IO-Link Boosts Plant Productivity

In my previous blog, Using Data to Drive Plant Productivity, I categorized reasons for downtime in the plant and also discussed how data from devices and sensors could be useful in boosting productivity on the plant floor. In this blog, I will focus on where this data is and how to access it. I also touched on the topic of standardizing interfaces to help boost productivity – I will discuss this topic in my future blog.

Sensor technology has made significant progress in last two decades. The traditional transistor technology that my generation learned about is long gone. Almost every sensor now has at least one microchip and possibly even MEMs chips that allow the sensor to know an abundance of data about itself and the environment it which it resides. When we use these ultra-talented sensors only for simple signal communication, to understand presence/absence of objects, or to get measurements in traditional analog values (0-20mA, 0-10V, +5/-5V and so on), we are doing disservice to these sensors as well as keeping our machines from progressing and competing at higher levels. It is almost like choking the throat of the sensor and not letting it speak up.

To elaborate on my point, let’s take following two examples: First, a pressure sensor that is communicating 4-20mA signal to indicate pressure value. Now, that sensor can not only read pressure value but, more than likely, it can also register the ambient temperatures and vibrations. Although, the sensor is capable of understanding these other parameters, there is no way for it to communicate that information to the higher level controller. Due to this lack of ambient information, we may not be able to prevent some eminent failures. This is because of the choice of communication technology we selected – i.e. analog signal communication.

For the second example, let us take a simple photoeye sensor that only communicates presence/absence through discrete input and 0/1 signal. This photoeye also understands its environment and other more critical information that is directly related to its functionality, such as information about its photoelectric lens. The sensor is capable of measuring the intensity of re-emitted light, because based on that light intensity it is determining presence or absence of objects. If the lens becomes cloudy or the alignment of the reflector changes, it directly impacts the remitted light intensity and leads to sensor failure. Due to the choice of digital communication, there is no way for the sensor to inform the higher level control of this situation and the operator only learns of it when the failure happens.

If  a data communication technology, such as IO-Link, was used in these scenarios instead of signal communication, we could unleash these sensors to provide useful information about themselves as well as about their environment. If we collect this data or set alerts in the sensor for the upper/lower limits on this type of information, the maintenance teams would know in advance about the possible failures and prevent these failures to avoid eminent downtime.

Collecting this data at appropriate frequencies could help build a more relevant database and demonstrate patterns for the next generation of machine learning and predictive maintenance initiatives. This would be data driven continuous improvement to prevent failures and boost productivity.

The information collected from sensors and devices – so called smart devices – not only helps end users of automation to boost their plant’s productivity, but also helps machine builders to better understand their own machine usage and behaviors. Increased knowledge improves the designs for the next generation of machines.

If we utilized these smart sensors and devices at our change points in the machine, it would help fully or partially automate the product change-overs. With IO-Link as a technology, these sensors can be reconfigured and re-purposed for different applications without needing different sensors or manual tunings.

IO-Link technology has a built in feature called “automatic parameterization” that helps reduce plant down-time when sensors need replaced. This feature stores IO-Link devices’ configuration on the master port as well as all the configuration is also persistent in the sensor. Replacement is as simple as connecting the new sensor of the same type, and the IO-Link master downloads all the parameters and  replacement is complete.

Let’s recap:

  1. IO-Link unleashes a sensor’s potential to provide information about its condition as well as the ambient conditions, enabling condition monitoring, predictive maintenance and machine learning.
  2. IO-Link offers remote configuration of devices, enabling quick and automated change overs on the production line for different batches, reducing change over times and boosting plant productivity.
  3. IO-Link’s automatic parameterization feature simplifies device replacement, reducing unplanned down-time.

Hope this helps boost productivity of your plant!

Using Data to Drive Plant Productivity

What is keeping us from boosting productivity in our plants to the next level? During a recent presentation on Industry 4.0 and IIoT, I was asked this question.

The single biggest thing, in my opinion, that is keeping us from boosting productivity to the next level is a lack of DATA. Specifically, data about the systems and the processes.


Since the beginning of time, we have been hungry for efficiency. While early man invented more efficient methods to hunt and survive, today we are looking for ways to produce more efficiently in our plants with minimum or zero waste. After exhausting all the avenues for lean operations on plant procedures and our day-to-day activities, we are now looking at how we can recover from unanticipated downtime quickly. I am sure in future we will be seeking information on how can we prevent the downtime altogether.

There are plentiful of reasons for downtime. Just a few examples:

  1. Unavailability of labor – something we might be experiencing these days, when the COVID-19 pandemic has reduced some labor forces
  2. Unavailability of raw materials
  3. Unavailability of replacement components
  4. Unavailability of assets
  5. Failures in machines/components

In this list, the first two reasons, are beyond the scope of this blog’s intentions and frankly somewhat out of controls from the production standpoint.

The next two reasons, however, are process related and the last one is purely based on the choices we made. These three reasons, to a certain extent, can be reduced or eliminated.

If the downtime is process related, we can learn from them and improve our processes with so called continuous improvement initiatives. We can only do these continuous improvements based on observable factors (a.k.a. data) and we cannot improve our processes based on speculations. Well, I shouldn’t say “cannot”, but it will be more like a fluke or luck. It is apt to say “ what can’t be measured, can’t be improved!”

