QIF Open Standard Supports Digital Transformation of Quality Inspection Processes

Author: Dimensional Metrology Standards Consortium/DMSC

The term “digital transformation” is used in business to describe the use of digital technology to radically improve the performance or reach of enterprises. It is, at its core, about process digitization and the strategic use of information to drive new value.

QIF (the Quality Information Framework) is an open ANSI standard for manufacturing, designed to support the digital transformation of quality inspection. QIF is a major step forward in enabling fully-digital quality inspection processes, and in supporting the strategic use of quality information throughout the enterprise and the product development process.

Fundamental Challenge
The fundamental challenge in transforming quality inspection into a fully digital process is interoperability. While interoperability is often thought of as merely the ability to exchange computer processable data between different components of a system, in the case of quality inspection, this is not enough. To be fully useful, the data must flow seamlessly into, through, and out of the system. If you examine a quality inspection system from a high-level perspective, it can be thought of as supporting four major activities, each implemented as a number of components:

What is not shown in the figure is that the major inputs to quality inspection are from design and manufacturing. In fact, quality inspection is where your design and manufacturing processes are made accountable.

With a fully capable quality inspection system, the inspection results for any characteristic of a part are not only traceable to the measuring equipment used to generate them, but also to the manufacturing processes used to create the characteristic, as well as the features in the authority data-set (CAD master model) used to define it.

This level of accountability cannot be achieved with a piecemeal approach to interoperability, focused on just
exchanging data. Interoperability and connected data is only possible with an integrated approach, focused on
semantics, traceability, validation, and the seamless flow of data. This integrated approach is what QIF is all about and showcases its benefits.

Enabling Digital Quality
A prerequisite for digitizing any process is to create a vocabulary (or ontology) of its terminology and semantics. This vocabulary must be carefully modeled, based on a thorough knowledge of the problem. Consider this simplified view of a typical quality inspection process, using digital measuring equipment (e.g., a CMM).

For each activity in the workflow, there is associated information, which may be inputs from external sources,
knowledgebase information, or outputs. This figure shows the most prominent categories of quality information associated with each activity:

QIF defines an extensive framework of interoperable data models, providing coverage of virtually all quality
information and semantics used in quality inspection workflows. It supports semantic interoperability throughout the extended quality inspection process, providing the flexibility to seamlessly combine applications from multiple vendors. QIF promotes quality inspection potential for innovation in areas such as planning, analysis, and optimization.

QIF also respects formal and de facto standards:
• It is implemented in standard XML, to make integration with other systems easy
• It interoperates seamlessly with DMIS, the international standard for digital measuring equipment programming
• Its product definition language (QIF MBD) is designed to work seamlessly with 3D CAD systems and other sources of product models, such as scanned data and drawings.

Digitizing Product Definition
Quality inspection doesn’t happen in a vacuum. It’s part of the larger product development ecosystem, including, most notably, design and manufacturing:

While quality inspection consumes data from multiple sources, its most critical input is the authority product model (or, more generally, dataset), describing a component or product, as designed. These data sources come in multiple forms, including drawings, scans of physical models, 3D printing files, triangulated meshes, and, most importantly, 3D CAD models—with or without PMI.

The QIF MBD data model:

  • Supports multiple geometric representations, including wireframe geometry, faceted 3D, precise 3D NURBS boundary representation (BRep), and 3D point clouds
  • Provides traceability to the authority dataset, including support for digital signatures
  • Supports direct mapping of geometry, topology, PMI, and annotations to major CAD systems (CATIA, NX, Creo, SolidWorks) and neutral formats (STEP AP242)
  • Supports fine grained object-level traceability to CAD model internal object identifiers
  • Allows the tracking of version differences in authority data
  • Includes extensive validation criteria
  • Includes full support for feature-based semantic PMI
  • Is expressly designed to enable the use of model-based definitions in digital data-driven Model-Based
    Enterprise (MBE) processes

QIF MBD is designed to support many use cases, including:

  • Authority datasets that are not machine processable
  • Authority datasets that are accessible only through an API
  • Authority datasets that are accessible only through derivative files
  • Authority datasets that are incomplete
  • Authority datasets that are revised/updated
  • Authority datasets without PMI
  • Authority datasets with partial PMI

Because QIF MBD has the capacity to fully represent the geometry, topology, PMI, and annotations of major native CAD file formats, and supports traceability, validation, and digital signing, it can serve as a trusted derivative file format for applications beyond quality inspection. Because it is a fully documented ANSI standard, it can even serve as a long-term archival format.

