Tag Archives: Variability

 

Remember when artificial intelligence (AI) was a glimmer on the horizon? And then ChatGPT stormed onto the scene and people were convinced every job out there was soon going to be replaced by a bot? Now it turns out, not so much.

As awesome (and we don’t use that word lightly) as AI is, it’s only as good as the data it has to work with. At USC Consulting Group, we’re finding this is especially true when we’re using AI for predictive analytics. AI doesn’t like variation, and there can be a lot of that in manufacturing processes.

Here’s a look into this issue and how to handle it.

A short primer into AI and predictive analytics

AI is a broad term describing computer systems that perform intelligent tasks, like reasoning, learning, problem solving, and more. Not so obvious is predictive analytics, which is the ability to forecast future outcomes using AI based on data. You’re already familiar with it, to a certain degree. If you’ve ever had a recommendation from Netflix based on what you’ve watched in the past, that’s it. In a nutshell.

Netflix’s use of predictive analytics created a seismic shift in consumer expectations. This technology also has the potential to transform operating procedures and processes for many industries.

It’s extremely powerful when dealing with processes in which multiple predictors are influencing outcomes. It has the ability to tell us which path to take in order to achieve a desired outcome, even when process patterns and trends are changing.

It means greater precision and accuracy, speed and increased efficiency, the holy grails for any manufacturer.

But there is a fly in this cyber ointment.

Variation.

AI doesn’t like it and – low and behold – that means humans are necessary in this process in order for predictive analytics to achieve its potential.

What is variation?

When we’re talking about manufacturing processes, what exactly does variation mean?

In manufacturing, variation is the difference between an actual measure of a product characteristic and its target value. Excessive variation often leads to product discard or rework.

When you’re dealing with high process variation and instability, it degrades efficiency, consistency and ultimately, profits. A key manufacturing performance objective is the establishment of stable and predictable processes that limits variation – minimum variation around target values.

A main focus for USC Consulting Group is to identify the root causes of variation and address them. Generally, it boils down to people, components and materials.

Some examples to causes of variation include:

It can be one of these factors, several, or something else. But whatever it is, it’s impeding our ability – and the bot’s – to predict outcomes.

Minimizing variation with our Customized Quality System (CQS)

Every situation is different. The cause of variation on one manufacturing line isn’t going to be the same on another. USCCG assesses and evaluates client processes, then applies a customized approach using a series of tools, techniques and methods that is most applicable in addressing the causes of variability. This customized approach enables USCCG to address variability in an efficient manner. We call it our Customized Quality System (CQS).

We review processes from “the cradle to the grave” and identify the highest-impact operations, then drill down to the tasks and steps within those operations until we uncover the culprits.

Although every situation is different, the general roadmap includes:

Removing variability through our CQS not only has an immediate impact on improved product conformance but also paves the way for AI to do its job in predictive analytics, i.e., we want predictions with minimum variability.

It’s just one way USC Consulting Group is using the human touch to make sure AI is up to the job.

Read more about this in our free eBook, “AI and Machine Learning: Predicting the Future Through Analytics.”

AI and Machine Learning - Predicting the Future Through Data Analytics eBook

 

Back to top ↑

 

Statistical process control (SPC) is a commonly used machine learning software in manufacturing that measures the consistency of a product’s performance based on its design specifications. Minimizing variability is a crucial part of avoiding defects and maintaining resilient manufacturing operations.

This guide outlines the different ways that businesses can effectively utilize SPC and reap all of the benefits this technology has to offer.

How Statistical Process Control Works

SPC is a tried and true technology that businesses have been using for more than 100 years to improve their manufacturing operations. It conducts ongoing statistical analyses, taking into account factors such as the materials, design, employees who handled the product and the machinery used to create the product.

SPC’s constant vigilance enables businesses to make swift and accurate resolutions to quality control problems. However, it’s not fully autonomous like other manufacturing software that can identify statistical correlations without human help. Instead, it relies on large amounts of training datasets that another source must manually input to achieve the desired results.

This form of machine learning is known as supervised learning. Businesses can input human-labeled datasets by themselves, or they can recruit another algorithm to automatically input statistics in a process called “machine annotation.” In either case, SPC needs to absorb as much raw data as possible to maximize its efficiency.

SPC displays its findings in easy-to-read control charts, and it’s the business’s responsibility to set the parameters for each chart by providing the software with enough information. This process includes six basic steps:

  1. Define the manufacturing process you want to monitor and control by establishing the input variables, output variables, equipment, materials and any other external factors that might affect the process.
  2. Collect the data that the software extrapolated from the variables you provided, then organize it into a digestible format — usually a chart or spreadsheet.
  3. Select and construct the control charts based on the type of data you’re using, such as weight, length, temperature and any defects that might have occurred.
  4. Look for patterns in the control charts that indicate special cause variations in performance due to underlying defects. You can calculate process variability through a capability index, such as C, Cpk, Pp and PPk.
  5. Investigate the root causes of the variations and make the necessary equipment, material or operational adjustments to correct them.
  6. Continue to collect and organize data to identify more variations, updating the control specifications as needed.

