20 replies to this topic

[Simon]

I am interested to hear views on the different methods people use to solve work related quality problems. When I say problem solving, I am talking about recognised structured problem solving approaches such as PDCA (Deming), 8D (Automotive), or DMAIC (six sigma). On the other hand, perhaps you use a different method entirely.

What do you use, how do you use it and how well does it work for you?

Regards,
Simon

[Penard]

Hi Simon,

I'm sorry, no special method, only experience and 'pragmatic method'! Sometimes you have no delay to take a decision in a factory, so you must take the best one to comply with urgent requirements.

I remember when I was a student, teachers said 'you won't have to be the fireman, but you have to forecast issues'. To me, the best way to anticipate problems is to have a good knowledge of your process, product and factory, GMP's and Haccp plan -a good quality system

Last, it depends on you too. Very important not to show somebody you don't know how to react, but have to prove quickly you're effective - thanks to your good knowledge! Because you have on the one hand quality systems, and on the other hand your colleagues - human relationship are the most important to solve problems.

Regards,

Emmanuel.

[Jarve]

Despite my qualifications I have never thought "I'll use this method to solve this problem", I always gather as much information as possible before making an informed decision, the key I think is being able to split problems into component parts and then clearly explain to those involved what they need to do to help solve the problem. Communication skills are a key part here and never let anyone see that you're struggling or flustered, stay calm and cool, think clearly and don't let people who are panicking destract you from the "real" problems by confusing you with unneccessary detail. e.g. Question would be "where is the affected product?" answers you're likely to get are "well "he" put some here and I think "he" put some there, and "she" was looking in to it" - to resolve this kind of issue, make one person clearly responsible for locating all affected product.

And of course common sense must always prevail.

[Simon]

Hi guys,

By using the word “problem” I think I’ve steered this discussion in the wrong direction. Of course managing a problem [/i]successfully[/i] requires a professional and thorough investigation and the implementation of effective corrective action.

However, I was thinking more of proactive quality or efficiency improvements, so really I should have said what methods are members using to identify and expedite process improvement projects.

So for example in DMAIC methodology.

Define - would include clear description of the problem to be addressed, benefits, start date, planned end date, team members, supporting data.

Measure – would include current process description, current performance, data collection.

Analyse – what do we know about process input variables, how do route causes affect process output, what does data say.
Improve – what are possible solutions, benefits / costs / risks, how will solutions be implemented.

Control – post improvement measurement, standardisation of changes, review of project – did it meet goals / benefits.

Various quality tools are used at each stage of the process.

The above is the six sigma DMAIC approach. Does anyone use such a structured plan for improvement projects, or maybe something completely different?

Regards,
Simon

[Penard]

Agree with Jarve, I don't use this kind of method - but maybe my previous factories were too little...They don't have time enough to set it - I tried at the beginning of my career, but I gave up because my bosses didn't consider it as an interesting way of working. Little factories have to respect a continuous improvement because of competition, even if they don't really know what 'continuous improvement' precisely means. There are a lot of informal methods, maybe as many as factories... Usually big fims, factories and consultancies used it to my mind!

[okido]

Hi Simon,

In the past I used PDCA. However it is my experience the terms PLAN, DO, CHECK, ACT are reason for much confusion.
I now use DMAIC for larger improvement projects.
For day to day operetions I have small list with questions that gives structure to the problem soving process.
Questions like: What is the problem, for who is it a problem, how large is the problem, who is involved by solving the problem, when did the problem start, who does what, how much do we want to improve, can we use the solution elsewere, etc.
My experience is that using this list irritates people every now and than, in general we like jump to solutions too easy.

Have a nice day, Okido

[Simon]

Hi Simon,

In the past I used PDCA. However it is my experience the terms PLAN, DO, CHECK, ACT are reason for much confusion.
I now use DMAIC for larger improvement projects.
For day to day operetions I have small list with questions that gives structure to the problem soving process.
Questions like: What is the problem, for who is it a problem, how large is the problem, who is involved by solving the problem, when did the problem start, who does what, how much do we want to improve, can we use the solution elsewere, etc.
My experience is that using this list irritates people every now and than, in general we like jump to solutions too easy.

