Schonberger on Discontinuous Improvement, Frustration Data and the Big Problem of Factory Infrastructure

What happened to the customer in manufacturing? If you read the literature about what is going on in manufacturing, the customer doesn't appear very often, says Richard Schonberger. Moreover, he adds, neither does the employee base.

That's in stark contrast to the literature of the 1980s and 1990s, including Schonberger's own Building a Chain of Customers: Linking Business Functions to Create the World Class Company.

"There was a lot of talk of customer-focused manufacturing back then, and today it's just efficiency -- efficiency and utilization of manufacturing resources. It's back to that even though the lean community knows better. It's supposed to be short linkages all the way to the end customer," Schonberger says.

Not many people can speak about lean, Six Sigma and continuous improvement with more authority than Schonberger. He's been steeped in it for well over a quarter of a century and has shared his learnings in such books as Japanese Manufacturing Techniques: Nine Hidden Lessons in Simplicity (1982), World Class Manufacturing—The Next Decade: Building Power, Strength, and Value (1996), and Best Practices in Lean Six Sigma Process Improvement: A Deeper Look . . . with Telling Evidence from the Leanness Studies (2008), to name just a few, as well as hundreds of articles on manufacturing principles.

He's been a practicing engineer, a professor, and his robust research efforts include the "Leanness" Studies, which track inventory turnover of more than one thousand companies across the globe, and the World Class by Principles benchmarking project.

IndustryWeek caught up with Schonberger recently to hear his thoughts on a range of topics. Here is some of what he had to say.

Inventory and Things Go Wrong in Lean Land

In this new millennia, the highly dominant pattern is getting worse on inventory turnover, which is the easiest-to-understand marker of whether you're lean or not. You go into a factory, there's a lot of inventory. That's not lean. You go into their warehouses, that's really not lean. It's shocking to see the numbers in industry after industry, country after country … and it's a pretty telling indicator that things are going wrong in lean land. So I figured the story needs to be told.

Discontinuous Improvement

The state of continuous improvement is off kilter. I am seeing a strong tendency, a fairly long-standing tendency -- toward discontinuous improvement. Also, improvement has strayed from its earlier meaning. To talk about its earlier meaning I have to invoke a term that lost its cache two decades ago -- quality circles.

If you're really talking about continuous improvement, you should be talking about people on the front lines involved in improvement all the time. Unfortunately, that isn’t the case. Most manufacturers once believed in that, especially when [W. Edwards] Deming and [Joseph] Juran were alive and well and advocating things like that. But today it tends to be discontinuous, irregular projects, and most of the project team members are professionals. So the frontline people are left out, and that’s too bad because they know more about their process than anybody else.

It's been getting worse and worse for a long time. I hardly ever go into any factories anymore and see statistical process control charts. That's just one of many tools to keep shop people active. It's not necessarily the best tool, but it's symptomatic.

The Problem of Incomplete Improvement

I'm not saying [big projects focus] is bad. I'm saying it's incomplete. The projects undertaken in manufacturing companies tend to involve multifunctional groups -- that’s good -- and tend to attack fairly significant competitive issues. Sometimes their projects are really grand, affecting the entire workforce or are interdepartment in scope, or intercompany in scope. We need those kinds of projects, but we also need the continuous kind of improvement that involves collecting data every single day.

By the way, project-by-project improvement is data deficient. An improvement project team gets organized and gets chartered and takes off and, 'Woah, we don’t have any data. What kind of data do we need? Where are we going to get it?' On the other hand, if Dr. Deming were still alive, he'd say this kind of data should be collected all the time, continually, by everybody in the workforce – all the offices too -- recording everything that goes wrong or is going astray. Actually, my preferred way to collect that kind of data specifically is 'frustration' data.

Find Frustration Data

If you're asking your employees to collect data, don’t ask them to collect data on defects and errors and mistakes. Those are negatives and they will inspire defensive attitudes and behaviors. But if you say, 'We want you to record every frustration that occurs to you during the day' even if it's something having to do with your family – you're frustrated because you're at work and you’ve got a sick wife or something like that.  

Mark Royal and Tom Agnew of the Hay Group wrote The Enemy of Engagement: Put an End to Workplace Frustration--and Get the Most from Your Employees. They label frustrations as silent killers. This word frustration in that book really makes sense because anything that people are frustrated with on the job probably has a negative impact on the company – 'I'm frustrated because the parts don't arrive on time. I'm frustrated because the tools aren’t kept sharp and in the right place.' You could personalize the continuous improvement process instead of treating it abstractly.

Defect Data Will Arise Naturally

[Defect] data should crop up naturally, especially if the lean concept of cells is in there. Whenever you have a cell, the cell team is trading jobs all the time. They are responsible as a team, and if you don’t do it right at your station or your machine fails, the next process is affected immediately -- so that would be a frustration and the frustration would be exactly the same thing in that case as discovering a defect. I discovered a defect, but more importantly, I'm frustrated because the preceding machine is turning out bad ones.

The Big Problem of Infrastructure

You can hardly find anything written about equipping factories, and that's infrastructure. Our roads and highways and bridges and waterways are deteriorating and there is a big national hue and cry over it. Nobody but nobody is saying that about factories.

Sure we pay a lot of attention to maintenance of existing equipment, but existing equipment tends to be wrong for the purpose. And that's why I've been writing so much about concurrent manufacturing. Factories need to be equipped with multiple pieces of equipment so that you can produce multiple products at the same time.

Here's an example. A sodapop factory has several hundred, maybe even a thousand SKUs when you consider all the different brands and sizes and bottles and so on, yet they have only two high-speed assembly lines that are down for repair two hours out of every shift, and they can only make sequentially all of these many SKUs.

The lean person should be saying, 'Tear those high-speed, very high-cost mile-long conveyors out and put in 15 cells, 15 small cells each running at the rate of use of your ultimate customers' and then link to actually sales points to drive the schedule on these small machines.' Some people would say, 'We can't do that.' I know they can.

That same concept of multiple small pieces of equipment is alive and well in some industries, but mostly not. Most companies are striving for localized efficiency and never mind if they can make what the customers are actually [requesting].