Data Types

Autumn Labs can ingest various forms of data from multiple sources. When data is correctly linked, Autumn Labs delivers meaningful insights that make decision-making a breeze. This guide explains the different data types and how they’re used.

At the highest level, data relates either to the Station or to the Unit being processed.

Station Data Types

Station States

Station States represent a Station’s operational status. They follow a single-timeline state machine. You can define custom states, but these reserved states have special meaning:

• active
  Station is currently processing something (e.g., “Moving a conveyor belt”).

• idle
  Station is ready but not processing anything (e.g., “Waiting for Part”).

• error
  Station has encountered an error (e.g., “Sensor Failure”).

• maintenance
  Station is under maintenance or calibration (e.g., “Routine Calibration”).

• custom_state
  Station is in a custom state you define.

Below is a timeline example illustrating how State and Description durations accumulate:

1. At 01:23:00

   • State: State 1
   • Description: Description 1

2. At 01:23:03

   • State: State 1 (continued)
   • Description: Description 2

3. At 01:23:05

   • State: State 1 (continued)
   • Description: Description 1

4. At 01:23:07

   • State: State 2
   • Description: Description 1

5. At 01:23:10

   • State: State 3
   • Description: Description 1

This results in:

• State 1: Total duration = 7 seconds (01:23:00 → 01:23:07)
  • Description 1 lasted 6 seconds
    • Occurrence 1: lasted 3 seconds (01:23:00 → 01:23:03)
    • Occurrence 2: lasted 3 seconds (01:23:05 → 01:23:07)
  • Description 2 lasted 2 seconds (01:23:03 → 01:23:05)
• State 2: Total duration = 3 seconds (01:23:07 → 01:23:10)
  • Description 1 lasted 3 seconds (01:23:07 → 01:23:10)
• State 3: Starts at 01:23:10 and continues until the next change.

Stations set states by calling set_station_state(...)

Tip: Always include a meaningful description when sending a state. It helps Autumn Labs understand the right context and offer more accurate insights. States of same name and description so the platform can group states and provide aggregate level information (for example, state="error", description="Conveyor Jam").

---

Station Modes

Station Modes let users specify the type of routine a station is running, so that the dashboard can filter metrics and calculations accordingly. For example, at the start of each day, you may need to calibrate the station and perform five dry-run cycles to verify optimal operation. During this period, setting the station mode to something like dry-run will prevent those five runs aren’t counted toward productive metrics (e.g., unit output volume).

Customers often set the station to online to refer to the station now being connected on the manufacturing line or offline to refer to station is now taken off the manufacturing line.

Stations set modes by calling set_mode(...).

---

Station Metrics

Station Metrics are key-value data points collected at the Station. They provide information about performance or environment. Examples include:

• temperature: Float (°C)
• pressure: Float (psi)
• motor_speed: Float (rpm)

Stations push metrics by calling push_metric(...).

---

Station Logs

Station Logs capture messages or debug information you want to record. They include:

• trace: Detailed debugging
• debug: Debugging information
• info: General information
• warn: Warning messages
• error: Error conditions
• fatal: Critical errors
• panic: System panic

Stations push logs by calling push_log(...)

---

Unit Data Types

Unit Metrics

Unit Metrics provide information about the individual units that the station is handling. They must include unit_serial_number and are crucial for calculations like CPK. Examples include:

• length: Float (millimeters)
• weight: Float (grams)
• test_voltage: Float (volts)
• color: String

Use the push_metric(...) command and pass in a unit_serial_number to record unit-specific metrics.

---

Unit Events

Unit Events track Unit movement through Stations. The two main types are:

• track_in: Unit enters the station
• track_out: Unit exits the station (Optionals: include a result, such as “passed” or “failed”)

Tip: Send all unit associated data after you have tracked the unit "in" the station and prior to tracking the unit out of the station. This helps Autumn Labs keep a clear audit history of all data collected by station for a specific unit.

Use the track_unit_in(...) command and track_unit_out(...) to track the units in the station.

---

Unit Logs

Unit Logs record messages or debug information tied to a specific Unit. They share the same levels as Station Logs:

• trace: Detailed debugging
• debug: Debugging information
• info: General information
• warn: Warning messages
• error: Error conditions
• fatal: Critical errors
• panic: System panic

Use the push_log(...) command and pass in a unit_serial_number to record unit-specific logs.

---

Asset

Assets record file uploads or binary data associated with a Unit. Common asset types include:

• Images
• Logs
• Test results
• Configuration files

Use the push_asset(...) command and pass in a unit_serial_number to upload unit-specific assets.

---

Putting It All Together

• Station Data: States, Metrics, Logs
• Unit Data: Metrics, Events, Logs, Assets

When you combine these data types, Autumn Labs help correlate and draw relationships between Station performance, Unit outcomes, and any file assets to deliver powerful insights.