Have a nice day, Okido

Thanks for sharing Okido. Human beings are riddled with assumptions and effective problem solving can be taken as the elimination of these assumptions, one-by-one-by-one. Herein lies the problem. It is a painstaking process. And as you say people prefer to jump to quick solutions. Then they wonder why it didn't work and file the problem in the 'too hard to solve' cabinet.

Thanks,
Simon

[Justin]

I have discovered that problem solving is circular. By that I mean that the problem you thought that you solved should be revisited on a regular basis. This certainly leads to continuous improvement but with the advantage that all the leg work has mostly been done. With DMAIC, to me, it seems as though it is unidirectional and once the project is handed over to the process owner, it is not revisited unless something drastic happens.

[Simon]

I have discovered that problem solving is circular. By that I mean that the problem you thought that you solved should be revisited on a regular basis. This certainly leads to continuous improvement but with the advantage that all the leg work has mostly been done. With DMAIC, to me, it seems as though it is unidirectional and once the project is handed over to the process owner, it is not revisited unless something drastic happens.

It's easy to forget to go back and check the solution is in place and is still being effective. Good point Justin.

Regards,
Simon

[Dr Vu]

Hi All

On principle i use DMAIC.. but modified with an Interim containment step (8d) after the define step.
But in reality it depends on the depth of the problem i am faced iwith at the time. Some problems are solved much easier and require a modification from the above but the bottom line is for all the problem solving i do- i always have an underlying PDSA principle injected into it to ensure its a corrective and preventative step in one.

Thanks
Vu

[Simon]

Hi All

On principle i use DMAIC.. but modified with an Interim containment step (8d) after the define step.
But in reality it depends on the depth of the problem i am faced iwith at the time. Some problems are solved much easier and require a modification from the above but the bottom line is for all the problem solving i do- i always have an underlying PDSA principle injected into it to ensure its a corrective and preventative step in one.

Thanks
Vu

Thanks for recycling this oldie. For those who like to keep it simple I find "WHY?" is a very good question to ask when trying to get to the root cause of a problem.

Any new ideas?

Regards,
Simon

[Simon]

Hi All

On principle i use DMAIC.. but modified with an Interim containment step (8d) after the define step.
But in reality it depends on the depth of the problem i am faced iwith at the time. Some problems are solved much easier and require a modification from the above but the bottom line is for all the problem solving i do- i always have an underlying PDSA principle injected into it to ensure its a corrective and preventative step in one.

Thanks
Vu

Thanks for recycling this oldie. For those who like to keep it simple I find "WHY?" is a very good question to ask when trying to get to the root cause of a problem.

Any new ideas?

Regards,
Simon

[GMO]

I'm a newbie to most 'lean' techniques which a lot of these are; however, I learned a bit about Pareto some years ago and as a simple idea, it does focus your mind.

Say you have 100 products and your overall micro is too high. Pareto says; put them in order of micro failures and then concentrate on the top 20 because they will be causing the most of your problems (80% is the number often quoted but whatever it is, it's high.) I did this on one site and found that the top 20 shared one process - fixed that over the course of about 2 weeks and dropped our micro OOS from about 3.5% to <1%.

Often people think that 5 minute assessment is a waste of time but it's not added time, it's actually making sure you don't waste time later.

I've never used the other methods mentioned but I'm learning!

[Simon]

Often people think that 5 minute assessment is a waste of time but it's not added time, it's actually making sure you don't waste time later.

I've never heard of this GMO, is it simply what it says on the tin?

We're all learning.

[Jean]

I commonly use the RCA technique. This helps to find out many probable or hidden causes of a problem than the real cause. The brainstorming sessions are interesting as you get to know many factors and each factor is looked into individually, in a way helps to learn more from the other experts and helps to dig deeper into a problem and ultimately identify the solutions. But one has to finally analyze the significant causes and find the practical solutions to it.

[GMO]

I've never heard of this GMO, is it simply what it says on the tin?

We're all learning.

I just meant Pareto takes 5 mins to do as with many of these 'lean' techniques!

