“The value of experience is not in seeing much, but in seeing wisely.” – William Osler
Performance improvement strategies that focus on value-added activities have been popular for at least a couple of decades. Executives, managers and workers alike appreciate the simple idea of eliminating waste by eliminating “non-value-added” activities. Unfortunately for the multitudes of companies relying on this strategy, it has serious flaws. This article discusses:
Managers should not rely only on intuition to evaluate “value-added” activities for improving financial performance and customer service. Overcoming the flaws of the value-added approach requires more than just experience and intuition. Using “gut instinct” decision making and the value-added approach creates the “value-added fantasy” and decision evaluation blind spots which get worse as complexity of operations increases.
In a Harvard Business Review (May 2003) article, Don’t Trust your Gut, Dr. Eric Boneneau wrote, “The trust in intuition is understandable…But it’s also dangerous. Intuition has its place in decision making—you should not ignore your instincts any more than you should ignore your conscience—but anyone who thinks that intuition is a substitute for reason is indulging in a risky delusion. Detached from rigorous analysis, intuition is a fickle and undependable guide—it is as likely to lead to disaster as to success. And while some have argued that intuition becomes more valuable in highly complex and changeable environments, the opposite is actually true. The more options you have to evaluate, the more data you have to weigh… the less you should rely on instinct and the more on reason and analysis.” This speaks directly to the value-added fantasy.
The value-added fantasy is the perception that decisions based on subjective “value-added” evaluations and eliminating non-value-added activities will provide the most profitable performance for manufacturing, supply chain or service operations. If companies really want to get value out of the value-added approach, it should be combined with analysis and reason rather than used as an intuitive shotgun approach.
The problem with most value-added approaches is that they are highly subjective and dependent on intuition alone. Here are some typical definitions of value-added:
Some good things can result from value-added analysis. For instance, it typically shows that most of the time a product spends in a process is spent waiting. Lots of companies, especially those where little process improvement has been done, have gotten great initial benefit from this revelation. They have been able to eliminate some of this wait—or wasted—time. Additionally, the value-added concept is a simple, seemingly intuitive approach—eliminate waste, focus only on activities that provide value. However, an approach that is too simple is a problem when facing the complexity and variability of real world processes and challenges.
In the vast majority of cases, the customer does not care what activities a company labels as value-added. The customer makes a value statement with an exchange of money for goods or services at the time of purchase. When purchasing gasoline for their cars, people rarely, if ever, consider the value-added vs. non-value-added activities that might be taking place at refineries. The goal of most companies is to make a profit over the long-term, hopefully in ways that are legal. The goal is not to provide the most value. What customers would value most highly are products with perfect quality and best features provided whenever they want—for free. Subjective value-added labels, at best, only provide loose alignment between performance goals and financial goals. Since there is some correlation, good results may happen but it’s not because financial results are dependent on having the fewest non-value added steps—good results are a random occurrence. Following is a powerful example of how “intuitive” value-added thinking led in the wrong direction. Application of practical operations science quickly diagnosed the error and provided a very effective solution.
A multi-billion dollar medical equipment manufacturer designed and installed a single-piece flow line to reduce non-value added time in its production process. The intuition was that this would reduce waste, provide shortest possible cycle time (the time from beginning of production to completion of the finished product) and, thereby, generate best possible profitability and cash flow. The product was a blood filtration kit consisting of a number of plastic bags and components that were assembled and packaged for delivery. The production line is illustrated in the diagram below. In applying the operations science of Factory Physics® concepts, a remarkable result occurred that was counter-intuitive to traditional value-added thinking. In the end, the company was able to increase output of the line by 30% while increasing non-value-added time. We see this problem and its solution very frequently with implementations of one piece flow.
Error detection and correction was achieved by performing a simple application of operations science that only required four pieces of data:
A picture is worth a thousand words and the Quick Flow Benchmark (QFB) graph to the left illustrates the situation very well. The QFB is a module of Factory Physics’ CSUITE Operations Analytics software however the calculations can be made by hand. The QFB is a graphical representation of Little’s Law, a fundamental relationship in operations science. It is stated as:
Cycle Time = WIP/ Throughput
For much more description of Little’s Law, see the discussion beginning on page 82 in Factory Physics for Managers. For this brief discussion, some explanation of the QFB graph is required. The red lines, scales and points describe throughput. The blue lines, scales and points describe cycle time—the time from beginning of the production routing until completion of the routing. In this case, cycle time is measured from when components are introduced to the line until a completed kit rolls off the end of the line.
