (42%) Material (28%) Maint (15%) Quality (10%) Other (5%)