Follow

Data dictionary for work order insights

This data dictionary provides definitions and calculations for fields in the work order insights reports and CSV files.

Fields in the reports

The following table list the fields in the work order insights reports:

Field

Report

Type

Definition

Calculation

Analyzed Closed Work Orders

Closed

Number

The number of closed work orders we analyzed that had risks.

Count(analyzed closed work orders where risk score >0)

Analyzed Open Work Orders

Open

Number

The number of open work orders we analyzed that had risks.

Count(analyzed work orders where risk score >0)

Code

Both

String

The code for the work order (from the CMMS). If you're logged in to the CMMS, you can click the code to navigate directly to that work order.

N/A

Delay Days

Both

Number

The total time (in days) that the work order was overdue (i.e. open beyond the suggested completion date).

In the report for closed work orders, the value in brackets is the expected delay, based on similar work orders.

Close date - suggested completion date

Description

Both

String

The description for the work order (from the Summary of Issue field in the CMMS).

N/A

Duration Days

Both

Number

The total time (in days) that the work order took to complete.

In the report for closed work orders, the value in brackets is the expected duration, based on similar work orders.

Close date - open date

Excess Duration

Closed

Number

The total additional time it took to close all work orders, beyond what was expected.

Sum((work order duration - expected duration) for all work orders at that risk level or site)

Excess Failures

Closed

Number

The total number of abnormal failures we detected. These are cases where corrective maintenance work occurred before the mean time between failures (MTBF), and they represent work that could have been avoided.

Count(work order where work order is indicated as being an abnormal failure)

Most Common Risk

Closed

String

The most common risk we identified for a risk level or site.

N/A

Risk

Both

String

The most serious risk we identified for this work order. To learn more about risks and how they're assigned, see Risks identified in work order insights.

N/A

Risk Score

Both

Number

A number between 0-1000 that represents the relative level of abnormality for this work order when compared to similar work orders.

Normalized machine learning (ML) score based on aggregate abnormality of all work order fields

Risk Type

Both

String

The risk level (high, medium, or low) that represents how abnormal a work order is compared to similar work orders.

0-333 = Low

334-665 = Medium

666-1000 = High

Site

Both

String

The site assigned to that set of work orders.

N/A

Fields in the CSV file

The following table lists the fields found in the work order insights CSV file:

Field

Type

Definition

Calculation

Assets

Number

The number of assets associated with the work order.

Count(assets)

Code

String

The code for the work order (from the CMMS).

N/A

Contributing Work Orders

String

The work orders we feel are highly correlated with the occurrence of an abnormal failure. Effectively, these work orders caused the abnormal failure to happen.

N/A

Delay

Number

The total time (in days) that the work order was overdue (i.e. open beyond the suggested completion date).

Close date - suggested completion date

Description

String

The description for the work order (from the CMMS).

N/A

Duration (Days)

Number

The total time (in days) that the work order was open.

Close date - open date

Expected Assets

Number

The number of assets we expected to be assigned to the work order, based on similar work orders.

The expected value that ML determined based on the work order cluster

Expected Delay

Number

The number of days we expected the work order to be delayed, based on similar work orders.

The expected value that ML determined based on the work order cluster

Expected Duration

Number

The number of days we expected the work order to be open, based on similar work orders.

The expected value that ML determined based on the work order cluster

Expected Steps

Number

The number of times we expected labor to be logged against the work order, based on similar work orders.

The expected value that ML determined based on the work order cluster

Expected Tasks

Number

The number of tasks we expected to be associated with the work order, based on similar work orders.

The expected value that ML determined based on the work order cluster

Expected Techs

Number

The number of technicians we expected to be assigned to the work order, based on similar work orders.

The expected value that ML determined based on the work order cluster

Risk Color

String

The color that represents the assigned risk type.

Red = High (666-1000 risk score)

Yellow = Medium (334-665 risk score)

Green = Low (0-333 risk score)

Risk Score

Number

A number between 0-1000 that represents the relative level of abnormality for this work order when compared to similar work orders. In other words, this number indicates how much a work order differs from similar work orders.

Normalized machine learning (ML) score based on aggregate abnormality of all work order fields

Root Cause

String

The abnormalities we found for that work order when compared to similar work orders.

ML algorithm computes both field-specific and holistic deviations for all work order fields that have data populated

Site

String

The site associated with the work order.

N/A

Steps

Number

The number of times labor has been logged against the work order.

Count(steps)

Tasks

Number

In the report for closed work orders, this is the number of tasks that were completed in the work order.

In the report for open work orders, this is the number of tasks that are associated with the work order.

Count(tasks)

Techs

Number

The number of technicians assigned to the work order.

Count(techs)

Root causes

The following table lists the possible values in the root cause column:

Root cause

Description

Abnormal failure

The asset failed earlier than the MTBF.

Asset count

The number of assets associated with the work order was different than for similar work orders.

Delay

The work order was overdue (past the suggested completion date) longer than similar work orders.

Estimated time

The estimated time for the work order was different than for similar work orders.

Maintenance type

The maintenance type associated with the work order was different than for similar work orders.

Site

The site associated with the work order was different than for similar work orders.

Spent time

The time spent (i.e. hours logged) on the work order was different than for similar work orders.

Step count

The number of times labor was logged against the work order was different than for similar work orders.

Task count

The number of tasks associated with the work order was different than for similar work orders.

Tech count

The number of technicians assigned to the work order was different than for similar work orders.

Unexpected configuration

The configuration for a number of fields was different than for similar work orders.

Unexpected duration

The number of days the work order was open (i.e. the number of days between opening the work order and closing it) was different than for similar work orders.

Was this article helpful?
0 out of 0 found this helpful

Comments