Manufacturing execution systems allow assemblers to “see” exactly what’s happening on the plant floor.



In the fifth act of William Shakespeare’s Macbeth, the title character complains of life being a tale “full of sound and fury, signifying nothing.”

Well, that may be true for treacherous Scottish generals and their ambitious wives. But it is most definitely not the case for manufacturers. The hubbub of activity taking place in a modern assembly facility is filled with implications. Time, materials, scrap, workflow, even the sequences in which products are assembled, all have a very tangible impact the bottom line. In an increasingly competitive global economy, getting a handle on all that sound and fury can mean the difference between making a profit and ending up like, well, like poor old Macbeth.

Manufacturing execution systems allow engineers to extract real-time data directly from the assembly line. Photo courtesy Ford Motor Co.

Taking Care of Business

It is with this in mind that many assemblers have begun implementing some type of process-monitoring system to track everything from individual parts to uptime and performance metrics for individual assembly machines.

Similar to the enterprise resource planning (ERP) systems-which are used to rationalize and plan overall business operations by combining everything from accounting to marketing and logistics in a single database-today’s manufacturing execution systems (MES) allow engineers to cut through the clutter and “see” what’s really going on. By pulling data directly from the plant floor and processing it with the system’s analytic software, engineers can get a real-time snapshot of what is happening at either the workstation level or plant-wide. Engineers can also study long-term trends or past performance with an eye toward fine tuning their processes or uncovering the true source of otherwise hard-to-pin-down quality problems.

Finally, by linking specific products with process data, MES makes it possible to better control warranty issues. No longer does an automobile company have to recall tens of thousands of cars because of, say, “clutch” problems. Now, ideally, the company and its supplier can pinpoint the specific batch of clutches exhibiting the defect. Better still, they can even pinpoint the specific machine, workstation or component responsible for the problem by studying the clutches’ processing records.

The key here is “interlocking traceability,” says Tom Comstock, senior vice president of product management at Apriso Corp. (Long Beach, CA). By combining quality and processing data with a specific part, manufacturers are able to practice a “spill management” approach to warranty issues, identifying and containing the full nature of the problem as quickly and efficiently as possible.

Comstock-whose company has developed a series of MES products, including FlexNet Global Manufacturing Suite and FlexNet Production-says MES can also help a supplier defend itself in the event it is caught up in a warranty problem. Specifically, by documenting “how” it manufactured its product and proving it did so correctly, it can prove the source of a product’s problems lie elsewhere.

Rich Breuning, product manager at automation company GE Fanuc (Charlottesville, VA), agrees. Genealogy and traceability, or a “birth certificate” for each part produced, makes all the difference in the world when it comes to handling these two sides of a recall situation, he says.

At this inspection station, the operator is inputting product data to ensure each truck has received the correct options before it is shipped to the customer. Photo courtesy ASI DataMyte

Getting in Tune

As is the case with ERP systems, MES relies on software and a network of computers connected with one another, the production line and possibly a wider ERP system via Ethernet. Data is then extracted from the shop floor using handheld or fixed code or tag readers; a digital tool’s own control system; visual inspection software and operator interfaces; or a retrofitted sensing system installed in an older analog machine.

A digitally controlled press or screwdriver, for example, can be tied seamlessly into a plant’s MES software, thanks to the compatibility that naturally follows from the widespread use of OPC communications standards. This allows a two-way flow of information, so that the individual tool is providing production data at the same time it is being directed by the MES.

In the case of a screwdriver, the MES can alert the controller as to the specific product being worked on, after which the screwdriver’s sensors and controller record the torque settings that were achieved during actual assembly. This value is then permanently recorded for the component as part of its birth certificate. In addition, the time and duration of installation, among other metrics, are also logged in the interest of monitoring total throughput for the shift, day, week or any other time period-all data that can be used by engineers to enable better line balancing and eliminate production bottlenecks.

Of course, a screwdriver alone, while it is great for measuring things like torque and depth, can’t identify the assembly it’s working on. This is the job of some kind of tracking system, employing a bar code, 2D data matrix code or RFID tag. These systems keep track of where a particular assembly is in the system, and the way in which is has been processed. They can also specify what tasks need to be performed to a particular assembly if there is more than one type of product being processed on a single line.

This latter information is generally displayed on a machine or workstation operator interface, so that the operator knows exactly what to do with the assembly in front of him. “Go, no-go” functions further error-proof the process by disabling a piece of equipment if it isn’t being used in accordance with the displayed work instructions, or if a prior step has been omitted. Basically, the system requires that every step be correctly completed before the assembly is released to another workstation for further processing.

Another option is to tie in a visual inspection system, like the Inspect Software system developed by quality assurance specialist ASI DataMyte Inc. (Plymouth, MN). This system, which replaces traditional paper route sheets, feeds inspection and quality information directly to a central database, where managers can study everything from the status of an individual product to plant-wide processing and rework times.

