A model-based enterprise (MBE) is not simply an organization that replaces engineering drawings with 3D models. It is an approach to managing and using digital product information across the lifecycle—from systems engineering and design through manufacturing, inspection and sustainment. For industrial automation teams, the central challenge is making that information usable across people, software and factory equipment.
The 2018 Model-Based Enterprise Summit, hosted by the National Institute of Standards and Technology, brought together work on this challenge under the theme of an integrated lifecycle. Its proceedings span systems engineering, model-based definition, manufacturing, quality and inspection. Together, these topics point to a practical lesson: digital transformation depends on connected processes and well-managed information, not on adopting a single modeling tool.
What a model-based enterprise needs to connect
In an MBE, digital artifacts must support decisions beyond the team that created them. A product model may contain engineering information, but downstream users also need appropriate ways to access, interpret and validate that information. Manufacturing engineers, operators, quality teams, suppliers and sustainment organizations may have different tools and responsibilities, yet they rely on consistent product data.
This is where the digital thread becomes important. It provides a way to relate information across stages of a product’s lifecycle, rather than treating design, production and maintenance as disconnected data islands. Summit presentations described the thread extending from engineering and manufacturing into factory operations, where digital information can be provided to shop-floor personnel and used to compare as-designed and as-built configurations. Connections among systems such as product lifecycle management (PLM), manufacturing execution systems (MES) and enterprise resource planning (ERP) were also identified as part of the broader integration challenge.
For automation leaders, the implication is that connectivity must serve defined work. Linking systems is not an end in itself: the information needs to reach the right users and devices in a form they can use.
Model-based enterprise implementation starts with use cases
Model-based practices can create value in different ways, but implementation should begin by identifying the problem to solve. A systems engineering model, a manufacturing model and a model used for inspection do not necessarily answer the same questions. The required level of detail, data structure and validation approach depend on the intended use.
That principle is illustrated by work on 1D computer-aided engineering (1DCAE) for mechatronic products. At the conceptual design stage, teams may need to explore product functions before detailed geometry is available. The described modeling guidance starts by clarifying the target function and purpose of the analysis, then defining the model’s scope, parameters and required accuracy. A gear train model intended to examine rotational transmission, for example, may need only a gear ratio; analysis of vibration requires a representation of gear-tooth stiffness.
The process continues through implementation, verification and validation. Verification checks whether the model operates as intended across design-variable ranges and whether its results meet accuracy expectations. Validation compares model behavior with measurements from a subsystem, machine or experimental setup. Models also require documentation, promotion to their intended users and ongoing maintenance as design needs change.
This sequence offers a useful discipline for automation projects: define the decision a model will support before investing in its construction, and plan for validation and ownership as part of deployment.
Interoperability makes digital information usable
A lifecycle model is useful only when downstream systems and users can exchange and interpret its information. The summit proceedings describe this in both factory integration and technical-data delivery.
One factory-level interoperability effort explored using MTConnect and a bridge to the Robot Operating System (ROS) to coordinate robot teams and machine-cell devices. The aim was to investigate open technologies for integrating equipment and supporting applications such as in-process inspection, part management and machine servicing. The work reflects a recurring automation concern: devices may communicate effectively inside an individual machine or cell, while integration across a facility remains more difficult.
Another project addressed the delivery of engineering data to users who do not work in CAD. Its proposed workflow connected PLM with tools for generating viewable 3D PDF technical data packages and for converting and checking CAD data. The intended users included organizations involved in provisioning, cataloging and sustainment. The broader point is that model-based definition must account for how non-CAD roles consume technical information, not just how designers author it.
Standards, data exchange practices and verification procedures help establish that foundation. They also make it easier to define responsibilities when information moves between suppliers, program offices, manufacturing sites and other partners.
From connected data to adaptive operations
Factory sensors generate data, but collection alone does not establish operational value. The summit’s manufacturing discussions emphasized the need to determine how production data can support analysis, process and operations modeling, and digital representations of manufacturing activity. They also identified workforce and infrastructure needs as considerations in developing next-generation digital twins.
Related work on evolvable assembly systems considered how production resources might respond to changes in products, processes or operating conditions. Its shared-system-model concept provides a common basis for coordinating resources and supporting distributed decision-making. The research described applications at different levels, including fixtures, end effectors, workstations and assembly cells. This illustrates why consistency of context and information matters when systems are expected to adapt: individual resources need a coherent understanding of the wider production system.
Practical priorities for automation teams
Organizations moving toward an MBE can use these themes to guide implementation:
- Choose a specific use case. Define the users, decisions and lifecycle stage the model or data connection must support.
- Set the required information and accuracy. Avoid modeling detail that the application does not need, while documenting assumptions and limits.
- Plan verification and validation. Establish how digital data will be checked and, where relevant, compared with physical measurements.
- Design for downstream access. Consider how operators, quality personnel, suppliers and sustainment teams will view and use the information.
- Address interoperability and maintenance. Specify how systems exchange data and who will update models and interfaces as requirements evolve.
Building an integrated lifecycle
A model-based enterprise takes shape when trusted digital information can support work across organizational and technical boundaries. The proceedings show that this requires coordinated attention to model purpose, data quality, interoperability, user access and lifecycle maintenance. For plant and engineering leaders, the practical priority is to connect these elements around real operational needs. That is how digital models and factory data can become useful inputs to engineering, production and quality—not isolated assets in separate systems.