Dataset Viewer
Auto-converted to Parquet Duplicate
benchmarks
list
summary
dict
[ { "fab": "semiconductor_frontend", "load": "baseline_mix", "disruption": "nominal", "strategy": "hybrid_rolling_horizon", "mean_cycle_time_h": 214.83, "p95_cycle_time_h": 260.67, "on_time_delivery_pct": 0, "avg_wip": 13.3, "throughput_lots_per_week": 10.4, "machine_utilizatio...
{ "project": "ReentFlow", "version": "1.0.0", "scenarios": 8, "algorithms": 9, "primary_metrics": [ "mean_cycle_time_h", "p95_cycle_time_h", "throughput_lots_per_week", "avg_wip", "on_time_delivery_pct", "machine_utilization_pct", "tardiness_h" ], "fab_modes": [ "semiconduc...

ReentFlow Fab Scenarios

Synthetic reentrant manufacturing fab scenarios for semiconductor, pharmaceutical, and precision optics production.

Schema Fields

Field Description
lot_id Unique lot identifier
product_family Product family (logic, memory, pharma, optics)
route Ordered operation sequence
operation Operation ID within route
machine_group Lithography, etching, cleaning, inspection, deposition
qualified_machines Tools qualified for the operation
processing_time Base processing time (hours)
setup_family Sequence-dependent setup family
release_time Lot release time
due_date Customer due date
priority Priority level (1 = expedite)
batch_compatibility Whether operation supports batching
maintenance_windows Scheduled tool maintenance
failure_probability Stochastic failure rate

Contents

  • sample_*.json — Full scenario + optimization results
  • manifest.json — Dataset manifest
  • eval_results.json — Benchmark metrics across scenarios
  • space-bundle/ — Gradio console bundle

Usage

import json
from huggingface_hub import hf_hub_download

path = hf_hub_download("alirezaaminzadeh/reentopt-fab-scenarios", "sample_semiconductor_frontend_high_wip_surge_nominal.json", repo_type="dataset")
scenario = json.load(open(path))
Downloads last month
62

Space using alirezaaminzadeh/reentopt-fab-scenarios 1

Collection including alirezaaminzadeh/reentopt-fab-scenarios