A successful digital transformation program steps an organization through a process that transforms its current state to a desired future state. For production operations, this often involves moving from a local, siloed procedural operating model to a more comprehensive model that is integrated to the enterprise, responsive to customers and market changes, and optimized in the context of overall business performance.
In practice, this often means thinking about production operations in a new way. It means stepping away from the practices and models that have enabled an organization to operate as well as it does today – but may need to change to achieve tomorrow’s goals.
Know the Limitations of Your Conceptual Models
Many industrial companies have relied upon derivatives of The Purdue Enterprise Reference Architecture: A Technical Guide for CIM Planning and Implementation, introduced in 1992 by Theodore J. Williams, as the basis for establishing processes and data flows for plant sensors, equipment, and controls and software systems, and for integrating to business systems. The model has proved useful, and often serves as a basis for dis-cussing where plant systems fit and how they interact. But when it comes time to innovate or transform today’s production operations, the model’s limitations may work against such initiatives.
Limitations of the Purdue Model
Did you notice the term “CIM” in the title referenced above? CIM is an acronym for computer integrated manufacturing. If you remembered that, congratulations on your upcoming retirement! It is a term that dates to the time when business software ran in batch mode on mainframe computers and industrial companies were beginning to figure out how to utilize computers and some digital controls in their plants. While it has been quite helpful, the Purdue model embodies a view from the plant floor up, with limited externalities. At the top level, it focuses primarily on planning and logistics.
The model’s hierarchical structure was well-suited to the computing and networking systems available at the time. In the ensuing decades, it has proved useful for constructing the “plumbing,” or connections between systems in the plant. However, it may not be ideal for new classes of plant data and technologies emerging today. Today, computing and software are increasingly deployed at the edge, rather than just at the higher levels of the model. And the model doesn’t contemplate the possibility of introducing other connected technologies beyond the plant equipment and associated sensors. There is no provision for wearables, smart tools, vision or voice systems, augmented reality goggles, or the like.
This does not mean there is no value in the Purdue model or its more contemporary incarnations such as ISA 95. But don’t let its limitations constrain your digital transformation initiatives unnecessarily.
The 21st Century Industrial Operations Ecosystem
Leading industrial companies are actively engaged in transformation programs that will reshape their production operations to be more integrated, responsive, and optimized to meet business and customer needs. Realizing these innovations requires an understanding of the emerging 21st century operations ecosystem. Connected environments surface heretofore unavailable machine data, enabling new business models. New systems may monitor the assets and new actors may interact with the assets in new ways. Industrial plant operations, typically siloed and fairly isolated today, will be reshaped as the core of a 21st century industrial production operations ecosystem. Field operations, such as mining or agriculture, are experiencing a similar evolution. In other cases, such as automotive, where the assets operate in public spaces and both the assets and ecosystems are evolving quickly, other factors come into play.
In 21st century production operations, work is accomplished with a combination of internal and external actors (asset manufacturers, third-party machine monitoring services, spare parts suppliers, etc.). This puts new demands on data requirements and cybersecurity strategies. New types of data are being generated (from wearables, vision systems, machine health sensors, etc.) and newly available digital twins and ma-chine learning systems can work at various levels to optimize the over-all system in sync with the needs of customers and business operations.
Connected industrial products such as pumps, blenders, compressors, robots, and packaging machines will spend most of their respective as-set lifecycles operating in production plants or factories. Traditionally, using and supporting these assets has been the province of plant operations and plant maintenance and engineering personnel. They use available software tools and other methodologies to maximize the asset’s contribution to the production goals of the plant. These often include asset management or maintenance applications, as well as execution applications such as planning, tracking, quality, and reporting.
The availability of connected machines is potentially disruptive to these established production and maintenance scenarios. The figure, “The 21st Century Industrial Production Operations Ecosystem,” depicts this for industrial plant operations.
It is important to focus on the “operate” phase, because this is where the asset’s value is delivered. This phase is also where many emerging, disruptive technologies can have a transformative impact on operations and business performance. It is no longer enough to merely address the asset’s availability, reliability, and risk. It may also be appropriate to consider potential new business processes enabled by smart, connected assets.
It is useful to frame any analysis of production operations requirements for a product (asset) with a broad, comprehensive view of the 21st century ecosystem. Unlike the hierarchical (Levels 0–4) Purdue model, which posits a bottoms-up perspective focused within a plant and with a limited notion of business systems interactions (primarily planning and logistics), the 21st century perspective should explicitly consider multiple plants and the possibility of more robust interactions with business systems, engineering systems, supply chain systems, expanded services, and customers.
Unlike supply chain models (Design –) Source – Make – Deliver, or asset lifecycle models Design/Build – Operate/Maintain – Retire/Dispose, the primary focus here is on real-time, day-to-day production operations, where disruptive technologies are poised to make a significant impact, not only on efficiency, but also responsiveness and operating performance. It’s not that any of these models were wrong or no longer valid. To be sure, they have provided the framework to get us to where we are today. But they may not adequately capture the requirements going forward.
ARC Advisory Group clients can view the complete report at ARC Client Portal
If you would like to buy this report or obtain information about how to become a client, please Contact Us
Keywords: Digital Transformation, 21st Century, Operations Ecosystem, Transformation Framework, ARC Advisory Group.