Tag: Statistical Process Control

The Point of No Return

The “Point of No Return” is a common expression that typically means you’ve reached an unrecoverable state if you continue to proceed with the current course of action.  When I clicked the “Publish” button for this post, I reached a point of no return (action).

From an accounting perspective, the term “Break Even” point is used to define the point where Total Costs equal Total Revenue.  The break even point translates to the quantity of parts that must be produced and sold to turn a profit.  Stock exchanges around the world serve as a constant reminder that investors are only concerned with PROFIT and return on investment (PROFIT).  In this context, a point of no return (profit) also exists.

Businesses exist for the customer or consumer.  Poor quality, missed delivery dates, short shipments, warranty returns, and poor customer service all lead to higher costs and may eventually cause customers to reach their “point of no return”.  Customers understand that the lowest price is not always the lowest cost option in the long run.  Business depends on repeat customers.

What’s the point?

In the simplest of terms, our actions must yield a return that is greater than the investment required to achieve it.  Delivering VALUE to the customer is one of the underlying principles of lean thinking and is measured by our ability to provide the highest quality products and services at the lowest possible cost, on time, delivered on time and in full.

This all sounds great on the surface but there will come a time where the cost to improve your systems and / or processes will exceed the return on investment – another point of no return.  Alternative, lower cost, solutions must be found to meet your continuous improvement objectives.

Where a significant capital investment is required, your company may require a payback period of one or two years.  A capital investment for a program that is soon to become obsolete is not a feasible option.  The point of no return (investment) is reached before any funding can even be considered.

The Bottom Line

Understandably, the team will become extremely frustrated when the very solution they proposed is rejected or declined.  While they may not doubt their own ability to provide viable solutions, they will doubt the company’s commitment to pursue excellence and continually improve.

For this reason, it is essential for the team to understand the reasons why.  It also underscores the need to identify and respond to improvement opportunities quickly and as early as possible during the launch cycle of any new system, process, or product.

Embrace Rejection

Rejection can sometimes be a gift.  As I have stated many times before, “There’s always a better way and more than one solution.”  Could it be that sometimes bad things happen for a good reason?

Rejection provides (forces) us with the opportunity to consider the present circumstances from a fresh perspective.  If the premise for the proposed solution was to “fix” the current system or process as it’s is now defined, perhaps a radically different and innovative system or process could better serve the company in the long term.

Is it possible that a new and lower cost alternative exists that could be at least as effective and perhaps even more efficient?  There are numerous examples of systems, processes, and technologies that exist today that were discovered by removing the limits that we unconsciously place on the scope of the problem that in turn limit the solutions we are able to develop.

The real problem with problem solving is the idea that the only solution is a “fix” to a system or process that is already be flawed from the onset.

Be Inspired

TED Talks are rife with examples of problem solving that yield radical and in some cases simple solutions.  The following TED Talks may serve to inspire you and your organization to look at problems and their solutions from a different perspective:

These TED Talks present problems on a different scope and scale than we may be accustomed to, however, the very discussion of alternatives alone should serve to inspire radical thinking that in turn inspires radical change.

You may have noticed from these TED Talks that some of the solutions presented were found outside of the context or circumstances from which the problem originated.  Is it possible that a “surrogate” solution exists elsewhere?

“Problems cannot be solved by the same level of thinking that created them.”

The point of no return is significant and literally requires “out of the box” thinking.  Many companies no longer grace our communities or employ our neighbours, losing business and opportunities for growth to lower cost manufacturers and distributors to continually emerging global economy.  The difference could very well be how we embrace the point of no return.

Consider that Toyota, as a new company to the North American automotive market, implemented innovative supply chain,  inventory management, and production techniques to remain competitive.  Radical change and innovation does not imply higher cost or investment.  At best it should simply imply “different”.

Other companies like Apple and GE managed to change their futures under the leadership of Steve Jobs and Jack Welch respectively.  Was it always pretty? Likely not from the books I’ve read.  However, the outcomes are undeniable.

The courage of Steve Jobs to solicit support from Microsoft’s Bill Gates was an extremely radical decision at the time.  This video, “Steve Jobs, Bill Gates and Microsoft – It’s Complicated“, clearly demonstrates the challenges faced in the relationship between Apple and Microsoft.  As for GE, I highly recommend reading “Straight From the Gut” by Jack Welch to best understand the radical changes in business and company culture during his tenure there.

