All 8 entries tagged Piuss

Six Sigma and Service industry

I was talking to dad today and the topic of what I have learnt over the past 5 months came up. I told him business improvement is a underlying theme throughout most of our modules. In particular, I bring up Six Sigma telling him how I am doing a dissertation on it. Then I was surprised when he said he heard about it while citing the success GE had with it. But he didn't know exactly what it is except that it is being used in many large corporations.

Then I went into a bit of effort explaining (poorly) the bell shaped curve and the DMAIC methodologies . And he suddenly throw a question at me "So how will it help my company?". I was a bit taken aback because I wasn't expecting it. So I fumbled a few words like "Well, your company is a importer of milk powder, and your main customer are pharmacies... but six sigma is more suitable to manufacturing... but it may be able to help with the admin side of the business...". It was easy to see I was totally unprepared for his question.

Later that night, I spent some time reading an article on application of six sigma in Seagate UK (PC hardrive manufactuer) and it made the following comment

"Thus, the case findings, consistent with the literature, conclude that six sigma lacks evidence of success when applied outside the mass manufacturing environment" (McAdam & Lafferty 2004; p546)

So it had me wondering how useful Six Sigma is outside the manufacturing industry. I remember during the PIUSS module, the question of "Six Sigma in Service" came up several times, and the general idea at the time was "Yes! Six Sigma can work in non-manufacturing (service) industry too". But now I am less convinced and would like to see some evidence of that statement somewhere.

McAdam, R. and Lafferty, B. (2004). A Multilevel case study critique of six sigma: statistical control or strategic change? International Journal of Operation & Production Management. Vol24. (5). p530-549.

: 31 Mar 2009 23:37 |  Tags: Mbe Piuss |  Comments (3) |  Close comments |  Report a problem

Experimental Planning – in the context of Taguchi

I remember when I did my undergraduate research project. I often rushed to carry out the experiments without thorough considerations beforehand. The end result is I realise there is a flaw in my experimental design, and make my result difficult to interpret or impossible to draw a valid conclusion. The consequence is plenty of re-do's and re-work which were both time consuming and expensive.

Today (Wednesday) I continued working on Taguchi experiment. I find the initial phase of research and experimental planning to be very time consuming. I needed to consider many aspects of experiment (selecting arrarys, control factors, assigning factors) all before I even start folding the planes. This is to ensure I do not make costly mistakes later on in the conducting experiment. This is the same idea as DFSS, where we want to make all the changes as early as possible to prevent costly changes later on. One book I frequently refer to for this PMA is Taguchi - Hands on approach (Peace 1997) is quite useful in this respect because it takes me through a step by step process, emphasising pitfalls and common mistakes, ensure all necessary precautions are taken to avoid making mistakes.

But the truth is even if I gave my best shot at paying careful attention to the experimental design, there always seems to some slip-ups somewhere in your experiment which you realise much later. For example, after I determined my design variable, and started folding planes. I forget to consider the effect of combinantorial (?) layout of orthorgonal arrays, which means some of the control factors are mutually incompatible with each other. Specifically, I couldnt have  planes with a 6 inch wing span AND a 1.5 inch high wing tip both at the same time (due to paper size constraint)! This means I had two planes that turned out to be impossible to build according to my design specifications. It's too late to change things now because that would take too much effort. I will simply point that out in my pma reflection.

I guess the important lesson to take away from this, is the importance of spending time and effort on the initial experimental planning. Making sure you have considered everything that can go wrong in your experiment and take steps to mitigate the effects in case they happen. In most PMA's this is not so important because we can easily make changes to our essays on the computer. But for the project, especially if it involve surveys or experiment. It is handy to study books on experimental designs , many of which are mentioned in the REME module.

: 12 Mar 2009 05:11 |  Tags: Mbe Piuss |  Comments (1) |  Close comments |  Report a problem

Recommend a book on Lean production

Today I came across reading a book on Lean production "The Machine that changed the world" James P. Womack, Daniel T. Jones and Daniel Roos; 1990; Macmillan. I got this book initially to read about supply chain management but realised this is actually a book about Lean production of Toyota.

