Analytics header

Saturday, July 19, 2008

Research about Agile Manufacturing Systems

I recently read a post in a Yahoo group that I am a member of thought was interesting in

.


Anthony Reardon posted:

"The first thing that came to mind when I read the goal of HMS at the entrance to this site was something I saw on National Geographic I believe. They were doing some deep water exploration and encountered a previously undiscovered form of life. Simply put it was a collective of single cell organisms that had adapted to work together- thereby creating an organization yet not classifiable as a self contained organism in the usual sense. Some had adapted a focus on bringing in materials, others to process, and others to move through. The benefits of the work were distributed throughout the organization. Could represent some of the earliest activities of holonic organization and I think a great starting point for discussion."

My response:

You bring up a very important point with the analogy ot biological systems. I would like to share some classification of the different approaches that have been present specifically relating to manufacturing systems. Most of this research was initiated in the 1990’s and has unfortunately have not evolved much since then as the original researches went in different direction including myself. All these approaches share a common goal - to achieve agility by imitating systems that exhibit chaotic dynamics and emergent behavior, which are the key to agility and of course adaptability. Emergent behavior is how these systems are able to deal with unanticipated situations; in fact this behavior is how they survive, competing with other organism/systems. This may be obvious when you consider the characteristics of agility or, better yet, observe agility in nature. The organisms that you describe have a goal: To sustain themselves and grow. These “unintelligent” living cells achieved symbiosis to create a new life form – that is emergent behavior. Interestingly some of the research in this field focused on “cellular automata” also related to research in the field of Artificial Life.

In general there have been 3 threads of research specific in the manufacturing system domain. They are:

Biologic Manufacturing System: This research was spearheaded by Kanji Ueada in the 90’s. The idea is quite far reaching and tries to imitate the behavior and mechanism of biological systems in a manufacturing systems. The main concepts are “Genetic Coding”, and “self Organization”. In a BMS the cell “attracts” the sub-components it needs, and “repels” the sub-components it produces. Ueda points out that the main problems that present manufacturing system have are caused by the separation of artifacts and information. Thus products and corresponding information are separately processed. He therefore he suggests that like in the realm of life, artifacts and information should be inseparable. This makes the coexistence of such contradictions as flexibility and autonomy possible. Read more here (http://www.cam-i.org/displaycommon.cfm?an=1&subarticlenbr=70)

Holonic Manufacturing Systems: The main notion of this concept is to imitate human social behavior in manufacturing system, as I posted in a post in this blog and in the Yahoo Group of course.

Fractal Factory: The theory regarding the Fractal Factory has its roots in Germany and was initiated by H.J. Warnecke – as introduced in his book “The Fractal Company (see it on Amazon). Warnecke suggests that the manufacturing systems of the future have to be seen by embracing the whole of reality by using a holistic approach. The Fractal Company attempts to reduce the consideration and phenomena in industry to a common denominator by applying this holistic approach. The term fractal is used because of the nature of fractals, which according to Warnecke should be imitated in the manufacturing systems. It is an independent entity within a corporate structure whose goals and performance can be precisely described. Warnecke lists the following characteristics for fractals: Self-similarity, Self-organization, Self-optimization, Goal-orientation, and Dynamics.

There is also an excellent paper written by some folk at the CSIRO Manufacturing Science & Technology, Preston, Victoria, Australia that compares these difference concepts.

Much of the research at the time was focused on realizing some of the mechanics and behavior of these system using Multi-Agent systems – since the structure and behavior of these systems resembled the same attributes that made Bbiological, Holonic, and Fractal so agile.

My personal area of research during these times was Holonic Manufacturing System as I pointed out in earlier post in the Yahoo group. Take a look at my Blog for some more on HMS as well. If you are interested in more information let me know I can share some publications that I have on these topics.

Monday, June 2, 2008

Discussion about metrics and KPIs

The following are snippets of a discussion that I have had with Scott Stephens (read his blog ) that started as an answer to a LinkedIn question. I think that it has some important and interesting points about metrics and KPIs.

Question from Scott:
What is your source for KPIs or business metrics? There is a challenge identifying which business measurements are important, how to measure, and how often to measure.

