However, that tends not to be the case when the same production is outsourced, because each supplier then needs a separate team of operators. Where multiple suppliers produce the same components, volumes are lower and the likelihood of finding operators with the specialized skills and training necessary to achieve optimum quality and efficiency is reduced.

In addition, the coincident trend toward smaller batch sizes across the entire spectrum of manufacturing makes the situation even more critical. Smaller batches imply more flexible manufacturing systems, which, by their very nature, require more highly skilled operators to cope with their increased complexity and technological sophistication.

Finally, manufacturers in developed countries are faced with an aging workforce, particularly among the more highly skilled, and a diminishing supply of new talent as younger workers gravitate to other career paths. Demographic trends in North America and Europe guarantee this situation will be with us well into the foreseeable future.

Automation isn’t enough The best solution to this challenge is to find ways to make the existing pool of operators more efficient, while replicating their skills and experience among their less skilled and experienced colleagues. Fortunately, this can be accomplished with today’s technology.

We can enhance the effectiveness of the best operators by providing them with artificial "senses" in the form of sophisticated measurement and detection systems that add unprecedented precision to their existing skills. Additionally, we can capture their experience and knowledge in applying those new "senses" in software systems to support others performing the same tasks.
Here’s how it might work.

Building the "sensory support system"
Equipping a manufacturing operation with the gages and signal processors necessary to implement the enhanced set of aforementioned "senses" is well within today’s state-of-the-art capabilities. It doesn’t require a breakthrough in artificial intelligence technology or the creation of a science fiction cyborg operator. It can be done using existing gaging and process control systems.

The gages exist, the signal processors or amplifiers exist and the knowledge needed to integrate them into production machinery exists. Creating a functioning "sensory support system" requires only a few basic decisions.

The most basic of these decisions is whether the application requires a custom system or a standard, off-the-shelf solution. While the ultimate answer may not be as simple as the question implies, the procedure is the same for both options.

Start with the part geometry, the measurements required and the tolerances involved. These three parameters will define the type of gages required, which will, in turn, help you determine whether an off-the-shelf solution is available from your preferred supplier.

Today’s most advanced gaging systems tend to be highly modular, which can blur the distinction between standard and custom solutions. However, that actually works in your favor by widening the range of economically viable choices.

Another advantage of modularity is that the same gauging components can be frequently used in a variety of different applications. For example, current generation "single-finger" gages work equally well on grinders, lathes and even machining centers for preprocess, in-process and postprocess measurement.

This application flexibility helps reduce cost by minimizing inventory requirements and simplifying training. Finally, modularity helps conserve capital, because modular systems are more easily reconfigured to accommodate future part changes or even completely different parts.

In any case, the gaging solutions used to build the "sensory support system" need to provide outputs that can be electronically captured for subsequent analyses and re-use. This is the key difference between a pure quality assurance system and an operator support system. Remember, the intent is to put in place a system to enhance the capabilities of your best operators, while capturing their skills and experience to support their less capable colleagues.

Virtual eyes
The first step in the process is to equip the operator with virtual eyes, fingers and ears to enhance the quality of the data being fed to the brain. Using grinding as an example, the eye’s function is best duplicated by an in-process gage that delivers real-time information on what’s happening to the workpiece while it’s being ground.

Based on what the operator sees through the gage, the machine cycle can be adjusted to achieve optimum efficiency while producing the required level of quality. Measurement values, rate-of-change data and process stability can be easily observed and used to make the decisions necessary to produce quality parts.

For example, knowing where the part surface is prior to cycle-start minimizes the amount of time spent grinding air at the beginning of the cycle. That initial size measurement can also tell a skilled operator whether the part is worth grinding at all. A part with too little stock at the beginning of the cycle will probably be scrap at the end, while a part with too much stock will take too long to bring to size and accelerate wheel wear. The same is true for a workpiece with excessive run-out, a condition the latest generation of in-process gages can also detect.

