I went a bit more in depth about common and special causes. These apply to more than just problems, per se. Almost any data can be more correctly viewed in this context and more information gained.
Gaining insight into the nature of variation, can greatly help to avoid “knee jerk” decisions. This may increase the number of positive outcomes and reduce chaos in the system. Looking at important figures in this manner can also help spot trends, judge the effects of change and greatly increase a manager’s ability to predict future occurrences.
The last post mentioned the more prevalent error of treating common cause as a special cause. This can result in over reaction, asking for explanations where none exist and vastly over-complicating the system. An example is making a rule for everything that happens. Soon there are so many rules that no one can remember, let alone follow them. This has the effect of stifling the system. More on this in a later post.
The second mistake is to treat a special cause as a common cause. That is to ignore the signal and attribute everything to the system. This results in missed opportunity for improvement. It may also cause the manager to fail to make needed adjustments.
An example might be Joe’s production figures from the previous post. Let’s say in the subsequent 9 weeks Joe produces 47, 43, 45, 44, 45, 44, 43, 45 and 44 hours respectively. Plotted on our chart, with the prior figures, it would look like this.
The mean is now calculated at 43. The square root of 43 is about 6.56. Taking three times [standard deviation] this amount, we get 19.68. Adding 19.68 to the mean gives an upper control limit of 62.68 and a lower limit of 23.32. This suggest, unless something changes, Joe will continue to produce weekly amounts between these two limits.
The chart indicates all points fall within the limits, but it also tells much more. The last nine consecutive points do not cross the mean. In statistics this is known as a trend or change in process. It is not statistically probable that this could occur by chance.
This is an opportunity to examine what has changed and see if the change [in this case favorable] can be further implemented. It may be that Joe is using a different method, can his method be taught to others? Perhaps he has chosen to work harder. If this is the case, appropriate recognition may help make the change more permanent. Failure to take notice may make his effort seem unwarranted, in his mind. Realizing something has changed at a later date may make finding the source of the change far more difficult.
The tool of statistical process control (SPC) can also be used to help judge the effects of new tooling added to the shop. It could also show the effects of improved scheduling, marketing programs and almost anything else. There is a huge amount more a manager could know about the topic of statistical process control (SPC) and this only scratches the surface. There is far more information in the suggested reading list, included in the ARTICLES section of this site.
Learning about SPC helped me a great deal as a manager. By using simple methods I was able to start making decisions based on data, rather than emotion and suspicion. It also provides a more factual means of confirming a theoretical change. There are several other simple tools that may be equally helpful. I hope to discuss these more in future post.
Joined: 15 May 2007 Posts: 30 Location: Bluefield, WV
Posted: Tue May 29, 2007 7:16 pm Post subject:
Quote:
The last post mentioned the more prevalent error of treating common cause as a special cause. This can result in over reaction, asking for explanations where none exist and vastly over-complicating the system. An example is making a rule for everything that happens. Soon there are so many rules that no one can remember, let alone follow them. This has the effect of stifling the system. More on this in a later post.
This is my last job, to a T. Everything that would happen had to be explained and would get a new rule. One time an air conditioner job came back because the week after I changed the compressor the blower motor went out. In the Monday morning meeting he had me stand up in front of everyone and explain why this happened. Joking I said maybe we should change the blower motors on every big AC job. He actually thought it was a good idea. I quit the next week.
Joined: 19 May 2007 Posts: 206 Location: Camp Verde, AZ
Posted: Tue May 29, 2007 10:45 pm Post subject: Re: Solving Problems or Why Bolts Get Left Loose III
louis wrote:
In the last post on this topic,
Learning about SPC helped me a great deal as a manager. By using simple methods I was able to start making decisions based on data, rather than emotion and suspicion. It also provides a more factual means of confirming a theoretical change. There are several other simple tools that may be equally helpful. I hope to discuss these more in future post.
Louis,
This is somewhat new to me. I have a very basic understanding of what you are talking about. I can get into Excel and probably make a chart. What things are important to look at? As a novice, what things should I be looking first, second, third and so on, Total billable hours per week, Gross Profit dollars per week, car count, average RO or what???
You are giving out some theory. How do we put it to use? As this discussion goes on I would like to have some of our actual data to look at, and learn from.
Facts are the best thing to base decisions on. This looks like a good tool to help make factual choices with.
