How to Calculate Machine Availability

How To

Summary

This document explains how Machine Availability is calculated for this OEM in Maximo Renewables. It clarifies which data fields are used, how the Status tag determines availability, and how data‑quality filters can impact the final percentage.

Note:
Machine Availability is an operational metric, not a performance metric.
Because of this, low power output, performance deviations, or power‑curve differences have no effect on availability.

Objective

To define:

The exact formula used to calculate Machine Availability
The Status tag logic that classifies a turbine as Available or Unavailable
Why performance indicators (Active Power, Wind Speed, Power Curve) do not affect availability
How data‑quality filters may influence the final availability value

Environment

IBM Maximo Renewables
OEM turbines providing a Status tag
10‑minute SCADA interval timestamp data (or implemented timestamp data)
Maximo Analytics data‑quality filtering layer

Steps

1. Data Source

For this OEM, Maximo uses the Status tag, which may include values such as:

Status Value	Meaning	Availability Classification
0	Error	Unavailable
1	OK	Available
2	Service	Unavailable
3	Pause	Unavailable
4	Communication Failure	Unavailable

Only Status = 1 is counted as "available."
All other Status values are considered "unavailable."

2. Availability Formula

Maximo uses the standard industry formula:

\text{Machine Availability} = \left(\frac{\text{Number of timestamps where Status = 1}}{\text{Total timestamps}}\right) \times 100

(Number of timestamps where Status = 1 / Total number of timestamps) \times 100

This is fully aligned with:

\text{Availability} = \frac{\text{Total Time – Downtime}}{\text{Total Time}} \times 100

Where:

Downtime = timestamps where Status ≠ 1

3. Important: Performance Data Is Not Used

Machine Availability does NOT use:

Active Power
Wind Speed
Power Curve values
Efficiency or PR
Any AP threshold (15 kW, 19 kW, 40 kW, etc.)

These belong to turbine performance metrics, not operational availability.

4. Impact of Data‑Quality Filters

The Analytics layer applies filters that remove rows containing:

Missing critical tags
Repeated/duplicated rows
Invalid or corrupt values

Filtered rows:

Do not count as “available”
May reduce the availability numerator
Can significantly impact the final availability %
Are not visible in user export files

As a result, availability may decrease even if:

There are no breakdowns
No communication failures are displayed
AP/WS appear normal

5. Example Interpretation

If a turbine shows:

144 total timestamps
139 timestamps with Status = 1
5 timestamps with Status = 3

Raw availability = 96.53%

However, if additional timestamps were filtered out due to missing critical tags, the number of valid "available" timestamps decreases, which can produce a final availability of 87.75%, even though only 5 Status‑3 intervals appear in the export file.

Additional Information

Machine Availability is an operational metric, not a performance metric.
Only the Status tag determines availability for this OEM.
Filtering explains gaps between exported Status data and the final availability percentage in the portal.

If additional clarification is needed, please feel free to contact the support team.

Document Location

Worldwide

[{"Type":"MASTER","Line of Business":{"code":"LOB77","label":"Automation Platform"},"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSV0CLD","label":"IBM Maximo Renewables"},"ARM Category":[{"code":"a8mgJ0000000ASPQA2","label":"Analytics-\u003EData Governance"}],"ARM Case Number":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":""}]

Tips