Asking the right questions, open minds, radical thinking, and strong leadership coupled with a commitment to pursue excellence, continually improve, and solve problems may help everyone realize that the point of no return can be one of the greatest gifts you’ll ever receive.

To quote Albert Einstein, “A clever person solves a problem.  A wise person avoids it.” and so we … “look before we leap.”

Your feedback matters

If you have any comments, questions, or topics you would like us to address, please feel free to leave your comment in the space below or email us at feedback@leanexecution.ca or feedback@versalytics.com.  We look forward to hearing from you and thank you for visiting.

Until Next Time – STAY lean

Versalytics Analytics
Advertisements

Method Matters and OEE

English: This figure demonstrates the central ...
English: This figure demonstrates the central limit theorem. It illustrates that increasing sample sizes result in sample means which are more closely distributed about the population mean. It also compares the observed distributions with the distributions that would be expected for a normalized Gaussian distribution, and shows the reduced chi-squared values that quantify the goodness of the fit (the fit is good if the reduced chi-squared value is less than or approximately equal to one). (Photo credit: Wikipedia)

Tricks of the Trade

 

Work smarter not harder! If we’re honest with ourselves, we realize that sometimes we have a tendency to make things more difficult than they need to be. A statistics guru once asked me why a sample size of five (5) is commonly used when plotting X-Bar / Range charts. I didn’t really know the answer but assumed that there had to be a “statistically” valid reason for it. Do you know why?

 

Before calculators were common place, sample sizes of five (5) made it easier to calculate the average (X-Bar). Add the numbers together, double it, then move the decimal over one position to the left.  All of this could be done on a simple piece of paper, using some very basic math skills, making it possible for almost anyone to chart efficiently and effectively.

 

  1. Sample Measurements:
    1. 2.5
    2. 2.7
    3. 3.1
    4. 3.2
    5. 1.8
  2. Add them together:
    • 2.5+2.7+3.1+3.2+1.8 = 13.3
  3. Double the result:
    • 13.3 + 13.3 = 26.6
  4. Move the decimal one position to the left:
    • 2.66

To calculate the range of the sample size, we subtract the smallest value (1.8) from the largest value (3.2). Using the values in our example above, the range is 3.2 – 1.8 = 1.4.

 

The point of this example is not to teach you how to calculate Average and Range values. Rather, the example demonstrates that a simple method can make a relatively complex task easier to perform.

 

Speed of Execution

 

We’ve written extensively on the topic of Lean and Overall Equipment Effectiveness or OEE as means to improve asset utilization. However, the application of Lean thinking and OEE doesn’t have stop at the production floor.  Can the pursuit of excellence and effective asset utilization be applied to the front office too?

 

Today’s computers operate at different speeds depending on the manufacturer and installed chip set. Unfortunately, faster computers can make sloppy programming appear less so. In this regard, I’m always more than a little concerned with custom software solutions.

 

We recently worked on an assignment that required us to create unique combinations of numbers. We used a “mask” that is doubled after each iteration of the loop to determine whether a bit is set. This simple programming loop requiring this is also the kernel or core code of the application.  All computers work with bits and bytes.  One byte of data has 8 bit positions (0-7) and represents numeric values as follows:

 

  • 0 0 0 0 0 0 0 0 =   0
  • 0 0 0 0 0 0 0 1 =   1
  • 0 0 0 0 0 0 1 0 =   2
  • 0 0 0 0 0 1 0 0 =   4
  • 0 0 0 0 1 0 0 0 =   8
  • 0 0 0 1 0 0 0 0 =  16
  • 0 0 1 0 0 0 0 0 =  32
  • 0 1 0 0 0 0 0 0 =  64
  • 1 0 0 0 0 0 0 0 = 128

To determine whether a single bit is set, our objective is to test it as we generate the numbers 1, 2, 4, 8, 16, 32, 64 and so on – each representing a unique bit position in binary form . Since this setting and testing of bits is part of our core code, we need a method that can double a number very quickly:

 

  • Multiplication:  Multiply by Two, where x = x * 2
  • Addition:  Add the Number to Itself, where x = x + x

These seem like simple options, however, in computer terms, multiplying is slower than addition, and SHIFTing is faster than addition.  You may notice that every time we double a number, we’re simply shifting our single “1” bit to the left one position.  Most computers have a built in SHL instruction in the native machine code that is designed to do just that.  In this case, the speed of execution of our program will depend the language we choose and how close to the metal it allows us to get.  Not all languages provide for “bit” manipulation.  For this specific application, a compiled native assembly code routine would provide the fastest execution time.  Testing whether a bit is set can also be performed more efficiently using native assembly code.