I recommend this book because

• I believe it is one of the first texts on lean, according to the book the authors coined the term lean
  

• It is easy to read. It begins with a history of automobile manufacturing from craft production to mass production which has ills and inefficiencies that eventually give rise to Japan's lean production. The clear logical flow allow you to see why lean become important. What I find most useful is it clearly contrast the management philosophies between the West and Japan
  

  Mass Production	Lean Production
  Produce everything in mass leads to inventories	Produce only what is needed, aim for zero inventory Kanban system
  Market style relationship with supply chain	Seek long term relationship with suppliers
  Clear separation of design with production	Provide suppliers only with performance specs, but allow supplier to come up with product specs
  Tolerance mentality toward defect, defect is seen as 'inherent' and cannot be eliminated	'5 why's' problem solving system (similar to root cause analysis). root cause of defect is eliminated and will never happen again
  Firefighting or 'fix it' mentality toward problems	Kaizen gradual improvements invovling small teams and devote time to reflect and solve problems

One of the amazing idea I find is

In Western style assembly line, management promote maximum throughput and allow no stoppage in assembly line. So line managers have no incentive to fix minor problems because his ass is on the line if assembly stops. This result in small problems snow-balling into big problem in end product, and within production process.

Lean production devolve responsibility down to floor worker,  any worker is free to stop assembly line if a problem is discovered, and group of workers will work on solving problem together.

Not surprisingly, when each worker is free to halt the process, the assembly proces actually almost never stops because the process defects has been eliminated. Whereas to impose non-stop assembly always end up stopping the assembly because of various small problems during production. And the end product often has  many defect which needs to be checked and fixed, incurring significant cost on repair and quality checks.

• Detailed description of lean and how it relate to production, supply chain, customer and so on. Page 55-57 describes lean principles in assembly line which resembles six sigma (teamwork, quality circle, root cause)

Verdict: Even though I have only spent a couple hours reading this book, I can already grasp the keys ideas about lean (compare this to some books you read for hours but still have no idea what the topic is about?). This text is filled with rich examples from Toyota company and is very easy to read. I'd recommend this to people studying lean for PIUSS or for their project, if they havnt already come across it.

: 08 Mar 2009 12:20 |  Tags: Kbam Mbe Piuss |  Comments (1) |  Close comments |  Report a problem

Flight testing II

Follow-up to Flight testing from Kang's blog

After some more testing today incorporating some design changes, I have isolated factors influencing flight performance

• Wing span
• Wing angle
• Head weight
• Folding wing tip
• Fold wing trailing edge
• Cut out plane head

Especially the ones in bold made significant impact on performance. Once this is realised, I spent sometime with my next door neightbour discussing why this is so (who by the way showed me how to fold his 'ultimate paper plane' ; see picture below) .We finally agreed that

1. Head weight helps to keep plane head down (lower pitch angle) during flight (too much pitch angle create stalling).
2. Folding wing tip helps focus airflow above the wing surface. This in turn increases 'above' wing speed and hence lower 'above' wing air pressure according to Bernoulli's principle
3. Folding wing trailing edge up increase climb, whereas fold down makes plane dive. Folding one up and the other one down makes plane turn corners.
   

Ideally to keep plane aflot, the plane should not climb too much so that stalling occurs. Equally, it should not dive. So to keep plane 'leveled' , different adjustment to the head weight and folding trailing edge is needed. I think this could be a potential source of 'interaction' in Taguchi experiment. Another observation I got from my neighbour's design is plane can go through a series of climb-dive fly patterns to stay in air longer, I have yet to figure out how this happens.

I think these discoveries are good, I feel I finally discovered some significant control factors. Now I need to proceed into my next stage- experimental design, and finding a way to measure time, and find a big space to carry out my experiment.

I have spoke to sport centre and I think I can use the basket ball courst in the moring (730am) during the week. But I still someone to do measuring time for me. I understand everyone will be busy this week with KBAM and PIUSS , is anyone able to help ??

First prototype - Cut plane head; folded wingtip and folded wing trailing edge; paperclip added to head (not visible)

Ray's Flying Champion - My neighbour's ultimate flying champion. According to him this one can fly for ages and ages. It's a very different design with much bigger wing aspect ratio. The head is folded several times to make plan head 'heavy'. Wingtip is 'rolled' instead of folded. Strangely this one has a habit of flying in circles and go through several 'climb dive' patterns.

