LinkedIn

WEBINAR: In-House Vs. Commercial MES Systems

Traceability: Using Data To Gain Control and Improve Line Performance

The clock is ticking. Every second that passes costs your company money. Your line is shut down and you feel the back of your neck starting to sweat. Imagine you’re responsible for manufacturing RF modules at a cell phone chip supplier, and you’re trying to find out why your production yield just dropped off a cliff.

The modules you make are physically small, about 1mm square, and consist of a few semiconductor chips, filters and passive components. Everything has been going along smoothly with acceptable failure rates, but now you find yourself in the middle of an urgent investigation that had better start producing some answers fast.

Your line techs suggest starting with the semiconductors. Since they are the most complex parts in the assembly, they have a higher likelihood of failure. After an engineering team spends days running reliability tests, all the chips check out fine. You move the testing along to the other parts. As each day passes with no answers, the pressure begins to build. Finally, after two weeks composed of many stressful hours, your engineers find the problem – an inductor!

You discover that the supplier had changed their procedure on making the component without telling you. The new process shifted the center frequency just enough to knock your module yield to zero.

Having resolved the inductor issue, you begin to calculate the losses. You realize in shock that the time for the engineering analysis alone is going to run at least a million dollars. You nervously start factoring the lost revenues and expedite fees. This fiasco ends up being a major hit that will not bode well for you or your company.

You may live in dread of a situation like the one just described, or you may have already been through something similar. Whatever the case, you are about to learn some ideas that will help you prevent a similar thing from happening to you. The powerful tool of “traceability” boasts among its superpowers the ability to significantly shorten the time it takes to spot potential problems on your line and prevent or correct them before they become costly blunders.

What is Traceability?

Data access drives the modern production environment. Manufacturing and the supply chain are full of interwoven complexities that require an automated system to keep up with everything. Traceability gives you the power to track and share data on the production line with ease. Traceability has two functions:
Newer technologies such as IoT and cloud computing stacked with advanced on-premise options enable an organization's MES to collect more detailed information and distribute it effortlessly. Traceability saves time and money by making any problems on the production line easier to identify and fix.

What Traceability Does

You’ve learned so far that companies need traceability to record data and then share it so they can eliminate lengthy and expensive trial and error troubleshooting. You may be wondering at this point precisely what type of data is useful to trace. Typically, the more information you can trace about both the process and the materials used the better.

Recording the Process and Materials

Here are some significant areas of the process that you ideally want your MES to keep track of for each part or component in real time on each manufacturing job.
Modern MES also possess the powerful capability of tracking detailed information about the materials you are using. Having the means to record the following information to each part is also essential to keep your line running smoothly and cost-effectively.
Now, you should start to see the potential of traceability. In the example of the “six-figure inductor” examined previously, traceability would have substantially cut the time needed to solve the problem. Depending on the level of traceability used, engineers could have zeroed in on the out-of-spec component and solved the problem in hours instead of weeks.

Commonality Analysis

The slight shift in the center frequency of an inductor created a “showstopper.” Big, sudden, radical shifts can occur from seemingly insignificant causes on the other end. The first thought was, “the drop in yield is significant, so most likely the problem lies in the most complex parts, the semiconductor chips.” That’s a logical assumption when you are working with limited or no information.

Unfortunately, in this case, the guess turned out to be dead wrong; the problem happened to be in a component less prone to problems. This wrong guess caused the team to waste valuable time testing chips that were functioning correctly.

When more information is available, engineers often refer to a better approach to troubleshooting as “commonality analysis.” The method involves answering a series of questions. Naturally, the more information available, the better a commonality analysis will work.

First, you identify good and bad populations of components. The idea is to answer the question, “What is uniquely common to the bad parts?” For example:

Whether you've heard it called by that name or not, commonality analysis is the heart and soul of problem-solving on a manufacturing line. Traceability, and more specifically the level of traceability, determines how effectively you can prevent problems from occurring and how quickly you can recover from the ones that do happen.