 

Method Matters

 

The above examples demonstrate that different methods can be used to yield the same result.  Clearly, the cycle times will be different for each of the methods that we deploy as well.  This discussion matters from an Overall Equipment Effectiveness, OEE, perspective as well.  Just as companies focus on reducing setup time and eliminating quality problems, many also focus on improving cycle times.

 

Where operations are labour intensive, simply adding an extra person or more to the line may improve the cycle time.  Unless we change the cycle time in our process standard, the Performance Factor for OEE may exceed 100%.  If we use the ideal cycle time determined for our revised “method”, it is possible that the Performance Factor remains unchanged.

 

Last Words

 

The latter example demonstrates once again why OEE cannot be used in isolation.  Although an improvement to cycle time will create capacity, OEE results based on the new cycle time for a given process may not necessarily change.  Total Equpiment Effectiveness Performance (TEEP) will actually decrease as available capacity increases.

 

When we’re looking at OEE data in isolation, we may not necessarily the “improved” performance we were looking for – at least not in the form we expected to see it.  It is just as important to understand the process behind the “data” to engage in a meaningful discussion on OEE.

 

Your feedback matters

 

If you have any comments, questions, or topics you would like us to address, please feel free to leave your comment in the space below or email us at feedback@leanexecution.ca or feedback@versalytics.com.  We look forward to hearing from you and thank you for visiting.

 

Until Next Time – STAY lean

 

 

 

Versalytics Analytics

 

OEE: The Means to an End – Differentiation Where It Matters Most

A pit stop at the Autrodomo Nazionale of Monza...
Image via Wikipedia

Does your organization focus on results or the means to achieve them?  Do you know when you’re having a good day?  Are your processes improving?

The reality is that too many opportunities are missed by simply focusing on results alone.  As we have discussed in many of our posts on problem solving and continuous improvement, the actions you take now will determine the results you achieve today and in the future. Focus on the means of making the product and the results are sure to follow.

Does it not make sense to measure the progress of actions and events in real-time that will affect the end result? Would it not make more sense to monitor our processes similar to the way we use Statistical Process Control techniques to measure current quality levels?  Is it possible to establish certain “conditions” that are indicative of success or failure at prescribed intervals as opposed to waiting for the run to finish?

By way of analogy, consider a team competing in a championship race.  While the objective is to win the race, we can be certain that each lap is timed to the fraction of a second and each pit stop is scrutinized for opportunities to reduce time off the track.  We can also be sure that fine tuning of the process and other small corrections are being made as the race progresses.  If performed correctly and faster than the competition, the actions taken will ultimately lead to victory.

Similarly, does it not make sense to monitor OEE in realtime? If it is not possible or feasible to monitor OEE itself , is it possible to measure the components – Availability, Performance, and Quality – in real-time?  I would suggest that we can.

Performance metrics may include production and quality targets based on lapsed production time. If the targets are hit at the prescribed intervals, then the desired OEE should also be realized.  If certain targets are missed, an escalation process can be initiated to involve the appropriate levels of support to immediately and effectively resolve the concerns.

A higher reporting frequency or shorter time interval provides the opportunity to make smaller (minor) corrections in real-time and to capture relevant information for events that negatively affect OEE.

Improving OEE in real-time requires a skilled team that is capable of trouble shooting and solving problems in real-time. So, resolving concerns and making effective corrective actions in real-time is as important to improving OEE than the data collection process itself.

A lot of time, energy, and resources are expended to collect and analyze data. Unfortunately, when the result is finalized, the opportunity to change it is lost to history.  The absence of event-driven data collection and after the fact analysis leads to greater speculation regarding the events that “may have” occurred versus those events that actually did.

Clearly, an end of run pathology is more meaningful when the data supporting the run represents the events as they are recorded in real-time when they actually occurred.  This data affords a greater opportunity to dissect the events themselves and delve into a deeper analysis that may yield opportunities for long-term improvements.

Set yourself apart from the competition.  Focus on the process while it is running and make improvements in real-time.  The results will speak for themselves.

Your feedback matters

If you have any comments, questions, or topics you would like us to address, please feel free to leave your comment in the space below or email us at feedback@leanexecution.ca or feedback@versalytics.com.  We look forward to hearing from you and thank you for visiting.

Until Next Time – STAY lean

Versalytics Analytics