I had just read the slides from Lean Reliability web seminar.
As a Reliability Engineer and Six Sigma Black Belt I can say that the Six Sigma methodology and philosophy and the Reliability tools fit together perfectly. Today all my reliability improvement projects are conducted as Six Sigma projects.
The RE tools and techniques can also give an additional help to the statisticians regarding the phisiscs of failures and having additional information to analyze besides statistical data only (only numbers).

Luiz F Ventura
Brazil

Anything to share with the forum then?

I attended a LEAN Six Sigma and two of the tools listed are FMEA and RCM.

Is six sigma another fad in maintenance or is there something different about six sigma?
I have done a small amount of research and not been trained ... To me, it seems to be very strict about evidence and using data. Problem is, in maintenance we dont have much data and what we have is not reliable usually, and at the end of the day, we do not want to collect data. We need to solve problems without much data so the six sigma rules about evidence and data seem to suggest six sigma is not the right tool for reliability professionals unless they are involved in expensive reliability growth programs (where experiements are set up) or companies with fleets of equipment operating in the same way.
I am interested in this topic because at the moment I am sceptical as to whether 6 Sigma has anything to add to RCM or the common RCFA methods.
Regards
Steve
www.pmoptimisation.com.au

Steve,

Like you I am not trained in 6 sigma but I have a good friend that is a black belt (indicating a high level of training). From discussions with him it seems that the main failing with 6 sigma is that by focusing on improving the existing process it misses out on the opportunity to apply a better process.

It kind of locks you in to data collection to improve something that may not be worth improving or that may be imporved far more simply.

Like all management fads (albeit a popular one) it will fade (remember Business Process Reengineering?)and ultimately good companies go back to fundamentals and solve problems.

I am a six sigma practitioner in our company, and I have completed Monash's Masters in Maintenance and Reliability, so I feel qualified to comment.

Six Sigma is, very simply, discipline. There is a "improvement" methodology, and a "creation" methodology. There are defined steps that guide a team to improve a product or process (or create anther one).

Sig sigma uses a collection of "tools" that have been in existance for decades, and would be familiar to industrial engineering or manufacturing people. Anyone who has been through six sigma training would realise there is nothing "new" about the tools, simply the way they are applied to improve a business situation. "Hard" tools include things like process mapping, Ishikawa diagrams, Gage R&R studies, Hypothesis testing, Root Cause Analysis, Design of Experiments, FMEA and control charting. You can even use Weibull analysis. "Soft" tools include things like communication planning, change management, risk management and training.

If the solution to "the problem" is known, then it is not a six sigma project, it is a Nike project ("just do it").

If you can't adequately define a problem and put a business case around doing the project, then you don't do it.

I have read the posts regarding data collection. Data collection ("Measure") is the second step in six sigma disciple. The reason for data collection is very simple. You establish a baseline before you "Improve". And you need to be able to demonstrate your improvement has actually been effective compared to the baseline measurement. In our company, significant decision-making based on opinions and emotions has (almost) been eliminated, and I guess our shareholders would appreciate that. The saying goes "In God We Trust. Everyone else bring data". Criticism of data and data collection is, I suggest (with all due respect), somewhat misinformed.

If there is no management support, then six sigma will become a "fad" and fade away like other business improvement initiatives we have all seen (even me in my short career).

Yes six sigma and its application is not perfect but nothing is. One significant problem I see is the reliance on the Normal distribution, which is fine for transactional and manufacturing processes, but not much value in reliability engineering where it rarely appears.

As an engineer, I like the discipline, and the process, and the tools. I also like meshing it with reliability improvement, whether you are improving a planning process, for example, or improving the reliability of a piece of plant. There is plenty of scope for using the six sigma discipline in reliability improvement, and I fully support the post above from Luiz.

It doesn't matter whether you use 8D, PDCA, Kepner-Tregoe or six sigma. The important factor is doing SOMETHING to improve your business' performance and competitiveness.

Six sigma is just one way of putting structure around business improvement.

Regards, Matt.

Good one Matt.
That seems to answer my question very well. Much appreciated.
Data collection, in my humble opinion is one of the least understood an poorly excetutes processes in maintenance. I have started to talk about these matters at conferences. What you have said about data seems to conform with what I think.
Regards
Steve

Hello to all,

Nice to hear from you again Steve, I just would like to share my experience, I am no black belt or aikido master, but one thing is very important here most specially for plants that have several in-house improvement initiatives such as six-sigma, lean manufacturing, TPM, Kaizen, RCM, RCFA etc.

It must be clear to all people as to which will be the main umbrella of the organization and that the rest of the improvement initiatives will be under that umbrella, if this cannot be well explained to the people, some will think that these improvements are just eating up my time through meetings and so on.

Ex. If lean will be the accepted umbrella, then 6 sigma, TPM, RCM will be the drivers and tools to carry out the improvement initiatives.

