Skip to content

Latest commit

 

History

History
199 lines (143 loc) · 6.4 KB

File metadata and controls

199 lines (143 loc) · 6.4 KB

Key Insights & Business Impact

Executive Summary

Analysis of 526 manufacturing records across 175 batches reveals significant opportunities for cost reduction and quality improvement. Main problems: material loss in coating stage, inconsistent operator performance, and humidity-related deviations.


Top 5 Findings

1. Coating Stage Has Highest Material Loss

Finding:
Coating stage loses an average of 3.5 kg per batch, which is 25% more than Granulation and Compression stages.

Why This Happens:

  • Uneven coating causes material waste
  • Peeling and sticking issues
  • Equipment calibration problems

Business Impact:

  • If we process 1000 batches per year, that's 3,500 kg wasted in coating alone
  • At $50/kg material cost = $175,000 lost per year
  • Fixing coating issues could save 15-20% of this loss = $26,000-35,000 savings/year

What To Do:

  • Check coating machine calibration monthly
  • Review coating parameters (spray rate, pan speed)
  • Train operators on proper coating technique

2. Humidity Issues Cause 40% More Loss

Finding:
Batches with humidity problems (below 40% or above 55%) show 40% higher material loss compared to normal conditions.

Data:

  • Normal humidity (45-50%): Average loss = 2.5 kg
  • Problem humidity: Average loss = 3.5 kg
  • Difference: 1 kg extra loss per batch

Business Impact:

  • About 30% of batches have humidity remarks
  • That's ~300 batches/year with extra 1 kg loss each
  • 300 kg extra waste = $15,000 per year
  • Installing better humidity control = ROI in 6 months

What To Do:

  • Install automated humidity control in production rooms
  • Set alerts when humidity goes outside 45-50% range
  • Schedule production of sensitive products during ideal humidity months

3. Machine CT-04 and G-03 Need Maintenance

Finding:
Machines CT-04 and G-03 have 3x more deviations than other machines.

Numbers:

  • CT-04: 45 deviations in the dataset
  • G-03: 38 deviations
  • Other machines: Average 15 deviations

Most Common Problems:

  • CT-04: Spillage, Uneven coating
  • G-03: Sticking, Peeling

Business Impact:

  • Each deviation means 30 min to 2 hours of delay
  • Delay in production = missed delivery schedules
  • Average 40 deviations × 1 hour delay = 40 hours downtime
  • At $500/hour production cost = $20,000 lost

What To Do:

  • Emergency maintenance on CT-04 and G-03
  • Replace worn parts (spray nozzles, seals)
  • Consider replacing CT-04 if repairs don't fix it (old machine)

4. Operator Performance Varies Significantly

Finding:
Some operators have 2-3x more deviations than others, even when using the same machines.

Best Performers:

  • Grace: 12 deviations out of 53 operations (23% deviation rate)
  • Eve: 14 deviations out of 48 operations (29% deviation rate)

Needs Improvement:

  • Alice: 38 deviations out of 55 operations (69% deviation rate)
  • Bob: 32 deviations out of 52 operations (62% deviation rate)

Why This Matters:
It's not just machine problems - operator skill makes a big difference.

Business Impact:

  • Training underperforming operators could reduce deviations by 50%
  • 20 fewer deviations = 20 hours saved = $10,000/year per operator
  • 3 operators need training = $30,000 total annual savings

What To Do:

  • Pair Alice and Bob with Grace or Eve for one week
  • Create standard operating procedures (SOPs) with checklists
  • Review why Grace and Eve do better (speed, technique, attention to detail)

5. Metformin and Amoxicillin Have Inconsistent Yield

Finding:
Metformin 500mg and Amoxicillin 250mg show high variance in loss - sometimes 1 kg loss, sometimes 6 kg loss in the same stage.

Variance Analysis:

  • Metformin: Loss variance = 4.2 (very unstable)
  • Amoxicillin: Loss variance = 3.8 (unstable)
  • Vitamin C: Loss variance = 1.1 (stable and predictable)

Why This Is Bad:
We can't predict how much material we'll need. Sometimes we order too much, sometimes too little.

Business Impact:

  • Unpredictable yield = poor inventory planning
  • Safety stock costs extra = $5,000-10,000 tied up in inventory
  • Rush orders when we run short = 20% premium on materials

What To Do:

  • Investigate why Metformin and Amoxicillin are inconsistent
  • Check if it's raw material quality (supplier issue)
  • Standardize process parameters for these two products
  • Consider splitting into separate SOPs

Overall Business Impact Summary

Problem Area Annual Loss Potential Savings
Coating stage waste $175,000 $35,000
Humidity issues $15,000 $15,000
Machine downtime (CT-04, G-03) $20,000 $18,000
Operator training gap $45,000 $30,000
Inventory costs (inconsistent yield) $10,000 $8,000
TOTAL $265,000 $106,000/year

ROI: Implementing all fixes costs ~$50,000 (training, maintenance, humidity control)
Payback Period: 6 months
3-Year Savings: $318,000


Action Plan Priority

Do This Month (High ROI, Low Effort):

  1. Emergency maintenance on CT-04 and G-03
  2. Operator training program (pair weak with strong)
  3. Set humidity alerts in production rooms

Do This Quarter (Medium ROI, Medium Effort): 4. Install automated humidity control 5. Review and update coating process SOPs 6. Investigate Metformin/Amoxicillin inconsistency

Do This Year (Planning & Long-term): 7. Consider replacing CT-04 if maintenance doesn't help 8. Build predictive model for yield forecasting 9. Implement real-time dashboard for production monitoring


How We Found This

All insights come from SQL queries in Pharma.sql:

  • Material loss analysis (Query #2, #3, #9)
  • Machine performance (Query #6, #12)
  • Operator performance (Query #7, #16)
  • Environmental impact (Query #14)
  • Product variance (Query #10)

Data Coverage: September-October 2025, 526 records, 175 batches


Recommendations for Leadership

  1. Invest in preventive maintenance - $20k now saves $38k/year
  2. Operator training is high ROI - $5k training saves $30k/year
  3. Humidity control pays for itself in 6 months
  4. Track these metrics monthly - Don't wait for quarter-end

This analysis shows that small improvements in manufacturing efficiency can have big financial impact. The data is already here - we just need to act on it.


Prepared By: Parth B Mistry
Analysis Date: December 2025
Data Source: Pharma_Process table (526 records)