Communicating Via the Digital Thread
The notion of a digital thread is generally thought of as a communications framework, connecting traditionally siloed elements in manufacturing processes, and providing an integrated view of an asset (e.g., a component) throughout its lifecycle.

While fully digital data-driven processes are a prerequisite for supporting digital threads, they’re only a
starting point. A key requirement is end-to-end traceability.

QIF includes extensive support for traceability, not as a consequence of design, but rather by intention because traceability is an inherent requirement of quality processes.

QIF implements traceability through built-in traceability elements, validation criteria, persistent identifiers, digital signatures, and semantic links to source (authority) data sets.

A single feature on an individual component can be traced back through its inspection process, through its
manufacturing process, and even back to the version of CAD file and specific geometry that was used to define it. Or, a single feature in a particular version of a CAD file can be traced forward, through the manufacturing and inspection process, to the specific components built using it.

The Digital Thread allows for fine grained feedback and feed-forward between design, manufacturing, and
quality inspection – enabling design for manufacturing, design for inspection, and many other possible process

Exploiting the Digital Twin
Dr. Michael Grieves, who introduced the concept of a Digital Twin, defines it as:
“A set of virtual information constructs that fully describe a potential or actual physical manufactured product from the micro atomic level to the macro geometrical level. At its optimum, any information that could be obtained from inspecting a physical manufactured product can be obtained from its Digital Twin.”

Digital Twins may contain a variety of information, including, but not limited to, fully annotated 3D models, bills of materials (listing current and past components), bills of processes, inspection results, service records, and operational states captured from sensor data.

The value of a Digital Twin comes from using the information it contains in innovative ways to gain insight and create actionable information. Implicit in this is a requirement that the information is structured in a way that it can actually be used throughout the product’s lifecycle.

While no single data format is extensive enough to represent all the information that might be contained in a Digital Twin, QIF provides significant coverage, including:

  • Fully annotated 3D models, with precise and lightweight representation, semantic PMI, and links to the
    authority (master) CAD model
  • Bill of characteristics (BoC) representing criticalities and measurable characteristics
  • Quality inspection and manufacturing process information
  • Nominal (as designed) and actual (as measured) values for physical characteristics
  • Extensive support for traceability, with unique identifiers and digital signatures
  • Well-documented XML schemas, designed to support easy data reuse and integration
  • Completely open standard

QIF provides an optimal data structure to represent physical part and assembly data in a Digital Twin. And, because QIF data is created as part of quality inspection processes, there is no incremental cost to use it as the foundation for a Digital Twin.

Designed for Openness
QIF was designed recognizing that information is a valuable resource and a strategic asset for manufacturers, their partners, and their customers. In order to take full advantage of its information resources, manufacturers must manage information as an asset throughout its lifecycle to promote openness and interoperability. Managing information as an asset increases operational efficiencies, reduces costs, improves quality and services, and supports business goals.

QIF is built using industry standard XML technology, allowing it to be automatically processed and read by computer, without any loss of semantic meaning, using easily available tools. Its data structures are welldocumented,
non-proprietary, and designed to be accessible, discoverable, and usable by not just professional software developers, but also end users. As importantly, QIF is IT friendly.

QIF data is semantic, contextual, traceable, validatable, and, as a result, highly reusable. It is easily integrated
into smart data, big data, advanced analytics, and other enterprise systems.

For manufacturers, QIF is a robust and well-tested standard. At the 2014 and 2016 IMTS exhibitions in Chicago, a group of vendors demonstrated perfect interoperability between multiple commercial-off-the-shelf QIF enabled

For Software Developers, QIF is designed to solve the non-value-added problems associated with developing
quality inspection or MBD/MBE applications, allowing them to focus on innovation. The entire QIF data model is open for use, with no proprietary structures or hidden ‘gotchas’.

How to Get Started
QIF was developed by the Digital Metrology Standards Consortium (DMSC), a standards body comprised of quality and engineering software experts from industry, academia, and government.

For more information: www.qifstandards.org