Joined: 16 May 2007 Posts: 7 Location: Gonzales, LA
Posted: Wed May 30, 2007 8:20 am Post subject:
Hi Louis,
This is very interesting for me. Looking at the chart I immediately notice the two low weeks and would have considered them the problem. You are saying they are not an issue but the last figures are? I would like to know more about this, if you don't mind.
Joined: 15 May 2007 Posts: 774 Location: Baton Rouge, LA
Posted: Wed May 30, 2007 9:26 am Post subject:
Eric J. wrote:
Quote:
The last post mentioned the more prevalent error of treating common cause as a special cause. This can result in over reaction, asking for explanations where none exist and vastly over-complicating the system. An example is making a rule for everything that happens. Soon there are so many rules that no one can remember, let alone follow them. This has the effect of stifling the system. More on this in a later post.
This is my last job, to a T. Everything that would happen had to be explained and would get a new rule. One time an air conditioner job came back because the week after I changed the compressor the blower motor went out. In the Monday morning meeting he had me stand up in front of everyone and explain why this happened. Joking I said maybe we should change the blower motors on every big AC job. He actually thought it was a good idea. I quit the next week.
Good post!!
Hi Eric,
With regard to dealing with cause, either mistake will give poor results. There are more common cause problems than special cause, so this mistake is more common.
Your ex-boss treated a common cause as a special cause. The master cylinder comeback was a system problem. There is no feasible method of testing every cylinder or way of knowing when they might fail. Do enough brake services and it will happen. The person doing the service has no part and only delivers the random problem. Asking him to explain what happened is futile.
Changing every cylinder every time, would only serve to maximize the cost to everyone to prevent an infrequent problem. Even doing this might not prevent every occurrence as a defective part might be installed.
To more likely solve the issue an accurate, quick and non-intrusive method of testing might help. This may not be possible with current technology, so the overall cost must be weighed against the risk. Of course more accurate means of inspection, record keeping on different model failure history, etc. can lessen occurrences.
If the worse case scenario can be lived with and the occurrence is low, it may make more sense to not solve this issue, until a better method can be found. In other instances [life or death, catastrophe] or more frequent occurrences, it would make more sense to take another approach, knowing cost will be greatly increased.
Joined: 15 May 2007 Posts: 774 Location: Baton Rouge, LA
Posted: Wed May 30, 2007 10:02 am Post subject: Re: Solving Problems or Why Bolts Get Left Loose III
Hi Dave,
Quote:
This is somewhat new to me. I have a very basic understanding of what you are talking about. I can get into Excel and probably make a chart.
I found hand drawing and calculating charts helpful, until I became familiar with the process. Once familiar, there are a large number of SPC program, and Excel, that can do the math and chart for you.
Quote:
What things are important to look at? As a novice, what things should I be looking first, second, third and so on, Total billable hours per week, Gross Profit dollars per week, car count, average RO or what???
Almost anything that is important and makes sense to you can be plotted. To start pick something you would like to know more about, plot it and see if it helps. Some things may not be worth the effort, other things are quite beneficial. What it can do is greatly increase your odds of making proper decisions. It also give a means to determine common and special causes, and the better method used to deal with them.
In time you may want to familiarize your techs with the techniques. Our guys plot there own production and use it as a guide when trying new methods. The historical data are particularly helpful when evaluating new equipment.
For instance, Jeff used his diagnostic time chart to evaluate the effects of a new scan tool we were considering. The change of process was very evident and we felt comfortable with the purchase.
Quote:
You are giving out some theory. How do we put it to use? As this discussion goes on I would like to have some of our actual data to look at, and learn from.
I will try to use specific applications in future post. Also reading the books in the suggested reading list may help a good deal. I have a friend that is far more knowledgeable about SPC than I am. If there is enough interest, I may be able to get him to write a few post.
Quote:
Facts are the best thing to base decisions on. This looks like a good tool to help make factual choices with.
Thanks,
Dave
SPC can greatly increase the odds of decisions being correct. Nothing will guarantee a perfect decision every time, but this method greatly stacks the odds in your favor. It's another tool managers can use to improve their performance.
Thanks Dave, I appreciate your interest and will try to write more on the topic.
Joined: 15 May 2007 Posts: 774 Location: Baton Rouge, LA
Posted: Wed May 30, 2007 10:26 am Post subject:
Hi Cecile,
Cecile wrote:
Hi Louis,
This is very interesting for me. Looking at the chart I immediately notice the two low weeks and would have considered them the problem. You are saying they are not an issue but the last figures are? I would like to know more about this, if you don't mind.
