Initial Situation
The company had a traditional Quality Management System used mainly for documentation and compliance, but it was not integrated with shop-floor production activities.
Operators reported defects through:
paper forms
Excel spreadsheets
end-of-shift quality reports
Quality engineers reviewed these reports hours or sometimes days after production events occurred.
This delay meant that many defects were discovered after entire batches had already been produced, increasing rework and scrap costs.
The company leadership knew quality problems existed but lacked real-time visibility into when and where defects were being generated.
Baseline Production Metrics (Before Session Pilot)
Production Capacity
Average daily production:
1,200 units/day
Working schedule:
2 shifts per day
5.5 days per week
Monthly production output:
1,200 × 26 ≈ 31,200 units/month
Baseline Quality Performance
Internal quality audits conducted over a two-month period established baseline KPIs.
KPI | Baseline Value |
Defect rate | 4.2% |
Rework rate | 3.1% |
Scrap rate | 1.1% |
First Pass Yield (FPY) | 95.8% |
Defect Volume
Monthly defects generated:
31,200 × 4.2% = 1,310 defective units/month
Of these:
Reworkable defects:
31,200 × 3.1% ≈ 967 units/month
Scrap units:
31,200 × 1.1% ≈ 343 units/month
Rework Cost
Average rework cost per unit:
€18
Total monthly rework cost:
967 × €18 ≈ €17,406/month
Scrap Cost
Average production cost per unit:
€42
Scrap cost:
343 × €42 ≈ €14,406/month
Total Quality Loss
Monthly financial loss due to defects:
€17,406 + €14,406 ≈ €31,812/month
Annualized quality loss:
≈ €381,000/year
Baseline Quality Reporting Process
The quality reporting workflow had several inefficiencies.
Operator Reporting Delay
Operators often waited until the end of their shift to log issues.
Average defect reporting delay:
3–5 hours
Root Cause Investigation Time
Quality engineers needed to manually trace production data from multiple sources.
Average root cause analysis duration:
2.5 days
Defect Recurrence
Because defects were detected late, the same issue often propagated across multiple batches.
Average defect recurrence rate:
28%
This meant more than a quarter of defects repeated before corrective action was implemented.
Key Operational Processes Impacted
The absence of real-time quality visibility impacted several operational workflows.
Shop-Floor Defect Reporting
Operators had no easy digital interface to log issues immediately.
Quality Event Tracking
Defects were disconnected from production timestamps and machine states.
Batch-Level Containment
Quality engineers could not isolate the exact moment defects began.
Continuous Improvement
Production teams lacked sufficient data to identify recurring defect patterns.
The Solution
The company deployed Session Pilot’s AI-driven Quality Intelligence System (QIS) to digitize shop-floor quality reporting and connect quality events with production activity.
Session Pilot introduced several capabilities.
Real-Time Defect Reporting
Operators were given a simple digital reporting interface accessible through tablets at each workstation.
Operators could record:
defect type
workstation
timestamp
part ID or batch ID
optional images of defects
Reporting time per event dropped to less than 10 seconds.
AI Defect Categorization
Session Pilot used historical defect patterns and production context to automatically classify issues into standardized categories such as:
dimensional deviation
surface defects
welding defects
assembly misalignment
This ensured consistent defect classification across shifts.
Production Context Integration
Each defect event was automatically linked to:
workstation ID
operator shift
production batch
timestamp
machine state
This allowed engineers to identify exact conditions under which defects occurred.
Quality Monitoring Dashboard
Quality managers gained access to real-time dashboards showing:
defect trends by workstation
defect frequency by shift
recurring defect patterns
production batches affected
Results After 4 Months
Following deployment, several key performance indicators improved.
Faster Defect Detection
Average reporting delay reduced dramatically.
Before:
3–5 hours
After:
Immediate reporting (< 2 minutes)
Detection time reduction:
≈ 60% faster defect identification
Reduced Rework
Earlier defect detection allowed issues to be corrected before entire batches were affected.
Rework rate decreased from:
3.1% → 2.5%
Monthly rework volume:
Before:
967 units
After:
31,200 × 2.5% = 780 units
Reduction:
187 units/month
Monthly cost savings:
187 × €18 = €3,366
Scrap Reduction
Earlier intervention prevented defective components from progressing through production.
Scrap rate reduced from:
1.1% → 0.9%
Monthly scrap units:
Before:
343 units
After:
31,200 × 0.9% = 281 units
Reduction:
62 units/month
Monthly savings:
62 × €42 = €2,604
Root Cause Investigation Speed
With automated production data linking, quality engineers could analyze defects much faster.
Before:
2.5 days
After:
1.5 days
Reduction:
40% faster investigations
Operator Engagement
The simplified reporting system increased defect reporting activity.
Average weekly defect reports:
Before:
~55 reports/week
After:
~140 reports/week
Increase:
2.5× more quality data captured
This improved the accuracy of quality analytics.
Reduced Defect Recurrence
Defect recurrence dropped significantly.
Before:
28% recurrence
After:
17% recurrence
Reduction:
39% improvement
Financial Impact
Total monthly quality loss reduction:
Rework savings:
€3,366/month
Scrap savings:
€2,604/month
Total monthly improvement:
€5,970/month
Annual savings:
≈ €71,640/year
Operational Impact
Session Pilot transformed the company’s quality management process from delayed defect reporting to real-time shop-floor intelligence.
The platform now supports:
real-time defect tracking
batch-level containment
faster root cause investigations
data-driven quality improvement initiatives
Production teams gained continuous visibility into quality performance, enabling faster intervention and reduced defect propagation.