Skip to content

BottleneckDetectionEvents¤

Identify which station constrains a production line by analyzing utilization.

Module: ts_shape.events.production.bottleneck_detection Guide: OEE Analytics

When to Use¤

Use for multi-station line optimization. The bottleneck determines line output — finding it is the first step in improving throughput. This module compares utilization across stations and detects when the bottleneck shifts between stations over time.

Quick Example¤

from ts_shape.events.production.bottleneck_detection import BottleneckDetectionEvents
import pandas as pd
import numpy as np

np.random.seed(42)
df = pd.DataFrame({
    "timestamp": pd.date_range("2024-01-01 06:00", periods=360, freq="1min"),
    "station_A": np.random.choice([True, False], size=360, p=[0.92, 0.08]),
    "station_B": np.random.choice([True, False], size=360, p=[0.98, 0.02]),
    "station_C": np.random.choice([True, False], size=360, p=[0.85, 0.15]),
})

bd = BottleneckDetectionEvents(df, timestamp_column="timestamp")
station_uuids = ["station_A", "station_B", "station_C"]
utilization = bd.station_utilization(station_uuids, window="1h")
bottleneck = bd.detect_bottleneck(station_uuids, window="1h")
shifts = bd.shifting_bottleneck(station_uuids, window="1h")
print(bottleneck)

Key Methods¤

Method Purpose Returns
station_utilization(station_uuids, window='1h') Compute per-station uptime percentage per window DataFrame with station, window, and utilization ratio
detect_bottleneck(station_uuids, window='1h') Identify the bottleneck station (highest utilization) per window DataFrame with window and bottleneck station identifier
shifting_bottleneck(station_uuids, window='1h') Detect when the bottleneck shifts from one station to another DataFrame of shift events with before/after station
throughput_constraint_summary(station_uuids) Aggregate summary of how often each station is the bottleneck DataFrame with station and percentage of time as bottleneck

Tips & Notes¤

The bottleneck has the highest utilization

Counter-intuitively, the station with the most uptime (not the most downtime) is usually the bottleneck — it is running flat-out trying to keep up. Stations with lower utilization are often waiting.

Related modules

See Also¤