Me
Metrics are used to gauge and measure how a business is performing. Performance is how effectively a business is achieving its defined goals. Hence metrics only make sense in the context of business goals. In order to find the important metrics for your business you should ask yourself what are my business goals and how do I know that I am achieving or performing to the defined goals. Dr. Deming once said that: “Running a company on visible figures alone is one of the seven deadly diseases of management. Manage what you can't measure“. If you are looking for financial metrics the Balanced Scorecard methodology should be a starting place. My suggestion would be to make sure you well defined goals and then figure out how you would measure you are performing against these goals. Those will be the metrics that matter to you.
Scott: If I was specifying an SOA approach to systems, I would want to know what is the required operating spec and how would I state it. So if I know that to achieve a production schedule I need raw material availability, what is the metric to measure raw material availability? What is the definition? Is there a resource that I can share with my supplier? Would they agree that it is a “standard?” Does CAMSTAR software recognize those standard metrics? If so, where do they come from?
Me: Manufacturing KPIs are not practical to standardize. I have done some research into the matter and what I have found is that although there are some definitions of general (or meta) standard calculation their implementation differ. This phenomenon is not only related to KPIs in manufacturing but also to Manufacturing System that are notoriously heavily customized. It is my contention that a standard Manufacturing System, as well as manufacturing KPIs that can be applied broadly is an illusion. The main reasons are a direct result of the dynamic nature of the manufacturing shop floor. Even in companies that try to standardize you will find difference between production lines that manufacture the same product or product family. This is because manufacturing systems evolve; driven by the need to optimize and reduce waste. That is of course is also part of the continuous improvement philosophy. Hence KPIs that are defined today to measure performance toward a goal may not be valid tomorrow since the goal and/or environment have changed.
Scott: I agree that based on my research there is little standardization in the world of business measurements but I am not sure that this is a function of the uniqueness of the users or simply not much work in defining and developing metrics resources. I also agree with the continuous evolution of measurement and process but (and maybe this is my background in systems engineering) I believe that a top-down approach organized by high level measurements will support a results driven business and accommodate the change that you high light. Schedule performance, for example, is a metric that I think crosses multiple industries and organizations. The broad definition and method for calculation will not change although the mechanics for driving change in the measured result will.

Saturday, April 19, 2008

Mission and Vision

This topic has always been close to heart since I have experienced both the good and bad side of it. Recently I answered a question on LinkedIn on this topic and thought it would be a good idea to share it.

Here is what I replied:
"I believe that it has to do with leadership. Vision and Mission statement like many other things are part of a company’s culture and organizational development, and have everything to do with motivation and leadership. In order for Mission and Vision to have an effect in an organization, the people have to
1) Know them
2) Understand them
3) Believe in them

In most cases Mission and Vision are known at best, rarely are they understood by anybody else other than the team that came up with them, and even more rarely to people truly believed in them. Intriguingly, when people believe in these statements they are not really necessary, since at that point they are simply part of the companies’ culture. In such cases you will also observe a correlation between strong leadership and organizational focus.

So it really doesn’t help to keep pushing these to people to remember or memorize the statement, you have to motivate them to know, understand, and believe in it."

You can view the whole thread here.

But really what prompted me to write this posting is a proverb I came across:

"Tell me, I forget. Show me, I remember. Involve me, I understand."
-- Chinese Proverb

Saturday, April 12, 2008

The Quality Monster

When I was fresh out of college and got involved in my first consulting gig I was somewhat confused to learn about the prevalence and power of the “quality department” in most of the companies that I was working for. Through my studies I learned that quality was a controlling concept that manufacturers used to make sure that everything was – if you will - ok with the product. It took my quite a while to understand and grasp the extent of this monster of an organization, which seemed to have a deliberate separate existence within each company.

Where did this quality organization come from? It seemed unnatural to have quality managed by a separate entity. My understanding at the time was that quality was inherently part of the process. It came from good engineering and good craftsmanship – with a sprinkle of discipline. If you separate the quality management from the process, you create a conflict of interests. The people involved in each department strive towards different goals.

Why then do we have huge quality organizations in regulated industries? Ah-ha “regulated” that may be the clue. It seems to stem from the push for compliance. Regulated bodies like to have confidence that there is compliance. How better to show this other than be dedicating a separate organization to secure compliance? So it seems that in fact its politics and not anything to do with good engineering or manufacturing practices.

In my mind quality and product are tightly connected. Quality products is what we want, there is no sense in product with no quality. Therefore it makes no sense to separate these to organizationally either. Like a good German engineered car (or Japanese nowadays) it is a result of an effective production system. In fact in quality is something you get free if you simply Lean-out your process. Quality is inherent to the process. However you can’t just tell your quality department to go home one day? Is it me that fails to see the benefits in what we consider best practices?