How do I calculate summaries (average, total, min, max) in PI?

PI System provides multiple ways to compute summaries over time ranges.

PI Web API Summaries

import requests

# Get summary statistics for a tag
response = requests.get(
    f"{base_url}/streams/{web_id}/summary",
    params={
        "startTime": "*-24h",
        "endTime": "*",
        "summaryType": "Average,Minimum,Maximum,Total,StdDev,Count"
    },
    auth=auth, verify=False
)

for item in response.json()["Items"]:
    print(f"{item['Type']}: {item['Value']['Value']}")

AF Analytics Expressions

' Time-weighted average over 1 hour
TagAvg('TagName', 't-1h', '*')

' Total over 8-hour shift
TagTot('TagName', 't-8h', '*')

' Min/Max
TagMin('TagName', 't-1d', '*')
TagMax('TagName', 't-1d', '*')

PI DataLink (Excel)

=PIAdvCalcVal("Sinusoid", "t-1d", "*", "average", "time-weighted", 0, 1, "PIServer")

Summary Types

Type	Description
Average	Time-weighted average (accounts for uneven timestamps)
Total	Time-weighted total (useful for flow totalization)
Minimum	Smallest value in the range
Maximum	Largest value in the range
StdDev	Standard deviation
Count	Number of events in the range
Range	Max - Min
PercentGood	Percentage of time with good-quality data

Important: Time-Weighted vs Event-Weighted

• Time-weighted (default): accounts for how long each value was held. Use for process values.
• Event-weighted: treats each event equally regardless of duration. Use for discrete events.

PI defaults to time-weighted — this is correct for most process data scenarios.