Data quality

Data quality commonly refers to how accurate, complete, consistent, timely, and reliable data is for its intended use. In industrial and regulated manufacturing environments, it describes whether data captured across OT systems, MES, ERP, PLM, QMS, and supporting tools can be trusted for production control, compliance, analytics, and decision making.

Key characteristics of data quality

While specific models vary, data quality in manufacturing typically considers whether data is:

• Accurate: Correctly represents the real-world value (for example, actual torque values, lot numbers, inspection results).
• Complete: Required fields and records are present (no missing traceability links, signatures, or inspection records).
• Consistent: Aligned across systems and time (the same part, revision, or NC ID is represented the same way in MES, ERP, and QMS).
• Timely: Available when needed for operations, release decisions, reporting, and audits.
• Valid: Fits defined formats, ranges, and business rules (for example, date formats, controlled vocabularies, pass/fail codes).
• Traceable: Can be linked back to its origin, including who created or modified it and under what conditions.

How data quality shows up in industrial operations

In regulated manufacturing, data quality is closely tied to execution control, quality management, and compliance. Typical touchpoints include:

• Production records: Electronic travelers, batch records, and build histories that must be accurate and complete for each unit or lot.
• Traceability and genealogy: Correct mapping of materials, components, process steps, tools, and operators to final assemblies.
• Inspection and test data: Reliable measurements, test results, and dispositions for FAI, in-process checks, and final inspection.
• Nonconformance and CAPA data: Clear, consistent defect codes, root causes, and actions that support analysis and closed-loop improvement.
• System integration: Aligned master data and transaction data across MES, ERP, PLM, QMS, and LIMS so that no conflicting or duplicated records are created.
• Audit evidence: Records that are complete, legible, and logically connected, making it possible to demonstrate what happened, when, and under which instructions and revisions.

What data quality is not

Data quality is about the condition and fitness of the data itself, not about network security or system performance.

• It is not the same as data integrity, which often focuses on preventing unauthorized alteration and ensuring records are attributable, legible, contemporaneous, original, and accurate over time.
• It is not a specific software product; it is a property of data that can be influenced by processes, controls, training, and tooling.
• It is not limited to analytics. Poor data quality can affect real-time routing, releases, purchasing decisions, and maintenance planning.

Operational drivers of data quality

Common operational factors that influence data quality in manufacturing environments include:

• Standardized data models and master data: Clear definitions for parts, revisions, routings, work centers, and defect codes.
• Controlled work instructions and forms: Structured data entry, required fields, and validation rules within MES, QMS, or electronic DHR systems.
• System interoperability: Mapped identifiers and controlled integrations between ERP, MES, PLM, and other systems to reduce manual re-entry.
• Sensor and equipment calibration: Measurement systems analysis and calibration practices that help ensure captured values are accurate.
• Governance and stewardship: Defined responsibilities for maintaining and reviewing critical data sets, such as part masters, BOMs, routing, and supplier records.

Common confusion

• Data quality vs. data integrity: Data integrity often focuses on protection, attribution, and audit trails; data quality focuses on correctness, completeness, and usability. In regulated environments, both concepts are related and sometimes discussed together but they are not interchangeable.
• Data quality vs. data governance: Data governance describes the overall policies, roles, and processes for managing data. Data quality is one of the outcomes data governance aims to manage and monitor.

Manufacturing-focused example

In an aerospace plant, a data quality issue might occur if the torque values recorded on an electronic traveler are entered in the wrong units or linked to the wrong serial number. Even if the system is secure and tamper-evident, that record has poor data quality because it does not reliably represent what happened on the shop floor. Correcting this may involve changes to work instructions, field validations, and how tool data is integrated into the MES.