: 07 Mar 2009 22:51 |  Tags: Mbe Piuss |  Comments (1) |  Close comments |  Report a problem

Flight testing

Today I began doing some preliminary tests on my paper airplanes (or gliders). I just want to 1) try a few small things like folding the wing tip up/down, changing the wing area, to see what effect they have on flight time. I was actually quite surprised at the result especially when some of the modifications I made produced unexpected results (like when I expected not to fly, it did, when I expect it to fly, it crashes etc). Some of the more consistent finding I observed were

1) The way I throw launch the plane has a significant impact on the flight time/ distance

Initially I did not expect this to make much difference because I thought after the launch phase, the plane will just glide according to its aerodynamic properties conferred by its design. But I noticed sometimes on the same plane I get different results. When I tried to make my throw consistent, I noticed a relatively horizontal throw (very little angled throw) will produce much better flight distance than a more vertical throw (aimed higher say around 30 degrees from horizontal).

This finding made me reconsider whether to use throwing angle as my 'noise' factor in my Taguchi experiment. Initially I thought from a customer's view , one may not pay careful attention to the way he/she throws the paper aeroplane, so a 'robust' plane could be something that could be thrown at any angle and still produce good flight results. But since any throwing angle of above 20 degree or so may potentially reduce flight performance significantly, there may not be much point changing the angle of flight throw (i dont know...) .

Then I read about several different launch method. One is to throw the plane vertical as high as possible and let it decend through gliding. The second is to launch it from a high place. The second method is better for me because I don't think I could get my throw consistently high every time. But both methods require a large space such as the basketball court in the sport centre. This availability of space could be a potential constraint for my experiement. This means I may end up throwing it horizontally allowed by my height in a large classroom, or in a long corridor (such as the corridor in Claycroft).

2) The flight result of two identical planes are surprisingly similar

This is a result I wanted but I didn't expect the flight result of two identifical planes to be so similar to each other. Initially I thought the variation in my throw could have significant impact on the result, but to my surprise, when I perform the same throw on two planes with identical design (with a little practice to get consistent throw), they both landed at the exact same location! This is good news because it suggests with a careful attention to my throw, I could actually eliminate or reduce the influence of my throw on the flight result. This means the changes in flight result will likely be attributed to the design changes.

3) Somtimes it hard to determine whether improved flight is due to deliberate change in design or unknown factor

This is something I observed when I tried to throw a randomly folded plane (still maintain the basic design but has its wing tip folded down for no particular reason), I noticed the plane actually 'glide' quite nicely (slow descent)  but in an upside down manner. But when I fold another plane properly with a much larger wing surface, it didn't glide but rather flies like a dart (very fast, no gliding) and descend very quickly. This suggest to me that for the first plane it wasnt the wing that was producing lift but when it turned upside down, the whole plane upside down was producing lift (confusing).

That's for now, I will think more about what control factors I will use for the Taguchi experiment to influence flight performance. I also need to think about the location where I perform the experiment. Initially I thought about doing this outside, but now I think the wind will have too much influence which could make my experiment invalid. So indoor is probably better option for now. But as mentioned before, there are also multiple options for indoor location.

: 06 Mar 2009 22:56 |  Tags: Mbe Piuss |  Comments (0) |  Close comments |  Report a problem

principles of flight

Principles of flight
This is a simplisitic summary of how I understand paper airplane aerodynamics

(Fig1)
How does plane fly?
Plane flys because when it flys lift is created on its wings.

How is the lift created?
This lift is created because when the air pressure is smaller above the wing and greater underneath the wing.

Why is airpressure different above and under the wing?
This happens because when wings travel through air, the air travels faster above the wing and slower underneath the wing.

Why air travels faster above wing?

(Fig 2a)
The ways airfoil is designed (curvature on the top) means the distance to travel from the front end to the rear end is greater on the top surface than on the bottom surface (more distance to travel at the top, less distance underneath). When to bottom air reaches the rear end of the airfoil, the top end is still 3/4 of the way (see fig 2a), this creates a vacuum of air (vortex) at the rear end that pulls the top air to make it go faster (like how a vacuum cleaner pulls air) to meet the bottom air at the rear. (Note as of 16:40 06/03 a mistake in the description about the 'distance difference as a cause of lift generation' is recongised and will be corrected in a future entry)

How is air pressure difference related to lift?
Consider this; a motionless object is simply in a state where all forces applied to is equal in all directions. Pressure difference creates unequal balance in force. As the pressure above the wing lessens, the downward 'push' by gravity and weight of atomphsphere is overcomed by the upward 'push' of a denser under-wing air (like a hydrogen balloon, air inside ballon is 'lighter' (less pressure) than the 'heavier' air outside the ballon (higher pressure), therefore the plane rises up.

