Skip to content

LineThroughputEvents¤

Count parts per window and check takt adherence from counter or trigger signals.

Module: ts_shape.events.production.line_throughput Guide: Production Monitoring

When to Use¤

Use for real-time throughput monitoring. Counts parts from monotonic counters and checks cycle times against takt time targets. Ideal for lines where a PLC counter increments on each part produced or a trigger signal fires per cycle.

Quick Example¤

from ts_shape.events.production.line_throughput import LineThroughputEvents
import pandas as pd
import numpy as np

df = pd.DataFrame({
    "timestamp": pd.date_range("2024-01-01 06:00", periods=480, freq="15s"),
    "part_counter": np.arange(1, 481),
    "cycle_time_s": np.random.normal(loc=58, scale=4, size=480),
})

tp = LineThroughputEvents(df, timestamp_column="timestamp")
parts = tp.count_parts(counter_uuid="part_counter", window="1m")
violations = tp.takt_adherence(cycle_uuid="cycle_time_s", takt_time="60s")
print(parts.head())

Key Methods¤

Method Purpose Returns
count_parts(counter_uuid, window='1m') Count produced parts per time window from a monotonic counter signal DataFrame with window start and part count
takt_adherence(cycle_uuid, takt_time='60s') Identify cycles that exceed the target takt time DataFrame of violations with timestamp and overshoot
throughput_oee(counter_uuid, window='1h') Compute OEE-style throughput metrics per window DataFrame with actual vs ideal counts and performance ratio
throughput_trends(counter_uuid, window='1h') Analyze throughput trends over time for shift or daily review DataFrame with trend statistics per window
cycle_quality_check(cycle_uuid) Validate cycle signal integrity: gaps, duplicates, out-of-range values Dict of quality check results

Tips & Notes¤

Handle counter resets

Monotonic counters may reset at shift boundaries or after a PLC restart. count_parts automatically detects and compensates for resets by looking at negative deltas.

Related modules

See Also¤