A Fresh Look at Industrial Automation Machinery
Readers looking for an explanation of “automation machinery” will likely come across a litany of articles providing the following same groups of system types: fixed, programmable, flexible and, in some cases, integrated. In our opinion, this is a classic example of marketing content that started with a common, older textbook definition, and became replicated by parties trying to rank for a topic without really understanding the involved underlying principles. We hope to provide a fresh perspective on the definition of automation machinery as used in industrial manufacturing, and to point out modern nuances that break apart the older definitions referenced above.
Where Older Definitions Fall Apart
The definitions below are what you might readily find in a web search for “automation types” or “automated machinery types,” but these terms are rarely used in actual practice today. When they are used, they have notably different meanings. Let’s examine each term below, the traditional definition that you’ll find for each in a Google search today, and why we feel these categories no longer apply.
| Traditional Automation Type | Traditional Definition | Why This Breaks Down |
| Fixed Automation | Also known as “hard” automation, fixed automation performs a repetitive series of tasks, is unchangeable and must be replaced when new tasks are needed. | In our opinion, this term is a carry-over from “hard-wired” or “hard-coded” systems of old, which relied on printed circuit boards, switching logic and onboard programs. Today, even single-task systems come with modifiable controllers, thanks to their low cost and ease of deployment. Hardwired systems of course still exist, but not because they categorically serve as the “entry level” option as they once did. |
| Programmable Automation | Performs batch production of varying products, requiring re-programming for product changes that are difficult and time-consuming, and having reduced performance based on the time needed to program change-overs. | Carrying on from the above, programmable automation is not limited to batch or multi-product systems these days. Also, these systems are definitely not difficult, costly or performance-impactful to change as they once were. |
| Flexible Automation | A more sophisticated form of programmable automation, allowing for automatic and quick change-over between different products. | Today, “programmable” and “flexible” automation are largely the same thing. We might consider the difference as simply including a recipe module or not: Systems with a recipe module can store, load and change production recipes from the machine interface screen (HMI), whereas those without need added logic added at the controller. These are not distinct systems anymore, but instead only a simple variant of the same platform. |
| Integrated Automation | Large system-wide or entire plant-wide automation that hands off most decisions to computer-controlled processes, eliminating most human involvement. | Arguably, every format of automation these days can be “integrated,” meaning that individual systems can coordinate with other systems, as well as collectively share resources to self-direct their own actions. This term is no longer an isolated, “top-tier” level of automation, especially not in a world where our $75 office coffee machines can re-order their own coffee. |
Overall, we find the above categories to be an outdated way of identifying or understanding automation systems. In our opinion, systems are best described by their primary functional application, which allows for a better understanding of what these systems actually do, as opposed to how the automation is arrived at internally. This divorces the system architecture design from the actual application. For example, a system using “programmable automation” can serve single-task, recipe-driven, continuous, batch, single-unit, high-volume or even custom one-off applications, and everything in between.
To drive the point home: A customer would not call an automation partner and request a quote on “one flexible automation system please.” Instead, the customer would describe the application they’re solving, such as “nut driving” or “leak testing.” The automation machinery vendor would know these to be either mechanized tool process or evaluation machines, respectively. The automation partner would then discuss details that would arrive at a controls architecture based on how complex, customizable and variable the machine needs to be.
Automation Machinery Defined
To step away from the above traditional definitions, let’s turn to Merriam-Webster dictionary to arrive at a plain-English starting point:
- Automation: The technique of making an apparatus, process or system operate automatically.
- Machinery: A set of powered devices that function to perform a collective task.
So, automated machinery can be defined as a powered device, process or system that performs a task automatically. That’s a very functional definition! Automated machinery simply completes work under power at the initial direction of a human, self-directing until its defined task is complete.
The next logical way to categorize automated machinery then is by what type of task it is able to complete, knowing that modern automation technology offers multiple different internal ways to arrive at that task type. The old definitions of automation types may have had clear lines of delineation between what each type was able to perform, but now, we can freely float across lines as a function of budget, technical objectives and long-term business goals.
To help define modern categories by application type, let’s consider the below automation categories:
- Physical motion systems
- Mechanized process systems
- Inspection and evaluation systems (aka pass/fail systems)
- Automated data systems (aka decisions systems)
Physical Motion Systems
Systems that primarily serve to physically manipulate the location, orientation or position of a product can be categorized as physical motion systems. This type of automation machinery can be found in nearly every industry, and makes up the fundamental building blocks of automated manufacturing, replacing physical human labor moving materials around.
