Interview of Jeffery Younginer, Nissan North America
Uncategorized — By Gladys on 6 January 2009 at 5:42 am
As you know, Measure Control is always looking for new points of view and perspectives to present you through our interviews. For the interview you will read today, we even crossed the Atlantic to find the right person to interview! Jeffery Younginer is the Director for Nissan Production Way and Manufacturing Support for Nissan North America, and he very kindly accepted to answer some questions. On behalf of all the Measure Control team, I would like to thank him very much for his motivation in giving us complete and interesting answers for this interview.
Measure Control: One of the most important steps within the 6-Sigma process is the measurement. Should this step be neglected, the whole 6-sigma process could end up being a failure. Do you agree with this affirmation?
Jeff Younginer: Yes. You have to know where you are in the current condition and know where you want to be in order to make progress. At the end of the activity you should measure your progress to the goal. For parts, you have to have some confirmation that your process is in control.
Which tools do you regard as most effective to carry out measurements during the mass-production of automotive parts – eg: 3D machines, checking gauges… – ?
Jeff Younginer: It really depends on the process that you need to control and how critical/precise it is to ensure control (based on scrap cost, detectability, etc). In some cases random sampling to a fixture or human inspection is sufficient. In other cases I have seen lasers and ultrasonic measurement being used to ensure body panels are in the correct location and that the welds are adequate. It boils down to the rigidity of your process vs. the risk/cost of repair as to how far you go with automated measurement.
Measure Control: Do you regard checking fixtures as indispensable for the quality of your parts or the parts produced by your providers?
Jeff Younginer: Yes. You must have some way to check/confirm your parts. Whether it is a fixture or some other method depends on the situation.
Measure Control: Which technologies could replace checking fixtures in a close future? Could computer vision cameras be a substitute on a short term?
Jeff Younginer: At Nissan we use several methods to check parts/part locations; check fixtures, vision systems and laser scanners. They all have their pros and cons. Fixtures are the easiest/quickest to use but they are very expensive and not very flexible. If you make a change to the part, it is very costly to change the fixture. It also gets very expensive to move fixtures from one location to another (if the part moves locations). Vision systems take care of some basic verification but they have their limitations on how detailed they can be. Laser scanners are expensive on the front in and require some expertise but the flexibility of the scanners gives them longevity and lower operating costs for the future. Another big advantage with laser systems is that we can look at multiple parts and their interfaces.
Measure Control: What are the most important criteria for the choice of a provider of checking tools? – eg: proximity, short deadlines, prices, quality –
Jeff Younginer:
1. Quality/reliability of the measurement tool (it needs to accurately measure what I need it to)
2. Cost is going to play an important role. Fixtures are easy to train to use and can be used quickly in line but they are expensive to maintain (changes and relocating). Lasers take longer to set up and you have to develop/maintain some expertise but for the most part, you have a one time sunk cost with long term flexibility of use.
3. Ease of use ( I don’t want to hire a crew of PHDs to use/maintain it)
Measure Control: Which method do you use to collect and analyse the measurement data gotten from your measuring devices? (Eg:an ERP to analyse the Cp/Cpk studies for a series of parts).
Jeff Younginer: We have several softwares, depending on the type of data that need to be analysed:
- We use a software called Minitab for general statistical computing
- In the layout area where we sample vehicle/part fits, we use a software called MSRD.
- For some of our inline measurement we use software from Fanuc and CM4D
- For some of the inline flush/gap measurements we use software called AccurX to collect the measurements
- And finally we have also developed some excel based programs to compute CP/CPK information
Leave a comment
RSS for comments on this article
8 Comments
Dear Jeffery,
Once again, thank you for your answers. Antonio Onteniente put a comment in the Spanish section: he congratulates you for this article, and would like to know if you have an idea of how much your plant was able to save thanks to the implementation of the 6 Sigma process. Could you give us an estimation please?
