Typical Failure Data Challenges

No system of record for equipment

Data is maintained in multiple disparate manual and electronic repositories with no processes in place to ensure consistency in tagging or synchronization of tags and attribute data between repositories.

Problems include:

  • No complete equipment list in any one system or data source
  • Inconsistent identifiers for the same equipment in different systems
  • Inconsistent identifiers for equipment characteristics in different systems
  • Characteristic data values can be different in different repositories (this was not validated as part of this assessment, but is almost certainly an issue)
  • No automated synchronization or exception checking processes exist to identify and/or reconcile differences between repositories

No consistent process for defining equipment use/location data

Use/Location data are attributes that describe service conditions in which equipment is used (also known as operating context). These data give apples-to-apples comparison of equipment performance by allowing users to specify detailed criteria for selecting equipment in similar service when generating failure metrics.

Many companies mistakenly believe technical object structuring differences prevent comparisons between different facilities. However, structural differences will not affect merging and assessment of data from different locations provided there are comprehensive use/location data for equipment and primary equipment reporting levels and equipment boundary envelopes are defined within the technical object structure. The functional location structure itself cannot represent use/location data. Instead, this is done via classification of individual tag numbers. The upper part of functional location structure is comprised of grouping levels we can call functional area locations (FALs). The FAL structure is a framework for cataloging functional equipment locations (FELs) (FELs correspond to P&ID tags).

The FAL structure is like a phone book. A phone book organizes people so you can find them. It doesn’t tell you things about the people (age, race, income, etc.). Likewise the FAL structure cannot tell you a piece of equipment is used for hot standby purposes, has been de-rated to 75%, and operates in an air-conditioned building.

No structural definition of equipment boundary envelopes

Equipment structures are logical representations of equipment interrelationships, but are inconsistent with ISO 14224 equipment boundary envelopes. This skews breakdown metrics unless special selection criteria are entered on an equipment-by-equipment basis.

No differentiation between equipment functional and physical aspects

Functional location structures typically extend to facility areas only (in almost no cases do functional location structures extend to the equipment level). Equipment structures pick up where functional location structures leave off. Only one object type (equipment) is used to represent both functional and physical aspects of equipment (versus the Relationship of Coincident Individuals promulgated by ISO 15926-2 (ISO 15926-2, 2003). Ramifications include:

  • Functional characteristics must be assigned to the equipment object because there is no other object where they can be assigned
    • This greatly complicates failure metric reporting and creates a large volume of data maintenance requirements related to equipment moves
  • Anytime the equipment object moves, the following must be updated to reflect process conditions at the new point of installation:
    • All functional characteristics
    • Equipment risk data and criticality ranking
    • Preventive maintenance scope and/or frequency
  • Functional equipment breakdown metrics are meaningless when service conditions change for given technical objects.
    • This type of reporting is the primary requirement for equipment reliability reporting.
    • Meaningful breakdown metrics in SAP would require complex custom reporting and needless complications in data architecture, e.g. equipment characteristic change tracking).
  • SAP Equipment IDs do not match P&ID tag numbers; P&ID tags are entered in the equipment Technical ID field, a free-text field with no data validation
    • When equipment objects cannot be easily located, users revert to high-level facility areas as reference objects for notifications
    • Data joins (need for system-generated analytics) are ineffective when used with free text fields having no validation

Incomplete and inconsistent equipment failure data set

The equipment failure data set is such that meaningful reporting of basic failure metrics is not possible without data mining efforts. Breakdown reports will not deliver meaningful results at any level (corporate, region, business unit, or single equipment item). This is due to following issues:

  • Reference objects are not equipment or component specific: many notifications are assigned to functional locations that serve as grouping levels
  • Many failure data required for equipment breakdown reporting are missing, including malfunction start and end dates, breakdown duration, and equipment start-up dates
  • Lack of a consistent approach to setting the breakdown indicator and general misunderstanding of how the breakdown indicator should be used
  • No definition of a reporting level, lack of standard equipment boundary envelope definitions, and no process for rolling-up component-level failures to the parent equipment unit (reporting level)
  • Equipment operating context is unavailable
  • Failure codes are from an outdated version of ISO 14224
  • Failure consequences are not being captured as part of failure reporting
  • Repair costs are inaccurate or non-existent

No classification of failure events

Classification of failure events is used to capture consequences of failure over and above equipment repair costs. Production loss and HS&E incidents that are equipment-related should be ascribed to the specific failure events that caused the consequences. In addition, near-miss data should be captured.