All entries for December 2015
December 28, 2015
These are some of the things are learned during the module:
- Any successful improvement implementation requires alignment with the company strategy, senior management support, people involvement and capital investment on the change.
- Lean and Six Sigma are more than just a set of tools, they are ways to pursuing operational excellence and demand a change in the ways problems are approached and things are done in a company.
- It is much better to treat improvement methodologies as ways of changing the organisational culture (long term view) than just as localised problem solving tools.
- People is the key for sustaining (maintaining over time) the improvements a company makes and for creating a CI culture.
- Customer satisfaction goes beyond giving the customers what they tolerate, compliance with the client tolerances is not enough.
- Even though DMAIC has its origins in Six Sigma, it can be applied as a logic for process improvement and problem solving since it provides a structured framework to follow.
- Don’t pretend to solve a problem without first understanding it, learn and measure the current process before attempting to change it.
- Waste, either defined as non-value added activities or variation in a process, need to be reduced or eliminated to have more efficient operations and better quality in products.
- SIPOC (Supplier-Input-Process-Output-Customer) helps me understand which the contributors of a process are.
- First get the process stable and then improve it.
- It is always important to test your theories, if you think you have a solution, test it before implementing it.
- Lean and Six Sigma, instead of being rivals, are methodologies that can complement each other.
- If the strategy of your company is innovation (e.g. apple) maybe six sigma is not right for you.
- The idea of Six Sigma is process on target with minimum variation. Improving on the basis of that concept gives at the end better process capability.
- Process improvement is not difficult, finding a solution is not a big deal, what is difficult is the implementation and sustainability of the change associated to that improvement.
- The belt hierarchy of Six Sigma create an elite of people that can perturb the involvement and engagement of people from lower layers of the organisation structure.
- Taguchi loss function teaches us that the more a process/product moves away from the target (its best), the more money the company loses.
- Averages by themselves don’t mean anything, you need to see the spread (variation) of your data and the control limits to see if the process is under control or not.
- The PDCA starts with study not with plan.
More than just a set of tools, Lean, Six Sigma or the combination of both are ways to create cultural change in the organisations that implement them, they represent an opportunity for change in the way everyone at the organisation think and behave, and also an opportunity to be better every day and pursuit operational excellence.
Before I started this module, I thought I like Lean but I don’t like Six Sigma. I had the idea that Six Sigma was just a big list of tools that required a lot of statistical understanding to make sure companies had great quality, I thought it was a matter of quality engineers and not of process engineers like me. Due to my bachelor’s in Industrial Engineering and my job in process improvement areas I have an inclination for speed and flow (Lean Thinking); however, this module made me see the whole picture.
The thing is, the two are approaches are not rivals, they can complement each other. The whole module was about Lean Six Sigma and I realised it until the end, we learned in class and during the PMI course about people involvement, flow charts, one piece flow, streamlining, process capability, voice of the customer, statistical process control, voice of the process, etc., and all of that is a combination of Lean and Six Sigma.
Now, I understand that the gap between process engineers and quality engineers is really reduced, and both of us work for the same goal, to improve processes, have better quality and reduce costs. Now I have a clear view that there is no need for choosing one approach over the other, I can have lean and six sigma, I can understand flow and statistical control, I can have belts and normal operators working together for the same cause.
December 23, 2015
I have been asking myself how all this Six Sigma (SS) module fits with the EFQM idea that to get results you need to work on the enablers. My first answer was obvious, with six sigma you improve processes by reducing variation, so the use of SS directly affects the proccess enabler of the model.
Then I thought we also learned that a good implementation of SS will required people engagement, project alignment with the strategy and good leadership that supports change and empowers people, that shows that SS can touch other EFQM enablers (Leadership, People and Strategy), but what about resources?, well any change initiative requieres resources, people, money, etc. in the case of SS you need to invest resources to increace your process capability and get improve your performance.