That is correct, the two low weeks fall well within the range of statistical probability. Any number of things could contribute to the cause. For instance, working on an couple of large jobs and not completing them until the following week or any number of other things.
What is not statistically probable is nine-weeks that do not cross the mean. This indicates a change of process. If we act quickly the reason may be able to be learned. As time passes the reasons are far more difficult to track. We may also miss the opportunity to congratulate someone for putting out a great effort. Perhaps it’s time for a raise?
Many years ago, my oldest daughter brought home her report card. She had worked very hard to get her grades up. She had all A’s and one B. Immediately my eye was drawn to the B and I asked, “What happened here?” What an ignorant mistake, I missed the message and concentrated on the variation. Hopefully those we love forgive our ignorance.
This is somewhat new to me. I have a very basic understanding of what you are talking about. I can get into Excel and probably make a chart. What things are important to look at? As a novice, what things should I be looking first, second, third and so on, Total billable hours per week, Gross Profit dollars per week, car count, average RO or what???
Those are all things anybody can, and will, look at. Those numbers give words, but no story. There is no real, useable information in those numbers. In other words, they numbers can't tell you why any tech had a good or bad week, suffered few or many comebacks, why your new ad made money or not, and many other things a manager needs to know.
Louis is going to teach how to distill variance and trends from the usual mass of data and reports we see. He can demonstrate it in meaningfull, simple methods; well, once you get used to it, that is.
Those variances and trends will show you what to ignore and what to concentrate your time on, and will give you explanations of why things happen in your shop.
Joined: 15 May 2007 Posts: 774 Location: Baton Rouge, LA
Posted: Mon Jun 04, 2007 7:34 pm Post subject: Re: Solving Problems or Why Bolts Get Left Loose III
Hi Dave,
Dave wrote:
As a novice, what things should I be looking first, second, third and so on, Total billable hours per week, Gross Profit dollars per week, car count, average RO or what???
Dave
I apologize, re-reading your post, I am not sure if I answered this part of your question. I do not mean to be vague, but the philosophy I am speaking of is very "all encompassing." It is sometimes difficult to explain one thing, without the context of all the others.
I can say that I do not monitor any of the above mentioned figures. To me they are all historical data and can do little to help me. They are the result of other things, not the cause.
For instance, knowing billable hours, gross profit, car count and average R.O. cannot help me make any more money, in my opinion. These are figures that bankers find useful. They can compare one business to another and decide if they wish to loan money. Management trainers copy these figures from financial people.
What can help me is knowing my true cost. Establishing true cost per hour, gives a figure that will cover overhead and provide the profit I need. Once this figure is established, I simply concentrate on billing that amount for every hour of service I render.
Certainly this sounds easier than it is and I have written extensively on the concept. I will post on it in the near future. If all available hours are sold, profit will be as expected. If not, there is a marketing/quality issue, which is dealt with separately.
The only historical figures that I monitor are Dollars spent on marketing to maintain a three day backlog and gross sales dollars per total employees.
The former figure tells me how my quality is. Better quality increases referrals and repeat clients and lowers the amount. Three days was picked because I consider that a minimum healthy backlog. The figure is currently very low as we maintain five days with almost no marketing.
The latter figure measures how effective I am as a manager. My goal is $200,000.00 sales for every employee on payroll. I have four techs, myself and my wife. Six employees produce $1,200,000.00 annual sales.
If I wish to add another service advisor, they need to enable an additional $200,000.00 in sales to justify their position, the same with a technician. If true cost is properly established and billed as described, net profit will be consistent and where desired, in my experience.
SPC is a tool, useful to distinguish common and special causes. It is also useful to determine the effectiveness of proposed improvements. I will try to go much more in-depth with these topics in future post. Please let me know if this makes things any more clear. Again, I thank you for your interest and your reply.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
FAQ
Search
Memberlist
Usergroups
Profile
Log in to check your private messages
Log in
This is my last job, to a T. Everything that would happen had to be explained and would get a new rule. One time an air conditioner job came back because the week after I changed the compressor the blower motor went out. In the Monday morning meeting he had me stand up in front of everyone and explain why this happened. Joking I said maybe we should change the blower motors on every big AC job. He actually thought it was a good idea. I quit the next week.
