{"id":1781,"date":"2019-05-31T15:22:46","date_gmt":"2019-05-31T15:22:46","guid":{"rendered":"http:\/\/wiki.thomasandsofia.com\/?p=1781"},"modified":"2019-05-31T15:26:10","modified_gmt":"2019-05-31T15:26:10","slug":"six-sigma-transcripts-6","status":"publish","type":"post","link":"https:\/\/wiki.thomasandsofia.com\/?p=1781","title":{"rendered":"Six Sigma Transcripts 6"},"content":{"rendered":"
\n
\n
<\/div>\n
\n

Steps in the Analyze phase<\/h3>\n
\n
\n

– Have you ever wondered why problems <\/span> that were “supposedly solved” a few months ago resurface? <\/span> Why are we having this problem again? <\/span> Didn’t Mike and his team do this project <\/span> a few months back? <\/span> All too often, problems that were “solved” <\/span> were not really solved because the true root causes <\/span> were not properly diagnosed and addressed. <\/span> That’s why analysis has to be done correctly. <\/span> So, let’s discuss the Analyze Phase <\/span> and the steps involved. <\/span> The Analyze Phase is the third phase <\/span> of a Six Sigma project. <\/span> It is the A in DMAIC. <\/span> The purpose of this phase is to answer the W word, Why. <\/span> Why is there a problem? <\/span> We have to determine which Xs are the key factors <\/span> that impact the problem Y <\/span> in the equation Y = f(x). <\/span> There are five steps in the Analyze Phase. <\/span> Generate a list of potential Xs that impact Y. <\/span> Organize the potential Xs. <\/span> Shortlist and select the likely key Xs. <\/span> Develop a data collection plan for the analysis. <\/span> And, prove the key Xs in Y = f(X). <\/span> Let’s take a look at the first step, <\/span> generating a list of potential Xs that impact Y. <\/span> This is done using brainstorming <\/span> to generate a list of potential suspects. <\/span> In our example of the pizza chain <\/span> with complaints on pizza crust, <\/span> what impacts crust? <\/span> Possible Xs include the flour, the chefs, <\/span> procedures, tossing technique, <\/span> oven temperature, baking time, <\/span> delivery time, and the list goes on. <\/span> List as many potential Xs as possible. <\/span> Next, we need to organize the potential Xs. <\/span> To do this, organize the brainstorm list <\/span> using a tool such as a cause-effect diagram <\/span> or fishbone diagram. <\/span> After the diagram has been completed, <\/span> shortlist and select the most likely key Xs <\/span> that are impactful. <\/span> Use the knowledge, experience, and expertise <\/span> of those familiar with the problem <\/span> to identify which potential Xs <\/span> are the most likely culprits or key X factors. <\/span> Then develop a data collection plan for the analysis. <\/span> We need a data collection plan <\/span> before collecting any data. <\/span> Why? <\/span> So that the correct and sufficient amount of data <\/span> are collected to test the selected Xs. <\/span> In our pizza crust example, this means <\/span> that if you suspect oven temperature is a culprit, <\/span> you may need to run a regression. <\/span> If you’re not sure what regression is, <\/span> here’s a quick example. <\/span> Having good grades in high school <\/span> is a predictor of success in college. <\/span> And in the same way, oven temperature <\/span> can predict a perfect crust versus a burnt one. <\/span> Therefore, you should plan to collect sufficient data <\/span> on oven temperature and how well done the crust is. <\/span> The final step is to prove the key Xs. <\/span> Inferential statistics and hypothesis testing are used. <\/span> Conduct the necessary test such as the regression <\/span> we mentioned to validate the key X factors. <\/span> At the end of the Analyze Phase, <\/span> we have a list of proven Xs that were validated <\/span> by data-driven analysis. <\/span> In our example, let’s say there are three key X factors. <\/span> Tossing technique, oven temperature, and baking time <\/span> were proven to cause variation <\/span> and problems with our pizza crust. <\/span> In summary, the focus of the Analyze Phase <\/span> is on the question Why. <\/span> Using data-driven analysis, we prove the key Xs <\/span> that belong in Y = f(X). <\/span> The power and data-driven rigor of Six Sigma <\/span> has been put to work. <\/span> Correctly understanding the Analyze Phase, <\/span> its purpose, and the steps involved <\/span> will help you render a correct diagnosis <\/span> to get to the true root causes <\/span> or key X factors that impact the problem. <\/span><\/p>\n