Watching the rate of change during the grinding cycle also provides vital information. Because most automatic cycles are based on average stock and wheel conditions, an astute operator can tweak the process to take full advantage of a sharp wheel or a soft workpiece. Conversely, scrap parts and damaged wheels can be prevented based on rate of change information. And, of course, the rate of change in part size is a primary indicator of when to perform functions such as infeed rate reduction, finish dressing, dwell and end of cycle.

Finally, by keeping an eye on measurement stability, the operator can see how the part reacts to the grinding process. An unstable process can be caused by worn tools, misloaded parts, incorrect setups and a myriad of other operational circumstances. An erratic operation requires the operator to make the immediate decision whether to finish the cycle or abort it and call for maintenance.

Here’s a real-world example of how watching an operation with an in-process gage can improve the productivity of a Blanchard-type surface grinding operation. By using a single-finger gage to measure stock height of incoming parts and an amplifier to provide the necessary control logic, a customer is able to grind both sides of a part in a fully-automatic cycle with no changeover required.

The gage measures the incoming parts and the amplifier calculates how much stock must be removed from each side to bring the part to specification. The amount of stock removed from each side may be the same or different, depending on the part. The gage then monitors the grinding operation on the first side until it reaches the calculated zero point.

The parts are then flipped over and the machine control automatically grinds the second side to the new zero point, while the gage monitors the operation. At the end of the cycle, the control returns the machine to the original zero to accept a new part.

In addition to being a productivity-enhancing application in its own right, this also is a good example of the value of today’s modular gaging systems.

Virtual fingers
Metrologists know that vibration is the enemy of grinding precision, but it’s equally the enemy of grinding economy. Today’s advanced technology wheels can be ferociously expensive and at the spindle speeds they’re designed for, it doesn’t take very long for a little vibration to turn into a very costly problem.

Even when wheels are prebalanced, variations in mounting clearances can result in vibrations that lead to chatter in the workpiece, excessive wear on the wheel and dressing tools, and premature failure of the spindle. Automatic wheel balancing systems help the operator "feel" the vibration—even at extremely low levels—and make the required corrections.

The latest technology in this area features high-speed capabilities up to 12,000 rpm, and advanced design features to eliminate residual torque in the wheel head by placing both of the mass gravity centers in the same plane as the wheel center of gravity. The result is improved surface quality, more accurate part cylindricity and the elimination of chatter. These systems also use electronics, which are compatible with acoustic gap, crash and dressing controls that provide the "ears" of the "sensory support system."

Virtual ears
One of the most important lessons practice teaches a machine operator is exactly how a good operation sounds. That sense, too, can be augmented and improved with technology in the form of acoustic sensors, or virtual "ears."

Acoustic sensors are most commonly used today as gap eliminators to detect contact between the wheel and a workpiece or dresser, and as crash detectors to prevent catastrophic mistakes. However, they have a wide range of other capabilities. By combining multisensor inputs with sophisticated analytical algorithms, today’s systems provide new opportunities to integrate the operator with the process.

For example, it’s now possible to "hear" half a micron of material being removed from a wheel during a dressing operation, or to detect a chip or other discontinuity in a wheel’s surface that cannot be seen with the naked eye. What the skilled operator would have diagnosed through intuition can now be determined empirically and remedied before producing a scrap part. It’s also possible to "hear" the result of an uneven dress that leaves the wheel out of round. Soon we’ll be able to deliver information about spindle bearing condition and a host of other machine parameters simply by "listening" to the sounds they make.

Creating a virtual journeyman
Much of the hardware needed to implement a sensory support system already exists in most plants doing precision machining today. Everything described in this article is readily available to those plants not currently using these technologies. The missing piece is the system needed to capture the interaction between the operator and the control system, and pass it on to others in the plant. The solution to that requirement is almost certainly a database-driven networked software implementation that doesn’t exist yet. That’s the bad news.

The good news is that the sensory support system described in this article works very well as a stand-alone solution for an individual machine tool or a manufacturing system. In-process gauging, wheel balancers and acoustic sensors are all proven technologies. So are postprocess gages and closed-loop control systems. Integrating these technologies into your manufacturing processes today lays the foundation for tomorrow’s improvements and gives you a competitive advantage as the pool of skilled operators continues to shrink.