A friend of mine explained what SMED was the other day (a senior colleague is fond of saying "we'll 'SMED' this!") Anyway, that simply means "Single Minute Exchange Die" and it comes from a car manufacturer realising that their die changeovers took hours so they had to carry loads of stock of each part which is one of the "five wastes" (another 'lean' principle and one part of it is overproduction so basically having something in a store is costing you money in materials, space etc.) They tried to reduce their changeover time and got it down to 1 minute by changing the layout of the area.

Now think about that in a cleaning schedule. If you reduce changeover times, you potentially have more time to clean and also if you reduce strip down times, you also have more time to clean. These are very powerful tools which aren't difficult to understand and generally give results which QA / Technical people would like! After all, another of the 5 wastes is making faulty items which need rework. I think in the food industry though we have quite closed minds about all this and we need to open them; after all, car manufacturing have been doing this for 40 years or more!

[Charles.C]

Dear GMO,

Not sure if this is OT or not (querying the validity of stat. based options.)

car manufacturing have been doing this for 40 years or more!

I certainly agree one can only be awestruck at the minute details concocted within their ISO, etc standards. Is this really a desirable way to go for food ?

I also suspect that these techniques are only feasible since they are not dealing with natural products and the unpredictable intrinsic variations in many of the quality factors. I once spent 6 months trying to set up bacteriological control charts for receival of raw seafood from several suppliers but eventually gave up. Mother Nature refused to provide a situation of statistical control and the effort of doing ANOVA type analyses was prohibitive. Maybe software tricks now available might hv improved things but I rather doubt it.

I guess cleaning should be a more controlled process (via surface mic. counts for example) but seems to me not comparable to, say, weight control where statistical routines are common in the process environment. ?? Maybe you have seen actual factory examples ?

Rgds / Charles.C

[Simon]

I just meant Pareto takes 5 mins to do as with many of these 'lean' techniques!

A friend of mine explained what SMED was the other day (a senior colleague is fond of saying "we'll 'SMED' this!") Anyway, that simply means "Single Minute Exchange Die" and it comes from a car manufacturer realising that their die changeovers took hours so they had to carry loads of stock of each part which is one of the "five wastes" (another 'lean' principle and one part of it is overproduction so basically having something in a store is costing you money in materials, space etc.) They tried to reduce their changeover time and got it down to 1 minute by changing the layout of the area.

I'm impressed with your knowledge GMO. I know what you mean about the kind of process improvement snobs who just love to talk in acronyms, they get right up my nose. Another point on SMED is making things easier to change fo example less parts, standardisation, nuts that release after one turn etc. Also why not practice SMED on the cleaning schedule itself - more time for production.

Regards,
Simon

[GMO]

Another point on SMED is making things easier to change fo example less parts, standardisation, nuts that release after one turn etc. Also why not practice SMED on the cleaning schedule itself - more time for production.

Regards,
Simon

Very good point! I will suggest it to the team!

[GMO]

Yes. All of these things can be applied in a factory packing natural products. I've seen 5S implemented in a food factory for example and that doesn't matter what you make.

Despite the nature of the products we make, the food industry is becoming increasingly mechanised - who would have thought that years ago? The more mechanisation you have, the more important it is to ensure your changeovers and cleaning are the best they can be because automatically more machinery normally means fewer people to clean! Even in non mechanised processes there are opportunities to change your process and improve productivity. Then if you think about the products; yes there is variation but normally within a statistical range. I think everything is controllable to some level and then we'd make the food industry really world class rather than one of the biggest manufacturing industries because it's harder than some other businesses to export (note; it's harder but not impossible, UK Smarties are now made in Hamburg, Polos for Australia are made in Malaysia...)

I have just been in so many factories where a change has been proposed but then quashed because of some slight difficulty along the way. The real problem was that there wasn't the belief in the first place! Give it a go. You never know it might change the world!

[Simon]

In light of the current economic climate the need for process improvement and cost reduction become even more important for businesses. Managers who can use improvement tools and techniques effectively will have a great opportunity in the coming years.

Regards,
Simon