: 06 Mar 2009 09:42 |  Tags: Mbe Piuss |  Comments (2) |  Close comments |  Report a problem

Taguchi for PIUSS PMA

I read about Taguchi experimental design today for PIUSS PMA. It is about designing and conducting experiments to optimise the design of paper aeroplanes. Essentially this entails (lila's favorite word) repeating exactly what we did during PIUSS for paper helicopters except this time I need to do it on my own.

The main reason why this question appeal to me better than other questions is that it is different from all the pma's we've done before. This is more hands on, practical which could potentially be a lot of fun. And since I will be doing all of the work, I have a real sense of ownership for whatever outcome that comes out of this exercise. Well, I know this is only paper aeroplanes and not designing , for example, a new jet engine, but wouldn't it be nice to have a pma that we will enjoy doing?

Well, if you enjoy doing experiment you would probably learn a lot more. I have a slight advantage in this respect since I did an undergraduate degree in science which invovled a lot of experimentations. So I am quite familiar with experiments and the subsequent report writing (having doing hundreds of them in my undergraduate !). But honestly speaking, experiment is fun if we do it once in a while, it becomes real pain when it becomes like a job, it's really hard to get used to the repetition, tedious procedures, and most frustrating of all, crap results!

Going through Graeme's notes on Taguchi definitely made things much clearer now, espcially the calculations and why they are done in a certain way. The content isn't too difficult to follow, only about 37 pages and by the end you will get a good gist of how to do Taguchi.

Just while I am on the topic, I am also considering changing the subject of the experimental study, trying to find something else to improve it's design that is not just simple paper aeroplane. I am a bit short on the ideas for now, hopefully something interesting will come up before I get started on the actual experiments.

Unlike FACS (growl ~~) , I think I am going to enjoy this PMA :)

: 24 Feb 2009 23:33 |  Tags: Piuss |  Comments (5) |  Close comments |  Report a problem

Six Sigma and Success Factor

Today team 3 presented our findings from our research as well as ideas about making change which we elicited from members of the group. We thought we captured the idea that Graeme had us to do but somehow we still came out a bit confused about the project/initiative level change effort. The only team that did capture that differentiation was Apinya, Lila, and Mennu's presentation.

When Graeme popped the question about this fine difference to me and Luis, I was a bit taken aback by it. I never thought about change on an project or a personal level. I guessed I have assumed that change takes place in the higher level through the vision, strategy and personal influence of the leaders. But come to think of it, changing the attitudes and behaviors of individuals may require a different approach from one that is directed at  a large group of people.And that seem to need a deeper understanding of the change curve of every individual. As different people may be going through different stages of change (i.e. denial, acceptance, realisation), what is required to change people may differ from person to person. And that would mean different style of leadership should be directed in different situations. What would be interesting is some of the techniques, or motivational tools that are useful for this purpose.

This idea of of people on various stage of change nicely linked with what Graeme suggested about the four dimensions of followers. They were stars, cynical terriost, fan , walking dead. Stars are obviously most desirable. Whereas cynical terrorists are mos difficult to deal with, but they should be respected. But Fans and walking dead I am thinking are probablt not much good if what they lack of interest. Usually I'd rather work with people who may not disagree but still show the passion to care about a particular subject.

A few people including meself, mannu, and alan were a bit sceptical about what Graeme called "appreciative enquiry" which is you tell people to improve but do it in a very non-direct way. By first telling them about their strength (what they are doing right) followed by what they are not doing so well (weaknesses or areas of improvement). Hmm.. what should I say about this. It's a nice idea and probably the more desirable and more easily accepted way. But it really is hard to do in practice. Mainly I think everybody wants to hear other people talk god things about themselves (I would be the first to admit!). But it takes a truly reflective person to see his own weaknesses and for the others it would be difficult when it's not pointed out directly. Well, I guess this just vary from person to person.

: 24 Jan 2009 04:34 |  Tags: Mbe Piuss |  Comments (0) |  Close comments |  Report a problem