A good example for elaborating my point is change-over in the plant to produce a different product. Unless there is a good process in place for ensuring all the change-over points are properly addressed and all the change parts are correctly installed and replaced, the changeover time could and will likely lead to tremendous amounts of lost productivity. Secondly, if these processes are done manually and not automated, that is also a loss of productivity or, as I like to say, an area for continuous improvement to boost productivity based on observable facts. Sometimes, we take these manual change-overs as a fact of life and incorporate that time required as a part of “planned” downtime.  Of course, if you do change-overs once a year – it may be cost effective to keep the process manual even in today’s situation. But, if your plant has multiple short batch productions per day or per week, then automating the changeovers could significant boost productivity. The cost benefit analysis should help prove if it is continuous improvement or not.

Assets are an important part of the equation for smooth operations. An example would be molds in the stamping plant or cutting-deburring tools in metal working plants. If plants have no visibility or traceability of these important assets for location, shape or form, it could lead to considerable downtime. The calibration data of these tools or number of parts produced with the tool are also important pieces of data that needs to be maintained for efficient operations. Again, this is data about the system and the integration of these traceability initiatives in the existing infrastructure.

Failures in machines or components could cause severe downtime and are often considered as unavoidable. We tackle these failures in a two-step approach. First, we hunt for the problem when it is not obvious, and two, we find the replacement part in the store room to change it out quickly. And, as process improvement, we schedule preventative maintenance to inspect, lubricate and replace parts in our regular planned downtime.

The preventative maintenance is typically scheduled based on theoretical rate of failure. This is a good measure, especially for mechanical components, but, predictive or condition-based maintenance usually yields higher returns on productivity and helps keep plants running smooth. Again, predictive maintenance relies on data about the condition of the system or components. So, where is this data and how do we get to it?

Standardization of interfaces is another important component for boosting productivity. In my next blog, I will share how IO-Link as a technology can help address all of these challenges and boost productivity to the next level.

Changing the Paradigm from Safety vs. Productivity to Safety & Productivity

In a previous blog, we discussed how “Safety Over IO-Link Helps Enable Human-Robot Collaboration”. It was a fairly narrow discussion of collaborative robot modes and how sensors and networks can make it easier to implement these modes and applications. This new blog takes a broader look at the critical role safety plays in the intersection between the machine and the user.

In the past, the machine guarding philosophy was to completely separate the human from the machine or robot.  Unfortunately, this resulted in the paradigm of “safety vs. productivity” — you either had safety or productivity, but you couldn’t have both. This paradigm is now shifting to “safety & productivity”, driven by a combination of updated standards and new technologies which allow closer human-machine interaction and new modes of collaborative operation.

Tom_Safety1.pngThe typical machine/robot guarding scheme of the past used fences or hard guards to separate the human from the machine.  Doors were controlled with safety interlock switches, which required the machine to stop on access, such as to load/unload parts or to perform maintenance or service, and this reduced productivity.  It was also not 100% effective because workers inside a machine area or work cell might not be detected if another worker restarted the stopped machine.  Other drawbacks included the cost of space, guarding, installation, and difficultly changing the work cell layout once hard guarding had been installed.

We’ve now come to an era when our technology and standards allow improved human access to the machine and robot cell.  We’re starting to think about the human working near or even with the machine/robot. The robot and machinery standards have undergone several changes in recent years and now allow new modes of operation.  These have combined with new safety technologies to create a wave of robot and automation suppliers offering new robots, controllers, safety and other accessories.

Machine and robot safety standards have undergone rapid change in recent years. Standard IEC 61508, and the related machinery standards EN/ISO 13849-1 and EN/IEC 62061, take a functional approach to safety and define new safety performance levels. This means they focus more on the functions needed to reduce each risk and the level of performance required for each function, and less on selection of safety components. These standards helped define, and made it simpler and more beneficial, to apply safety PLCs and advanced safety components. There have also been developments in standards related to safe motion (61800-5-2) which now allow more flexible modes of motion under closely controlled conditions. And the robot standards (10218, ANSI RIA 15.06, TS15066) have made major advances to allow safety-rated soft axes, space limiting and collaborative modes of operation.

On the technology side, innovations in sensors, controllers and drives have changed the way humans interact with machines and enabled much closer, more coordinated and safer operation. Advanced sensors, such as safety laser scanners and 3D safety cameras, allow creation of work cells with zones, which makes it possible for an operator to be allowed in one zone while the robot performs tasks in a different zone nearby. Controllers now integrate PLC, safety, motion control and other functions, allowing fast and precise control of the process. And drives/motion systems now operate in various modes which can limit speed, torque, direction, etc. in certain modes or if someone is detected nearby.

Sensors and Networks
The monitoring of these robots, machines and “spaces” requires many standard and safety sensors, both inside and outside the machine or robot. But having a lot of sensors does not necessarily allow the shift from “productivity vs. safety” to “productivity & safety” — this requires a closely coordinated and integrated system, including the ability to monitor and link the “restricted space” and “safeguarded space.” This is where field busses and device-level networks can enable tight integration of devices with the control system. IO-Link masters and Safety Over IO-Link hubs allow the connection of a large number of devices to higher level field busses (ProfiNet/ProfiSafe) with effortless device connection using off-the-shelf, non-shielded cables and connectors.

Balluff offers a wide range of solutions for robot and machine monitoring, including a broad safety device portfolio which includes safety light curtains, safety switches, inductive safety sensors, an emergency stop device and a safety hub. Our sensors and networks support the shift to include safety without sacrificing productivity.