Skip to content

withme.md

Latest commit

 

History

History
104 lines (72 loc) · 2.99 KB

withme.md

File metadata and controls

104 lines (72 loc) · 2.99 KB
Raw

withme.md

Project Positioning

This is a simulated case study for portfolio use. It is designed to mirror real operational decision-making while avoiding proprietary code and confidential internal datasets.

Core comparison:

  • 4 workers x 10 hours
  • 6 workers x 6-7 hours

Target outcome of the comparison:

  • verify whether 6x6-7 can deliver better per-person efficiency,
  • improve stability (less backlog pressure),
  • reduce labor cost.

What We Used

Data foundation:

  • Incoming orders from Create Time windows
  • Product volume from Product Size (cm) L * W * H

Generated tables:

  • data/processed/daily_orders_for_staff_allocation.csv
  • data/processed/staff_allocation_daily_comparison.csv
  • data/processed/staff_allocation_efficiency_summary.csv

Generated visuals:

  • images/jan_hourly_average_orders.svg
  • images/jan_daily_orders_with_peak.svg
  • images/staff_allocation_comparison.svg

Modeling Logic (How We Simulated)

1) Daily demand mapping

Operation day is defined as previous day 09:00:00 to current day 08:59:59.

2) Volume calculation

For each record:

  • extract length, width, height
  • compute total_volume = L * W * H

3) Packaging-time random generator

Per-order handling time is randomly sampled by total volume:

  • <=250: 8-14 sec
  • 251-500: 10-15 sec
  • 501-1000: 12-16 sec
  • 1001-2000: 13-18 sec (bridge assumption)

  • 2000: 15-20 sec

4) Fatigue logic

For 4-worker model only:

  • first 6h at base pace
  • hour 6-10 gets fatigue penalty factor 1.3-1.4
  • effectively slower completion in late shift

5) Backlog carry-over rules

  • 4-worker model: above 2000/day goes to backlog
  • 6-worker model: above 3500/day goes to backlog

Backlog is carried to next day's total demand.

6) Cost logic

Hourly wage is fixed at $18/person/hour.

Daily labor cost:

  • workers * shift_hours * 18

What The Data Shows

From staff_allocation_efficiency_summary.csv:

  • 4x10 avg orders/person/hour: 32.6308

  • 6x6-7 avg orders/person/hour: 33.7127

  • Efficiency improvement: +3.32% (6x6-7 better)

  • 4x10 total labor cost: 21600.0

  • 6x6-7 total labor cost: 20836.8941

  • Cost reduction: -3.53% (6x6-7 lower)

  • 4x10 avg end backlog: 269.5 (8 backlog days)

  • 6x6-7 avg end backlog: 29.0 (1 backlog day)

Interpretation:

  • both models finish same total processed orders,
  • 6x6-7 does it with higher per-person efficiency and lower labor cost,
  • backlog pressure is significantly lower under 6x6-7.

Practical Narrative For Case Study

This simulation supports a practical staffing argument:

  • moving from long shifts to shorter shifts with more workers can outperform long-shift concentration,
  • once fatigue and carry-over backlog are modeled, the 6x6-7 structure is more stable,
  • decision quality improves when demand timing, fatigue penalty, and cost are evaluated together.

Disclaimer

  • This project uses simulated logic and non-proprietary computation.
  • Random generation is intentional to emulate operational variability.
  • Results are directional and intended for case-study demonstration.