The independent variable on the x-axis is work-in-process (WIP). This is a key concept. Given capacity, demand and variability, the level of WIP in a process determines output. It doesn’t matter whether the WIP is classified as value-added, non-value-added, manna from heaven or brimstone from hell. When determining line performance, WIP level is a design parameter. Too little WIP is as bad as too much WIP.
The solid red line plots throughput at various WIP levels in a perfect world with no variability. The solid blue line plots cycle time at various WIP levels in a perfect world with no variability. For more discussion on what a perfect world looks like and a contrast to operations science in the real world, see the previous Factory Physics blog article, Finding Your Perfect World as a Manager.
The red triangle line and blue diamond line represent benchmark performance curves for throughput and cycle time respectively. These benchmark curves show performance in the presence of moderate process time variability. Above the red triangle line is good throughput performance. Below the blue diamond line is good cycle time performance. Though not shown on the QFB graph (since it’s a basic analysis using only four pieces of data), be aware that a process’ performance curve for throughput vs. WIP runs roughly parallel to the red triangle benchmark performance curve. Same for cycle time and the blue diamond curve. In this example, the throughput vs. WIP curve for the production line would start at 0, run through the Original throughput point and the Final throughput point remaining above and parallel to the red diamond benchmark performance curve. Displaying the actual performance curves requires more comprehensive data but this is a such a simple and powerful analysis that it is always a great place to start.
Now to drawing conclusions from the QFB graph—the small red triangle (original throughput) and the small blue diamond (original cycle time) show where the production line’s performance was on initial analysis. The analysis was performed in one day and it showed the original WIP level to be way too low given a demand of 26,000 pieces/day.
The original one-piece flow WIP level in the line was about 380 pieces. Notice that this produced a very short cycle time (original cycle time) of about 21 minutes (0.02 days). This was great from a value-added analysis but the tradeoff was a large amount of wasted capacity. The original throughput was 20,000 pieces/day but the capacity of the line was about 28,000 pieces/day—a 71% utilization level.
The solution? Increase WIP in the line. At a WIP level of 4,000 pieces variability in the line was properly buffered and final throughput (large red triangle) went to 26,000 pieces per day—a 30% increase in throughput! The tradeoff? Cycle time went from 21 minutes to a final cycle time (large blue diamond) of 158 minutes, a massive increase in non-value added time. Since the plant only shipped once a day and the WIP was inexpensive, the effect on customer service and working capital was negligible.
It turned out that the entire line was hard-coupled. Whenever any one of the feeder lines or the main line stopped, the entire line stopped. To achieve the 30% increase in throughput, the feeder lines were decoupled from the main lines and WIP buffers were established between each feeder line and the main line as illustrated in the following figure.
Some might say, “Well, they should have just reduced variability in the line.” If that can be done, do it. However, such an approach usually ignores the realities of the situation. Reducing variability is hard, particularly when it is related to machine performance. A more practical approach is to use WIP to buffer variability until the variability can be dependably reduced. In this case, the company had been running the line in its original value-added one piece flow design for 15 years. There had been performance improvement efforts such as speeding up the glue machines from a 4 second process time to a 2 second process time but the resulting throughput improvements had only been incremental. The value-added approach to line design and operation had put blinders on the operations staff and reduced throughput by 30% for 15 years. To their credit, once Factory Physics analysis and training showed the benefits of buffering variability with optimal WIP the staff acted quickly and efficiently to realize the improved throughput.
Recommendations for Avoiding the Flaws in the Value-Added Approach
Finally, the simplicity of the value-added approach can be beguiling. Do fall under its spell. Eliminating waste is a great idea but it is vital to understand what is wasteful and what is not. The goal is to improve profit and customer service, not reduce waste. Apply practical operations science concepts to predictively conquer operations challenges.
Factory Physics Inc. provides award-winning science, management consulting, training and software to help managers optimize performance of manufacturing supply chains.
FACTORY PHYSICS® and LEANPHYSICS® are registered trademarks of Factory Physics Inc.