Introduction

In an era now being defined by the power of data-driven decision-making, we stand at the forefront of a transformative future, one where the vision of universal predictive maintenance becomes a tangible reality. Large enterprise companies are now championing this vision, emphasizing the need for predictive maintenance models that cater to the diversity of machinery on our shop floors, offering real-time, accurate insights into equipment health. This ambitious goal, shared by many in the industry, underscores a fundamental requirement: the creation and adoption of data that is universally meaningful — a common language for machine data, understood and utilized by all, not confined to individual corporate interpretations.

Here, is where MTConnect steps into the spotlight, and TAMS Model-Aware™ MTConnect makes it scalable and simple. Model-Aware™ technology not only ensures your data is correct, but it also simplifies deployment because it automates the modeling of a device into a semantic information model that makes data-to-information translation at scale possible. Most importantly the model is universally relevant, being recognized as an ANSI standard, and entirely open source so all companies can use it as their own underlying information model or create a translation layer (using adapters) to make their proprietary models universally compatible.

Understanding the MTConnect Information Model

MTConnect's role in modern manufacturing is not about data transmission; it's about translating data into a universal language. Imagine a scenario where every machine on your shop floor actually can speak to you, but they all speak different languages. The challenge isn't just in listening to the speech (data transmission and retrieval), but in understanding and interpreting what's being said into a language that everyone in your organization can comprehend. This is precisely what MTConnect does – it normalizes data from diverse machines into a standardized representation of information, making it universally understandable and actionable.

The Misconception of Direct Data Retrieval

One thing we have seen with some IoT providers in industry is that some are wearing the statement, "We don’t need MTConnect," as a badge of honor, boasting that they access “real-data directly from the controller.” This claim suggests that bypassing MTConnect leads to a purer form of data. However, is this approach genuinely more efficient, or does it miss a crucial point about the role of MTConnect in industrial data management?

The reality is that such claims overlook the essence of MTConnect or, worse, signal an attempt to create a vendor lock-in. Consider the example of Fanuc controllers. Data retrieval from these controllers universally employs the Fanuc Focas API. Whether or not this data is then mapped to MTConnect is a decision left to the solution provider. Those opting out of using MTConnect are almost certainly creating their own proprietary information model (language) for interpreting this data in their applications. It’s important to note that storing data in the format provided by Fanuc Focas is impractical and incompatible with efficient database structures. So, these companies are almost certainly not storing the data in raw Focas structures because it really wouldn’t work.

Hence, companies not converting to MTConnect are likely unaware of the real problem it solves. The critical point here is that MTConnect is not a data extraction method like the Fanuc Focas Api; it’s an Information Model that can be used to map the data extracted from Focas to make it universally meaningful. Providers opting to access Fanuc Focas and mapping to their own Information Model, instead of MTConnect, are opting to make the data meaningful only to their own application.

The idea that bypassing MTConnect gets you closer to the “real-data” is a silly misconception. Without MTConnect, data remains trapped in vendor-specific formats, limiting its utility and interoperability.

TAMS Approach and Non-proprietary Data Models

At True Analytics Manufacturing Solutions (TAMS), our approach also involves retrieving data directly from controllers (just like the guys that get the “real-data”), but we take a crucial additional step - mapping this data to the MTConnect Information Model. This method ensures that the data, once in the hands of the user, is not just a made-up data structure and semantics created by our developers for interpretation by only our application and long-term storage. Instead, it becomes a part of a structured, standardized dataset, universally significant and usable across various platforms and applications (not just ours).

This approach is more than just a technical choice; it's a commitment to open, standardized, and interoperable manufacturing data practices. By aligning with MTConnect, we ensure that our clients are not confined to proprietary data formats, thereby avoiding vendor lock-in situations. In the larger picture, the adoption of MTConnect as a universal language for industrial data makes the entire manufacturing ecosystem more cohesive, efficient, and ultimately enables the possibility for the universal predictive maintenance solutions large enterprise providers are now championing. A broad acceptance of the universal data-to-information translation (especially by these leading enterprises) is the most crucial step we can take towards achieving any solution that claims to provide universal predictive models for industrial device data in manufacturing.