This process sounds awfully similar to Statistical Quality Control (SQC), but there are some key differences. Statistical Process Control measures independent variables, while SQC strictly focuses on dependent process outputs. SQC also carries out acceptance tests by screening individual product samples, while SPC relies on large datasets and doesn’t have an acceptance testing feature.

Types of SPC Tools

Many types of analysis tools have developed during SPC’s century-long evolution. These tools are split into two main categories — basic tools of quality (7-QC tools) and supplemental tools (7-SUPP tools). Here’s a quick rundown of how businesses can use the 7-QC tools:

Stratification also often appears in the 7-SUPP tools category because of its versatility and importance to statistical analysis. Breaking up large datasets into smaller digestible chunks makes SPC software more accurate at identifying problems and reducing variability. Here are the other six 7-SUPP tools:

Today’s SPC software modules include all of these tools, allowing businesses to access dashboards that display the various charts and diagrams in one place. These insights can lead to identification of quantifiable improvement opportunities that maximize operational efficiency.

Benefits of Using SPC

SPC is one of the most effective machine learning resources for achieving consistent performance in manufacturing operations. Eliminating process errors allows businesses to simultaneously address the three biggest challenges in material handling — workplace hazards, equipment damage and carbon emissions — in many ways:

When these benefits combine, the final result is a more satisfied client base and a more profitable business. While SPC software can’t do all of the inspection work on its own, the tools and insights it provides are invaluable in a manufacturing environment.

Use Statistical Process Control to Its Full Potential

Business leaders who are willing to put in the necessary effort to provide SPC software with large datasets can use this technology to its full potential. They will gain access to numerous eye-opening statistics about operational inefficiencies and have all the knowledge they need to make accurate adjustments.

*This article is written by Jack Shaw. Jack is a seasoned automotive industry writer with over six years of experience. As the senior writer for Modded, he combines his passion for vehicles, manufacturing and technology with his expertise to deliver engaging content that resonates with enthusiasts worldwide.

Back to top ↑

Recently, I reached out to one of our most charismatic team members, Charlie Payne, and got him to tell me a little bit about his approach to new projects and, more specifically, his work in the food and beverage industry. Charlie is a Senior Operations Manager at USCCG, and has played an integral role in the development of our strategies and training programs since he came on board in 1990. He has years of experience in various industries including Food & Beverage, Mining, Oil & Gas, Life Sciences, and Manufacturing. Charlie’s innovative solutions and ability to build strong client relationships are the foundation for his long record of successfully completed projects. Here’s what he had to say:

When I start a new project, I certainly like to get face to face as soon as possible so we can get a sense of each other, understand the issues, and decide if we might be a fit. Usually, a half-day on site is enough to see if we want to do a more detailed two-week analysis where we’ll come in with a team and put together a business case and answer three questions:

  1. Is USCCG the right group to help drive results with this client – do we hit their team right?
  2. Are the issues we see addressable by USCCG in a timeline that makes sense? Is their management team open to change?
  3. Is there a viable business case? Usually we strive for a 2 to 4 return on the project costs, depending on the scope of the project. Generally, a bigger scope means a bigger return.

If the business case makes sense, then I like to go in to project mode quickly – the consistency of the team we use is important to us and a lot easier than having an extended decision process that means I need to acquaint a new team to the client.

As a Senior Operations Manager, I am responsible for delivering the results of the project and managing the partnership we have with our client. Working with the bench strength we have in our consultants – full time USCCG employees, many of whom I have worked with for years – can often make the projects successful and fun. To have fun with the client and my team is a key success factor for me.

food & beverage
I have worked with all types of clients, and I enjoy working in the food & beverage industry because it’s unlike other business sectors due to of the variability. In food processing, the input can often be so variable – size, quality, quantity – it can often make us wonder why we try to manage the process at all, when we are out of control right from the start! If you can’t affect the input, then in my opinion, it makes it even more important to control what you can – usually the process within the plant walls.

We’re certainly not making widgets and it’s not as precise as manufacturing. When dealing with nature we can be affected by harvest size, variations in bird or hog size, droughts, so you never know what you’re going to get. Combine that with a work force made up of a multitude of cultures, languages, and literacy skills and you have challenges getting consistency into the process. After 24 years and many businesses, the hardest position I have ever seen is to be a supervisor in further processing in a cold plant.

My work in the food & beverage industry has certainly been across the spectrum from large privately held to co-op, slaughter facilities to further processing. Lately, I’ve seen an increase in organic clients, as well as those expanding heavily overseas and wanting to right size their facilities.

The fierce competition and amalgamation from emerging markets are part of why our food and beverage clients look to us for opportunities to improve their processes to stay ahead of their competitors.

When we engage clients in this industry, we work across a number of fronts – from production plants with a focus on yield, throughput, and productivity – OEE type metrics, installing a Management Operating System (MOS) in the plant to bring consistency between shifts, lines, and plants.

Bringing in our LINCS Business Intelligence solution to allow for the key metrics to be known in a timely manner helps us focus management on the items they need to address in getting the performance they need. More recently, we’ve worked on reducing cost with a focus on the purchasing department and the spend on packaging materials, bringing great benefits to our food & beverage clients.

I believe good management practices are applicable across all industries and our solutions for our food & beverage clients can and have proven to be successful for various other clients.