In one instance, a truck manufacturer is using the system to track and confirm the installation of customer-specified options on otherwise identical vehicles. In operation, the inspector scans each vehicle, and the system poses a series of questions to the inspector to determine if it has been correctly assembled. The software then compares the answers to these questions to the options specified for that vehicle’s particular VIN. Automated test equipment can also be integrated into the system to ensure correct functionality. In either case, in the event of a disagreement, the vehicle won’t be released for delivery.

No matter what the exact means of accessing data, MES keeps the operator and anybody else who might be logged into the system abreast of what is happening through the machine’s operator interface or linked computer. Processing parameters, model numbers, part schematics, production levels, and even the customer’s name and address can be accessed this way, providing complete transparency.

In sum, by pulling together an interlocking network of controllers, readers, software files and processing equipment sensors, MES offers a powerful means of truly understanding exactly what is happening in a particular facility at any given moment or over any particular time period-from anywhere in the world.

This screen shot shows what the operator sees midway through processing a hub assembly. Note the status indicators in the upper right-hand corner and the checklist to the left of the schematic drawing. Photo courtesy GE Fanuc

Real-world Examples

To illustrate the possibilities of MES, Breuning, whose company offers what it calls its Proficy line of products, cites the example of the International Automotive Component Group factory in Göteborg, Sweden. Recently acquired from Lear Corp. (Southfield, MI), the plant manufactures cockpits on a just-in-time basis for a nearby Volvo plant. As part of the deal, the company pays a $10,000 penalty every time it screws up-either by being late or by providing the wrong cockpit or one that is defective.

To ensure this doesn’t happen, the company installed Proficy Assembly software and the requisite hardware to tie together the entire assembly process to ensure that each cockpit is built “correctly the first time, every time.”

“Let’s say an operator is installing the radio. They’ve got 30 to 40 seconds. There are 20 to 30 steps, and the workspace is 10 linear feet,” Breuning says. “The operator scans the cockpit. A display shows the operations to perform. He then scans the radio and torques its down with four screws.”

According to Breuning, as the operator is driving the screws, the digital screwdriver and MES generate a pass-fail reading and an exact torque value, which is included in the cockpit’s birth certificate. If any of the screws fail to meet the correct specification, the cockpit is flagged for removal or rework. The MES software simply won’t allow any additional work to be performed on the cockpit until the problem has been rectified or overridden by a qualified employee.

Equally important, the MES software flags those instances in which torque values or other parameters are getting close to the edge of acceptability. That way, the maintenance department can intervene before there is an actual problem, as opposed to reacting after things have already started going wrong.

“Once it’s broken down or you have quality problems, it’s already too late,” agrees Apriso’s Comstock, explaining the importance of these kinds of “watch out” metrics, as opposed to simple pass-fail inspections.

According to Breuning, this kind of linkage is also possible with old, legacy equipment through the use of specialized interface software. However, performing the necessary retrofits can be costly and generally only makes most sense when dealing with some kind of mission-critical application. In fact, most companies implement MES as part of a wider upgrade, which means new assembly equipment is also going on-line, making integration fairly straightforward.

Sharing Best Practices

Another benefit of MES is that by documenting the specifics of a particular process, it allows a process to be transported or more effectively monitored in another facility-whether that facility is on the other side of town or the other side of the world.

“Moving best practices from one place to another, is the single most important trend in the area,” Comstock says. “If I’m sourcing parts in China, I need to know quality is good. I need to know my product is being manufactured the same worldwide.”

Better still, once a set of processes is in place, it’s possible to compare how things are going in multiple facilities to determine new best practices once the line is up and running. No longer do engineers find themselves comparing apples and oranges. Instead, through the use of an enterprise-wide system, they have access to a synchronized and uniform record of their disparate processes, a “single version of the truth,” as it were.

“MES makes it possible to compare OEE (overall equipment effectiveness) across different lines. It doesn’t matter where the plants might be…. You can figure out why one piece of equipment is doing better in, say, Singapore than Chennai,” Comstock says.

Comstock adds that these kinds of analyses aren’t limited to highly automated assembly lines. In fact, he says by tracking material flow through the use of bar codes, Data Matrix codes and fixed or handheld readers, engineers can get a better idea of the efficiencies being experienced in mixed or manual settings as well.

This can be especially important when working with plants in low-cost regions, like Asia, Eastern Europe or South America, where something as simple as the way the plant is performing its kitting functions may be bogging down the line. By accessing uniform, up-to-date production data, engineers can identify and help diagnose the problem, even when it is multiple time zones away. As an example, Comstock cites a high-volume eyeglass manufacturer that was able to cut turnaround times from 5 to 2 days by better synchronizing material flow between the “lab” that ground the lenses and the part of the plant that installed the lenses in the correct frames.

Ultimately, Comstock says, whenever an assembler transports its operations to another location, it inevitably introduces a host of new variables that can have a profound effect on productivity. “In essence, they are taking [the process] out of a controlled environment and hoping the manufacturer assembles it correctly,” he says.

By keeping tabs via MES, however, that same assembler can gauge these same processes as if they were still being executed in the home facility.