Why Production Automation Does Not Always Work as Expected and How to Fix It

Nothing is more frustrating than making an investment in hardware or software that is supposed to increase sales, productivity or profitability, only to discover it does not. Too many companies make investments in new technology that never achieve their goals. In this article, we will offer one reason why this occurs and offer a possible solution.

Elon Musk, the CEO of the electric car company Tesla, invested heavily in the robots used to manufacture the company’s newest model only to find production had not improved but slowed down.

Discussed on a recent episode of 60 Minutes, Musk explained that this slow down almost put the company out of business. Musk thought he had created a world class manufacturing operation using robots. But unexpected problems occurred, and the robots kept breaking down, creating slowdowns or bottlenecks in production.

Source: inverse.com/

For most people this may sound counter-intuitive, but it is understood by believers of the 1984 best-selling book “The Goal” written by Goldratt and Cox. It is considered one of the best books on operations and manufacturing. Early in the book, the hero Alex Rogo meets his former operations Professor named Jonah at a manufacturing show and has this interesting conversion:

Professor Jonah: Alex what are you doing here?
Alex Rogo: I’m here on a panel on robots.
Professor Jonah: I’m into research on robots too.
Alex Rogo: Really, ours have been a god send. Since we’ve put those things in, we’ve had a 36% increase in productivity.
Professor Jonah: You mean you’ve earned 36% more money?
Alex Rogo: Not exactly- the increase was just in a certain area.
Professor Jonah: I don’t understand- if you didn’t earn more money how you can say you’ve had an increase in productivity? Have you sold more products as a result of these robots? Did you reduce costs? Did your inventors go down?
Alex Rogo: No.
Professor Jonah: Well then, if you didn’t sell more goods and you didn’t reduce costs and your inventories didn’t go down, then you can’t tell me you’ve had an increase in productivity.
Alex Rogo: Yes, but my price per part has gone down and my efficiencies have gone up.
Professor Jonah: Are you using the right measurement?

That is the bottom-line question. Are you using the right measurement? We see this in our consulting work all the time. Someone wants to make make an investment look good, even though it is not working as well as it should. For example, your new Print MIS system is not resulting in faster estimates or your Web-to-print is not reducing your staffs’ time for order entry, so some people create a new measurement. Changing your KPIs (Key Performance Indicators) to make your investment look good is the wrong approach. The right approach is to find the root cause and overcome it until your KPIs improve.

The Impact of Bottlenecks on Workflows

“The Goal” demonstrates the impact of bottlenecks on production workflows. It shows that every workflow has a bottleneck. Overcome that bottleneck and the entire workflow is more productive. The ideas in the book have become a formal management theory known as the Theory of Constraint which explains that workflows are filled with bottlenecks, not only the largest one but a series of smaller ones that are often masked by the larger bottleneck.

Not understanding that workflows have smaller bottlenecks often hidden by the larger bottleneck is one reason why attempts to automate fail. The attempt may only address the largest bottleneck but not the smaller ones. For example, a printing company invests in an e-commerce or web-to-print software solution, to overcome their largest bottleneck in order entry. As a result, more jobs flow into the shop faster, creating new bottlenecks in scheduling, job tracking and billing. If those new bottlenecks are not addressed, it can appear that the first investment in the web-to-print software did not solve their problem. Just as Elon Musk and Alex Rogo discovered, you must not only address your largest bottleneck but also the problems or resulting bottlenecks.

In this example, the problem is not that the web-to-print software did not work, but instead the other bottlenecks in scheduling, job tracking, and billing needed to be addressed. It is becoming clearer every day that software solutions need to address a number of issues such as web-to-print, scheduling, job tracking and billing.