How Not to Improve Manufacturing Productivity

Pharmaceutical manufacturing may be tougher than it has ever been before. It seems like once one has tackled everything, integrated best practices, purchased state-of-the art equipment, and hired consultants to help wring every last bit of profit out of the process, reduced margins throw a monkey wrench into the equation and the company’s leadership demands staff cuts, then slashes the capital investment budget.

Where can one look to for better profitability? Plenty of people are looking in some pretty strange places and doing some pretty unusual things. In pursuit of better margins, some are making colossal mistakes as they attempt to improve profits in today’s tough pharma environment. One thing’s for certain: No one can continue to do the same thing over and over again and expect different results — otherwise known as the definition of insanity. Like many people, trying one approach after another to lower costs is not an effective strategy. If that sounds familiar, it could be time to find new, more resourceful ways to get the job done, and discover more control over one’s destiny.

Pharmaceutical manufacturing and packaging has changed. Could it be time to make those adjustments one’s been putting off? Because in all this change there is opportunity, but such opportunity brings with it the inherent risk of failure. But failure is instructive and one can learn from the mistakes of others, so in that spirit, following are three common mistakes pharma executives make when attempting to increase manufacturing productivity.

Convincing someone that they “don’t know what they don’t know” can be extremely difficult. While speaking with plant operations people, GMs, VPs, etc., they often tell me how much they’ve spent on state-of-the-art equipment, how well they’ve adopted Six Sigma, how they’ve squeezed every last drop of cost out of their manufacturing process, etc., but remain frustrated because their lines aren’t living up to expectations.

Most of the executives running plants that I’ve encountered claim they track it, document it, analyze and minimize it — and that seems like so much “hogwash.” It’s not that executive managers don’t try, it’s because the process in which measuring and documenting downtime in most facilities is extremely inaccurate — and in many cases — not uniformly defined, or practiced from plant-to-plant or even line-to-line.

A lot of bonuses, ratings, and pats-on-the-back are tied to reporting good key performance indicators, and there is a deeply ingrained cultural bias against making downtime look too big or too bad. Over time, many plant operators have come up with methods for measuring efficiencies that omit the biggest losses. For instance, one can inflate one’s efficiency numbers if time related to clean-up, changeover, start-up, preventative maintenance, material shortages, breaks, meetings, training, etc. is omitted. In essence, if one only measures efficiency when lines are running successfully, one can report pretty good looking efficiency numbers. Everybody gets their bonus, but the company loses because this “look the other way” or “minimizing” approach conceals the underlying problems, that once fixed, could kick efficiency into high gear.

Many plants that are routinely reporting a line efficiency of 80-85% find that when implementing a more rigorous measurement criteria, such as Overall Equipment Effectiveness (OEE), that their true OEE is in the 50% range — or less. This can be a shocking discovery for middle managers who likely fear repercussions from management, so it’s critical that top management be involved in establishing a reward system based on accurate measurement of manufacturing productivity and foster a culture of improvement, rather than a culture of reporting the highest number. A lower starting number represents more potential for improvement. For example, if a line that is running at an OEE of 50% improves to 55% by developing rapid changeover methods, this correlates to a 10% boost in output.
 

Another common problem of human management is the under-reporting of downtimes. A situation that took the manager a reported five minutes to resolve may have actually have taken 20 minutes. What do you think gets reported? And here’s something else to ponder: Doesn’t it seem strange that all problems start at times like 10:10, 8:45, or 2:30, and are resolved in round numbers like 5, 10, or 45 minutes?

A very revealing phenomenon is to observe a line that implements a system with fully automated recording of downtime incidents. What do you think happens? Under these circumstances it is common for downtime incidents to increase ten-fold. Did the automatic reporting introduce problems? No! But it now faithfully reports every incident, in a very precise way: No emotion, no fudging. For instance, a typical pharmaceutical packaging line may have 1,000 short stop failures per week, averaging just 1-2 minutes in length, but each eats away at the line’s productivity. At first, this thought terrifies, but in time you have so much more feedback about your line, you can see and correct a whole series of problems that may have been hiding in the background. Conclusion? Systems that don’t automatically collect logged data significantly under report downtime. This makes it much more difficult to identify the real root causes. 
 