Thanks Antonio. Sorry for the delay in the response I have been out of the office for the past week. Nissan has it’s own 6 sigma type process called V-up (stands for Value Up). Since it’s inception 8 years ago, we have a documented net savings of over $89 million with just under 1300 projects closed out.
Jeffery, if it is not confidential, could you tell us which improvements does the V-Up process bring in comparison with the standard 6-Sigma DMAIC process? Do you have in Nissan the basic organization of Belts and Champions? It would be great if you could give us some information about a particularly successful case that you have had over the past 8 years. Thank you in advance.
Nissan’s V-up methodology consists of 3 major components. IDEA, DECIDE and V-Fasts
IDEA stands for:
Identify – the issue
Define – the scope, possible resolutions, paths to study
Estimate – the effects of resolving the issue
Accept – the challenge to work on the issue
The premise behind IDEA is to take major issues and break them down into bite size, workable projects for either DECIDE, V-FAST or Line management assignments. Usually facilitated by a V-expert.
DECIDE stands for:
DEC – Define, estimate create a team
I1 – Measure (current state)
I2 – Analyze
I3 – Improve
DE – Deploy and Evaluate
In comparison to DMAIC which stands for:
D – Define
M – Measure
A – Analyze
I – Improve
C – Control
DECIDEs are larger scoped projects usually taking 2-6 months to resolve. DECIDEs are facilitated/conducted by a V-pilot.
V-FAST are smaller scoped, quicker projects usually taking between 1-2 weeks to resolve but can take longer in some circumstances. They are facilitated by a V-fast facilitator.
Phase 1 – Define the issue and estimate the effects
Phase 2 – Compose a team to work on the issue
Phase 3 – Intensive discussion
Phase 4 – Implement the countermeasure/resolution
Phase 5 – Evaluate the effects – identify future actions
In comparing the levels of instructors, we utilize 3 levels; V-fast facilitators, V-pilots and V-experts.
• V-Expert (previously a V-pilot – 3 weeks of training multiple DECIDE project experience) ~ equivalent to a Master Black Belt
• V-Pilot (3 weeks of training and required to complete a minimum of one DECIDE project) ~ equivalent to a Black Belt
• V-Fast Facilitator (3 days of training) ~ equivalent to a Green Belt
As far as projects go, in the last 8 years (since inception of the program) I would estimate we have completed approximately 90 DECIDE projects and 1200 V-Fast activities. The topics range from implementing new processes or procedures to improving existing ones. I can’t share specific projects and their results – that’s one of the reasons that Nissan practices V-up (our own design of Six Sigma) verses Six Sigma itself. One of the requirements of Six Sigma is for them to validate a completed project to become certified to a certain level (Master Black Belt, Black Belt, etc.) which means they then have knowledge about possibly proprietary information which Nissan is not willing to do.
Hope this answers your questions…
Jeff, with such a detailed answer everything is perfectly clear. Thank you very much for these excellent results.
You are very welcome.
Hi Jeffery,
Your interview helps me to understand better the real problems met in production. When you compare the different checking technologies, you mention the limitations of the vision systems regarding the possibility to provide detailed information. I would like you to explain which details you are missing in the information these systems provide.
As I am a designer and integrator of artificial vision systems, this would be a valuable piece of information for me: indeed, I often dedicate myself to extend the possibilities of basic vision systems, by combining them with modern computer systems.
My exposure to vision systems has been of the simple variety. In my previous life with Michelin Tire (back in the early 90s) we were experimenting with vision systems to help us measure product lay positions and joint overlaps. We had issues with lighting, glare coming from the rubbers due to sheen, and just overall repeatability of the results so we abandoned what we were working on. Granted, that was a decade ago (a century in IT systems timing) so I would throw that recollection out the window now. Currently we have some vision systems here at Nissan but they are simple ones. All they do is confirm either a 2D barcode or confirm that an ID character on a part is the character that it is supposed to be installed in a downstream process. Again, I am sure there are more complex/capable vision systems that could do much more but my experience is limited to the lower end (cost)versions that simply give us a go/no go on what we are checking.
Hope this helps…
Jeff