When vendors avoid MTConnect and use their proprietary data models, they are essentially creating a system that makes sense only within their specific application. This approach leads to data silos where the information, though available, becomes limited in its utility outside the vendor’s specific system. Such data models, being custom-made, lack the universal applicability and interoperability that MTConnect brings to the table.

The Assurance of Model-Aware™ MTConnect Solutions

To delve deeper, it's not just platform providers who often misunderstand the main purpose of MTConnect as an Information Model. This extends across the Industrial IoT spectrum, encompassing machine builders, PLC manufacturers, academics, and even most MTConnect adapter developers. These stakeholders often simplify the MTConnect Agent, treating it merely as a tool for transferring XML data, and in doing so, they miss the fundamental goal of the standard. This oversight includes neglecting the critical role of precise semantics in MTConnect - that is, the specific labeling, categorization, and organization of data. This involves using accurate data tags, types, enumerations and structures to ensure that the information is not only correctly conveyed but also meaningful and interpretable in the broader context of industrial IoT.

Consequently, integration challenges are abound today even with widely available solutions claiming to provide data through MTConnect. Most of this is due to non-compliance with the MTConnect Information Model in solutions purporting to adhere to it. This is where TAMS approach comes in and the idea of Model-Aware™ MTConnect. The surest path to accurate and consistent mapping of machine data to the MTConnect Information Model lies in our Model-Aware™ commercial solutions, solutions that implement our open-source AdapterSdk or the small number of other forward thinking companies who are also implementing automatically transpiled code from the MTConnect Information model. Model-Aware™ guarantees that both the foundational source code and the software object model originate from code automatically transpiled from the MTConnect SysML model. This approach not only ensures adherence to the MTConnect standard, but it greatly reduces the development complexity by making the modeling automatic and eliminating the requirement of creating a translation file (the place where errors most frequently occur).

Why Choose Model-Aware™ MTConnect?

1. Universal Data Language: Model-Aware™ MTConnect ensures that data from all of your machines is correctly translated into a universally recognized format, making it comprehensible and actionable across various platforms. If it isn’t Model-Aware™ MTConnect it’s very possible that it is flawed in some way (significant or nuance), but the impact won’t be known until you have a use-case and you realize it wasn’t correct from the beginning.
2. Future-Proof Data: The data normalized to the MTConnect standard remains relevant and usable in the long term, irrespective of changes in technology or platforms. But, to be truly reliable and future-proof the data needs to be 100% correct, which is only guaranteed if it was developed with Model-Aware™ foundational source code.
3. Interoperability: With data in a standardized format, integrating with different systems, applications, and processes becomes seamless, fostering a more cohesive and efficient manufacturing environment.
4. Universal Predictive Models: The only ANSI recognized information model for Manufacturing equipment is MTConnect, which makes it universally significant. The universal predictive maintenance models of the future will almost certainly talk MTConnect, but they likely won’t talk the proprietary language of the many Industrial IoT providers who chose to make up their own models. MTConnect for future universal predictive models will be like English is for ChatGPT. Additionally, TAMS Model-Aware™ MTConnect will ensure that you don’t confuse the model with terminology that it never heard before.

Conclusion

In this era of data-driven innovation, the adoption of MTConnect, particularly the Model-Aware™ MTConnect being introduced by True Analytics Manufacturing Solutions (TAMS), represents a significant leap forward in universalizing industrial data. This approach transcends traditional limitations, ensuring data from diverse machinery is not only uniformly interpretable but also actionable across various platforms. By aligning with MTConnect, TAMS commits to open, standardized, and interoperable manufacturing data practices, paving the way for a cohesive, efficient manufacturing ecosystem. The universal predictive maintenance models, championed by leading enterprises, hinge on this very adoption of a universal data language, solidifying MTConnect's role as a cornerstone in the future of industrial IoT and predictive maintenance. As we embrace this transformative technology, we step closer to a future where our machines not only speak a common language but also contribute to a more integrated, intelligent, and innovative manufacturing world.