A manager has struggled through every asset utilization scheme he or she could find. That person’s optimized, been consulted, and Six Sigma’d until they were blue in the face. No matter what was tried, they just were not able to squeeze any more asset utilization out of the lines. Everything that can be done has been done, right? Research shows probably not. Most lines have an entire “new” layer of growth in asset utilization, hidden in full view. This layer lives in the following list; can you spot the areas that may need work?
 

The six major sources of lost productivity, per a TPM methodology, are:

1. Major breakdowns.

2. Setup and adjustments.

3. Short stops (idling).

4. Reduced speed.

5. Startup rejects.

6. Production rejects.
 

Do any these seem familiar? They happen on every production line, and can be minimized with the proper combination of accurate PLC (Programmable Logic Controller) data, good human oversight, and a tool that makes sense out of the mountain of data the line is generating.

Here’s an example: Sales are up: Good news — the wallet packaging line is at capacity. Bad news: It costs more than $4 million to add a new line. Interestingly, the secret to getting more production out of the existing line resides in the PLCs running the line. Most line managers see them as valuable tools in automating a line, but just as critically important — they generate heaps of data that if properly collected in a database and analyzed on-the-fly — will point the way to more savings.

Many lines are swarmed over by an army of clipboard-toting functionaries, obtaining the occasional error condition or fault report, but these folks and others fail to realize that the key to squeezing more productivity out of the line is to look at all the data the entire line produces. Manual process will always be inherently expensive and replete with human error, opinion and sleight of hand to insure conditions appear better than what they really are. Real-time knowledge is the key to uncovering the elements of any line that are impeding its productivity and manual data-collection methods — or even automated ones that result in reports days later — are a sure sign that the line is not performing optimally and not living up to its potential.

When business is good, real-time analysis of production can unleash the unused capacity of assets without a big investment in equipment and training. The growth that the board of directors and stockholders want is right there for the taking — and without the growing pains. And, when business is tough, this kind of real-time analysis can help scale back your production to only the necessary shifts, reducing overtime, waste, and help identify and eliminate underperforming assets in pursuit of lower depreciation and maintenance costs — especially on mature production lines where the easy gains have already been made. Finding ways to improve are much more subtle.

In real time, managers can immediately know what the real problem is, and walk (or run) out on the line and see how to solve the problem, perhaps fixing it just enough to get the line moving again.

Conclusion? Most lines inherently have substantial opportunity for improvement. In many cases, the expense of a new line can be pushed off into the future by practical analysis of current downtime, and focusing on major loss contributors.

Overall Equipment Effectiveness is music to most all pharmaceutical executives’ ears — the idea of having production lines producing at their peak — or at least close to it for hours on end. There are five essential components of a program to successfully optimize OEE, but if they are missing, productivity gains won’t stand a chance:

1. Management involvement to set the business objectives related to an improvement program and to maintain program focus. 

2. Ability to accurately measure productivity in real time.

3. Ability to accurately capture detailed reasons for efficiency losses (so corrective actions can be taken), including non-operating conditions like cleanup, changeover, breaks, meetings and preventative maintenance.

4. Ability to selectively involve the operators in key downtime events.

5. Web reporting and graphical Web dashboards for easy visualization of current performance information by Operations, Maintenance, Engineering and Quality teams.

Unfortunately, there are a lot more than three mistakes executive managers can make. Instead of putting the energy, effort and money into fixing any fatal mistake after the fact, it may be time to take a more careful look at how to intelligently and safely remove those impediments to high performance. Take advantage of the tools offered; learn and conquer the latest technology (it’s easier than you might think), and benefit from its promise.

The marketplace has changed — more changes are yet to come. All we know for sure is that it’s time to change with the market instead of waiting for the market to change. Those executives who make the investment now to improve their asset utilization and operational effectiveness — bringing them up to world-class levels — will assure that their companies are prepared for whatever the future brings.

Scott Klages is a vice president and senior manufacturing consultant at Parsec Automation Corporation. Klages began his adventures in manufacturing over 30 years ago as a struggling “jack-of-all-trades” in a small but energetic family-owned machine design and fabrication company based in Pittsburgh.