Data Integrity Matters
Klaus M. Blache | July 1, 2021
How much confidence do you have in your organization’s data when you need to make tough decisions?
Is the data you’re collecting useable (reliable and complete)? High-risk operations such as food products, nuclear, pharmaceutical, and airlines are keenly aware of the consequences of unreliable data. While other manufacturers don’t necessarily carry the same risk, unreliable data can have a significant impact at several levels.
U.S. Air Force core values are a cornerstone of why their maintenance missions are successful. Integrity in the aircraft-maintenance world addresses not only the individuals performing their duties, but aspects such as equipment and information. Reliable information enables good decision making while incorrect information results in missed goals or failure. The lack of accountability to uphold standards was prevalent in accidents, incidents, and documentation according to Chief Master Sgt. Douglas Ackerman, 43rd Airlift Wing Commander (published 12-3-2009). He wrote that, “Our core values were simplified to three short statements more than a decade ago: Integrity First, Service Before Self, and Excellence in All We Do.” For integrity, the simple definition was stated as doing the right thing when no one else is looking. Service before self is about doing one’s job and recognizing that the team can only succeed if each member does their job. Excellence in all we do addresses high standards of performance and sustaining it.
Here are signs of insufficient data integrity:
• Work performed not entered in the CMMS.
• Costs are not correctly assigned to assets.
• Not all maintenance events have accurate data.
• All events are recorded in the CMMS, but the data quality is poor.
• Metrics are not clearly understood, so values between departments or facilities are not fair comparisons.
• Your shipped maintenance training package is nowhere to be found. It arrives three days late. Tracking still shows it being processed in Atlanta and doesn’t know it arrived (this recently happened to me).
• PMs are skipped because there’s not enough time left in the month. Too often they are counted as completed.
• Parts are taken out of the crib for an emergency repair, but not recorded.
• Very old backlog jobs are just taken out of the system.
• You have kitted parts, the job is cancelled, and parts are not returned and logged in.
• Predictive-technology implementation improvements aren’t realized, but you’re not consistently collecting and analyzing sufficient data.
• Fault codes used are too vague and not entered consistently.
• Operations and maintenance have conflicting key process indicators, skewing data.
• Unclear guidelines result in inconsistent data entry by workers.
What is data integrity? It’s assuring that data is accurate, readable/understandable, consistently applied, retrievable, and correctly formatted over the life cycle of the data intent. If your data is important and you plan to use it for decision making, then you should audit it for accuracy. If people don’t trust your data, then what does that say about other things that you do?
With data usage and storage growing exponentially, it’s increasingly important to ensure the integrity of that data. When you have data that you trust, you can, with confidence, tap into the full potential of analytical tools and machine learning/artificial intelligence. EP
Based in Knoxville, Dr. Klaus M. Blache is director of the Reliability & Maintainability Center at the Univ. of Tennessee, and a research professor in the College of Engineering. Contact him at firstname.lastname@example.org.