Examples include conveyors, automated warehouse retrieval systems, robotic positioners and case packers.
These systems can be extremely simple in their automation, such as a simple straight conveyor length that runs continuously to move boxes from one end of a warehouse to another. These systems can use on-machine motor drives that simply turn on or off by a switch or sensor. Alternatively, these systems can be highly complex, utilizing incredibly sophisticated multi-axis robots with motion-centric logic controllers, separate safety controllers, ultra-precise positioning encoders and a host of data analytics that perpetually optimize motion paths.
Mechanized Process Systems
Any system that modifies, augments or otherwise performs “work” on a product can be categorized as mechanized process systems. From single-task machines up to entire production lines, automated systems that produce some discernible change in a material component are said to “process” that component.
Examples include plastic welding systems, thermal cooling tunnels and steam boilers.
In many ways, automation approaches to process machinery can be the most diverse out of our four categories here. A simple machine that drills a quarter-inch hole in a single wooden part, one at a time, may need no more than a few relay logic circuits and a start/stop button to perform the task. This might have been considered “fixed” automation in earlier times. However, we can also arrive at the same functionality using hobbyist-grade electronic controllers (a Raspberry Pi is a great example), which are technically programmable as well as able to be integrated. At industrial scales, we might consider an entire modern process line to be an aggregate mechanized process system controlled by a master programmable automation controller, performing all work steps automatically from inputting raw ingredients to outputting a finished, packaged, ready-to-ship product.
Inspection and Evaluation Systems
Inspection and evaluation systems are employed as checkpoints throughout modern manufacturing lines, and serve the purpose of assessing products for conditions that can lead to a “pass or fail” determination. This type of automation machinery works like the old-fashioned manual inspections that used human senses to deem a product good to go or in need of rework.
Examples include color and size sorters, leak test systems, metal detectors and special analytic systems.
Typically, an inspection system uses visual, color spectrum, heat, photo imaging, weight, X-ray, metal detection, harmonic or other analytical instruments to evaluate specific criteria important to the application. If a product passes inspection, it’s allowed to proceed forward in the production line. If the product fails inspection, it is physically diverted or digitally denoted for rework. Simple inspection systems can use low-complexity automation components to make their evaluation, such as a proximity switch gauging the presence or absence of a bottle cap. The switch activates a diverter gate to remove any bottles missing caps from a conveyor path. Advanced systems such as photo spectrometers may look for color abnormalities imperceivable to the human eye in critical products traveling at very high rates of speed, using advanced controls with integrated reporting and statistical tools required to meet regulatory compliance.
Automated Data Systems
Systems designed to utilize data and make appropriate decisions based on this data are found at the cutting edge of industrial manufacturing. These automated data systems blur the lines between mechanized work and knowledge work, taking input from physical systems on the plant floor and actioning that data via connected digital systems throughout an organization. We also refer to these as “decisions systems,” as their primary function is to draw conclusions from input data and take direct action in support of specific objectives.
Examples include:
- Continuous process improvement systems that monitor real-time process data from manufacturing lines in operation and automatically adjust specific variables on the fly in order to improve line efficiency, product quality and other valuable metrics.
- Interface gateways or software applications increasingly being deployed to interface with external resources based on conditions observed in local systems. In a manufacturing plant, an advanced refrigeration system may track its refrigerant volume, establishing and monitoring against a “normal” level value. In the event of a “below normal” level such as in the case of a major leak, the system can automatically trigger internal rapid shut-down as well as external emergency responses (such as notifying local hazmat and fire systems and receiving and implementing immediate instructions from those systems).
- Self-managing ecosystems, such as neural networks and IIoT (industrial internet of things), are the industrial variant of what consumers would recognize as “smart devices.” The industrial internet of things looks to provide maximal real-time process optimization, condition response and decision-making by integrating every possible device on a plant floor with one another. The belief is that “connected components” can generate much more data than what we have with current control systems, and that using this data in real-time can lead to more performative benefits extending from the plant-floor all the way up to enterprise management. Self-contained ecosystems such as these have yet to be fully validated and deployed at scale, but at least for now, they serve to identify one potential future path that industrial automation may take.
Automation Machinery for You
Are you ready to explore industrial automation industry machinery to optimize your assemblies? Check out our downloads for additional automation resources or contact us directly.