Levels of Traceability

Traceability has been around to some extent since modern manufacturing plants first started cranking up in the last century. What’s rapidly changing is the reach that advanced technology gives to manufacturing data. As you seek to control what is happening on the line, the more data you can collect and share gives you a better picture of what is happening at any given moment.

When a component is out of spec, it will tend to stick out like a sore thumb. If a particular operator or machine is performing poorly, a red flag will alert you before the damage is too great. Getting information to the right person or team is no longer a struggle; it can happen automatically.

Not every production situation needs the most advanced level of traceability. Depending on the circumstances, you may need fewer data points to make your component, part or product efficiently. Deciding what you need will usually come down to analyzing the available options and figuring out where your needs fit within the spectrum of what is possible. It can feel confusing, but traceability boils down to just two types:

Serialized Traceability

With the added advantage provided by technology, the era of serialized traceability is beginning to enter the mainstream of the manufacturing environment. Limited tracking is rapidly giving way to more advanced methods that fully automate data collection and accessibility.

Determining Your Traceability Sweet Spot

You know your operation and the requirements of your industry, so you probably already have a good feel for what you need. If you are unsure, here are a couple of factors to consider that may help you decide what level of traceability is right for you.

Using Serial Numbers - Consider the practicality of introducing serial numbers on the parts. Ask yourself if you have space for a printed barcode label? Can you print or laser scribe a barcode somewhere on the piece? Is there a chip on your assembly you could use to store the number electronically?

Difficulty of Implementation - Setting up any system requires some initial work. Serial-level implementation is no exception, but like any other great system, once it is set up, it starts operating at virtually no cost.

Planning out how operators will interact with the new traceability system requirements is often the biggest hurdle. For example, it may take a little time for a worker to get in the habit of scanning a barcode, and it may even initially add a few seconds to the operation.

However, when you think the process through, you may be able to use the barcode scan to take care of several tasks at once. Each scan can perform operations such as pressing “enter” on the keyboard or displaying information on a chart.

Over a surprisingly short amount of time, the process begins to save steps in the process. In turn, a thoroughly integrated traceability system provides invaluable real-time information about every aspect of your methods and materials. You can use this data to monitor continually and improve your production, and you can retrieve critical data instantly when a problem occurs.

Serialized Traceability: The Manufacturing Game Changer

The power of serial level traceability is simultaneously subtle and extreme. As data accumulates for every process in the production trail, the system starts to reveal previously hidden inefficiencies. Since authorized access to information is available through any internet connected device, the attention of the entire organization can focus on examining and improving every aspect of manufacturing. A web is created that connects all of the intricate moving parts – from weak supply chain links to inefficiencies in machine and operator – and enables cycles of learning on your line that were previously impossible.

With serialized traceability, getting RMAs are now a rare, easily solvable occurrence that usually leads to an immediate improvement in some area. Scanning the serial number calls up a complete history for the part and every component in it. Going deeper, you retrieve complete information for the specific components themselves.

If you find a problem with a particular group, you perform a "where used" analysis that shows you all products that contain the faulty component. You put a hold on any products containing the parts that are currently on the line and take corrective action. Likewise, integration with your ERP system allows you to do a " where shipped" analysis and address any parts that are already out in the field.

Serialized traceability pushes constant improvement. It allows you to avoid situations that start as minor problems and morph into financial disasters. When a problem like a product failure does occur, you can quickly track down the precise source and correct the problem. Keeping the information flowing between departments stops the tendency to cover up issues and creates more transparency and accountability throughout your organization.

The Smart Factory Era Is Here

Industry 4.0 refers to automation and data exchange that continues to redefine manufacturing processes. Cloud-stacking applications, IoT, cyber-physical systems, cognitive computing and other evolving technologies give operational managers a lot to digest.

Serialized traceability is the essential first step to get your organization moving in the direction of the future.

Contact Intraratio today, and let us help you decide what level of implementation will help get your organization leveraging current technology to your best advantage.