Egg retrievals distribution optimizer

Stim duration is the number of days from cycle start to retrieval. Please upload historical values so the model can learn how stim length relates to when retrievals occur.

The tool produces a weekly stim-start plan: how many stim starts to schedule each day so that simulated retrievals line up with your desired weekday retrieval mix (Monday–Sunday).

Please enter your target retrievals per weekday. If your durations have heavy tails (high variability or rare very long cycles), set an outlier cutoff in days (default 25). For refinement the model Monte Carlo simulation; you can choose how many simulated weeks to run (default 5,000).

Hopefully, these calculations will help with work distribution in Operation Theaters and IVF laboratories.

To my fellow lab mates: I hope this tool helps with your work distribution and eases weekend workload. :)

Questions or suggestions? Contact me on LinkedIn.

Next.js

1) Durations input

Please upload a CSV or Excel file (.csv, .xlsx, .xls) with historical stim durations in the first column of the first sheet.

Upload CSV / Excel

Number of data points uploaded: 0

Upload a file to see your distribution

Drag and drop a CSV or Excel file here, or click to browse. The histogram and stim-days table appear after your file loads with valid duration values (first column).

Note: If you don't have your clinic's historical stim data, you can use the bundled sample data set instead.

2) Planning inputs

Please enter the desired retrieval distribution during the week (Mon – Sun)

Mon

Tue

Wed

Thu

Fri

Sat

Sun

Total number of retrievals per week: 0

Accuracy priority

Refinement minimizes weighted MAE on simulated completions vs your targets. Your choice below is applied directly (unless you pick Auto).

Even across all daysPrioritize Mon–Fri (minimize weekday abs error)Prioritize Sat–Sun (minimize weekend abs error)Auto (cross-validation)

Other inputs to the algorithm (optional)

Smooth stim starts curveTighter predicted completion bounds plus a per-day cap on stim starts: ±8% of the weekly average per day when accuracy is even across days. If you prioritize Mon–Fri or Sat–Sun (or Auto picks that preset), the cap widens to up to ±40%, and the model also widens predicted bounds for that prioritized group so refinement can better reach your targets. May increase retrieval error vs targets.

Account for pre-retrieval cancellationsA random share of stim cycles ends before retrieval (e.g. ~1 in 10 at 10%). The model scales expected completions and increases recommended stim starts so your weekday TVA targets stay achievable on average.

Cancellation rate (%)Applied in the linear model and Monte Carlo (Bernoulli per stim start).

Remove all data points where stim duration is larger than:

Monte Carlo simulation (weeks):

Random seed (42 — is the answer to all questions in the universe)

Strata (empirical)

Egg retrievals distribution optimizer

1) Durations input

2) Planning inputs

Please enter the desired retrieval distribution during the week (Mon – Sun)

Other inputs to the algorithm (optional)

3) Results

Average Weekly Completion Distribution

TVAs target vs Monte Carlo simulation — retrieval distribution