I believe that 6 sigma is mostly recommended for manufacturing as this will contribute mainly on having close to perfect yield 99.999999%

My Warm Regards,


Rolly Angeles
Teacher

Matt, can you post a review article on 6 sigma? How to collect meaningful data for reliability improvement?

As a Six Sigma Green Belt (enough training to be dangerous) I think that making Six Sigma operate on a maintenance department as a stand alone process is a mistake. The first thing I was taught in Six Sigma is the Voice Of the Customer. There is no point in developing a Six Sigma program if we don't know what is important to our customers (internal as well as external).

In an ideal world Six Sigma should be part of the overall organizations process. There is no point in designing a production line and then have purchasing save $120 by buying cheap bearings. The cheap bearings then become the maintenance departments nightmare.

Six Sigma should be all inclusive in terms of design, operations etc. Then Six Sigma and CRM, TPM etc. make sense as part of a complete process.

I am a materials engineer with a post graduation course in quality engineeging (just lacking the final work project to finish). Now I work as an oil analyst in predictive maintenance in a commercial lab (not a company). I am interested in getting involved in subjects related with data analysis and numbers, like statistical control and operational research and thus eventually finish the post graduation on these matters. I will attend a 24 hours course in 6Sigma.

My question is if 6Sigma is a good choice to start dealing with these sunjects and if is possible to integrate it with my present professional activity in oil analysis (ferrography)?

I am a SSBB and hold an MS in Reliability Engineering. I have a department that has 15 belts and we do many reliability related projects using the Six Sigma methodology. It is a great fit for reliability. Six sigma is about eliminating defects and reliability is about eliminating failures (defects). Key tools in Six Sigma are mapping, RCA, FMEA, Pareto Analysis and, most importantly, strong statistical tools just like good reliability engineering requires. It is good for tough problems and it requires disciplined data analyis. Perhaps the biggest advantage are the cross functional teams that develop solutions more robust than the RE alone could do. I just wrote an article on this at this link http://www.engr.utk.edu/mrc/April%2007%20Newsletter.pdf . Maintenance is a little different. Improvements in maintenance are often associated with work processes. I have found that the application of lean principles are much more suited for maintenance. I wrote about this in the January newsletter. Look under MRC Newsletters at this address: http://www.engr.utk.edu/mrc/ . It is lean thinking and the application of lean principles that yield the biggest improvement in maintenance.

In your opinion what are the future areas in maintenance, reliability and processes (or services)analysis/management so that I could start focusing on some particular field(s) or tecnhology(ies)?

It is great to see responses here from 3 of the graduates from our postgraduate programs in maintenance and reliability engineering. Well done!

Ray Beebe
Monash University

quote:

...Ishikawa diagrams, Gage R&R studies, Hypothesis testing, Root Cause Analysis, Design of Experiments, FMEA and control charting. You can even use Weibull analysis.

All of the above 6 Sigma methods in theory could be used in maintenance RE and some of them in fact, such as RCA, RCM, control charting are being succesfully used by many in maintenance field.

Things are not that simple though when statistical methods should be utilized. The problem, IMO, is lack of valid data.

In RE we are dealing with failures and when a reliability model has to be developed: constructing distribution, calculating reliability, making prediction, statistical assessment of improvement,etc. we normally don't have the luxury that product reliability engineers have - reliability testing. It is required to have a fleet, as Steve mentioned, of similar machines and time between failures data for a every major failure mode. In a manufacturing plant I doubt many people have it.

Dave

Six sigma is best suited for manufacturing plants, most specially when competition among industries is feirce. What matters most to these industries would be their customer as they are the once who provide work for the industry. Hence, for industries that are keen on Quality, Customer Satisfaction and the like then six sigma is a perfect fit. Their goal will be to have a yield of around 6 defects per million products produce, I am just not been updated if this had been change.

But for plants with different vision such as safety like the airline industries, mining, oil ang gas plants, I think six sigma would not be as useful compared to manufacturing.

Kindly correct me if I am wrong about this, previously working in a plant interfacing with black belts and having some participants in my training which are green and black belts provided me this thought.

Again, what I believe is all these tools and strategies are useful if we know how to use them and not to overuse them.

My Warm Regards,

I am GB certified and found it helpful for many reasons. The big one for me is the "sales charts and back-up info" SS training really did help me when I have to justify plans, ideas, or positions with Sr. management.
I think Matt is spot on.
Rolly, 6-sigma is very relevant to aircraft, manufacturing, and business process.
Data is an issue for everyone. You use what you have and tell the story of "how good" the data really is and how it was leveraged into the final answer.
Textron has adopted a 6-sigma culture across the board and it is becoming a requirement for all leadership positions. One justification for that is being able to all talk the same langue across industries.