Finally, I thought if SS can be part of all of the enablers (directly or indirectly), it can be logical to think that SS will also be part of the results, then I asked myself what results I could get with a good* implementation of SS?, Financial results, client requierement compliance, employee engagement, better product value, less complaints, more training for employees, better process capability. In other words, SS touches 3 of the results of EFQM (Performance, People and Customer), pitifully it does not touches the social result.
I concluded saying SS is not perfect, it can be improved by mixing it with SOPK, Lean or other quality and operations improvement system; however, it can be classified as a way to drive Business Excellence in an organisation, since have the power to affect enablers and results of the EFQM model.
*Good implementation of SS includes avoiding the problems related with the hierarchy of belts.
December 21, 2015
The day we constructed the paper helicopter was long and enjoyable at the same time. I learned that day that we want to create robust products and processes, in other words, we want our products and processes to work consistently (with quality) no matter the noise (factors that affect the product or process). I learned that quality can come from the robust design of process/products and you can test that design by doing experiments; however, since in real life making experiments can be really expensive, we should do minimum experimentation and try to pursue quality in the most economical way (a few experiments, a few levels of control factors).
During the experiment we had to choose our control factors, construct the orthogonal matrix with the eight possibilities of helicopter and construct the helicopters. Something we discuss at my group at some point was that we needed to hold our horses and stick with the factors and levels we had decided before, even if we knew some of the combinations weren’t good. Sticking to the Taguchi process at the end took us to the best combination of factors, messing around with our factors and levels before following the procedure would had made everything messy. So first lesson of that day, follow the procedure.
We took the time each helicopter could fly adding noise (wings up and wings down) and calculated the S/N ratio (which helicopter performed in a more consistent way now that the noise was included), we analysed how each factor affected the result we were getting, we identified the best combination of factors and we calculated how predictable it was to get a performance like the one we got in the experiment again.
I liked the Taguchi method because by applying it, I could understand how to get the best combination of factors that will generate a consistent product performance and how some factors affect a performance more than others. Even though I don’t know much about statistics, It is always good to see that there are structured ways to experiment the combination of variables and get an answer of which is the best.
December 01, 2015
A problem I have seen during my working life is that many great process improvements don't stay in time, even though they are logical and structured, people might come back to the old ways of doing things if they are not convinced the change is beneficial or if they don't understand the reason behind it.
The thing is managing change with people is a huge challenge, the way people feel about the change determine their behaviour towards it which at the end could be translated into a sucessful implementation of an improvement or a failed one.
I think people should be considered since the first step of the DMAIC cycle, they should be included in all improvement projects, they should understand why things are being meassure, why things need to change and they need to be able to be part of the solution.
However, sometimes managers and process engineers consider employees as a problem when they reject change, when resistance to change is something natural in human beings. Maybe all we need (as process engineers) is to understand a bit of psychology and change management to make change last, maybe all we need is include our people, maybe that is all.
There is this tendency in us, as human beings, of having solutions for everything, our brain is always prepared to solve the problem no matter what it is. However, a problem can not be completely or properly solved unless we understand it, unless we are sure about the characteristics of our current situation.
I am an Industrial Engineer and in two of my modules back in my bachelor, I learnt that it is important to learn the process, diagnose it, propose an improvement, test that improvement, implement it (if everything in the test came as you thought in the test) and control it, it was common sense to me; nontheless the tendency is other, many of the people that I have worked for, pretent to ask me for solutions just for the sake of me working in the process department, I always have to explain I can not help in the solution if I dont understand what they are doing.
While studying this module PIUSS, the DMAIC cycle came as a remainder that you can not pretent to introduce solutions to a problem you still dont know, because that will just make things worse since you dont know your starting poin...how are you going to see an improvement if you dont know which is your gap? It is illogical.
This DMAIC cycle is a logical way to carry on with process improvement and I think it was what we were trying to do on the team work at class. We undestood the process, we measured it (time, defects, inventory), analysed it (made a list what were our problems and why they might be happening), improved it by proposing solutions, testing them and implementing them and I guess we will control it some other day.
I like process improvement because it is about common sense, I am enjoying PIUSS and I will try to learn as much as possible :).