\n

How to use the cause-effect diagram<\/h3>\n
\n
\n

– People are late for work. <\/span> Why are they late for work? <\/span> Well, maybe they missed the bus. <\/span> Or maybe they were on time for the bus, <\/span> but there was a huge traffic jam. <\/span> Or it was because of bad weather. <\/span> Or they simply woke up late. <\/span> These are just some possible reasons for being late. <\/span> And for each of these, there will be possible causes. <\/span> For example, for the first reason, <\/span> one could ask, why did you miss the bus? <\/span> And the list of possible causes goes on and on. <\/span> Wouldn’t it be nice to have a tool to <\/span> organize all the possible causes? <\/span> Well, in this movie, we’ll discuss the tool used <\/span> to the facilitate brainstorming and organize potential <\/span> causes or x’s in a six sigma project. <\/span> This tool is called the Cause-Effect diagram. <\/span> So, what is a Cause-Effect diagram? <\/span> A Cause-Effect diagram is simply a diagram <\/span> to organize the list of potential causes <\/span> on a particular effect. <\/span> In the above example, late for work is the effect. <\/span> The diagram is arranged based on logical relationships. <\/span> For example, I missed the bus, <\/span> therefore I am late for work. <\/span> Why did I miss the bus? <\/span> A number of possible or potential causes <\/span> for missing the bus comes to mind. <\/span> Take our pizza crust problem. <\/span> What can possibly cause the pizza crust problem? <\/span> It can be caused by x’s that are related to the oven. <\/span> Technique, delivery, raw materials, or the pizza chefs. <\/span> As for the oven, potential x’s, or causes, <\/span> are temperature and baking time. <\/span> And for each of them there are many contributing causes. <\/span> Here is the completed Cause-Effect or CE diagram. <\/span> This CE diagram takes the form of a tree diagram. <\/span> I used Excel to create this. <\/span> This is easy to revise, update and maintain <\/span> when you’re getting inputs from project team members, <\/span> any additional subject matter experts, <\/span> and process personnel in the brainstorming session. <\/span> The original CE diagram takes the form of a fish. <\/span> With the fish head, where the effect of pizza crust is. <\/span> And the tree diagram is replaced by a skeleton of the fish. <\/span> Showing its primary, secondary, and tertiary bones <\/span> arranged using a Cause-Effect launcher, like this. <\/span> It is also known as a Ishikawa Fish Bone diagram, <\/span> name after it’s inventor Kaoru Ishikawa, <\/span> of Kawasaki Works in Japan in the late 50’s. <\/span> He used the fish diagram since Japanese workers <\/span> can relate to it easily <\/span> because they are a maritime country. <\/span> The birthplace of sushi. <\/span> If this tool was invented in US, <\/span> it would have been a cow diagram with beef ribs. <\/span> Anyway, all the branches of bones contain <\/span> potential causes of theories. <\/span> Until they are proven with facts and data, <\/span> they remain just that. <\/span> The CE diagram should capture <\/span> all possible causes or x’s. <\/span> To do that, you want the fish to be a big fish. <\/span> Not a small fish. <\/span> The next time of Analyze Phase is to select <\/span> from the CE diagram the most likely theory’s or x’s <\/span> to test and validate with data. <\/span> This selection will be done using the knowledge <\/span> and experience of the project team <\/span> and subject matter experts. <\/span> Developing the Cause-Effect diagram <\/span> will help you brainstorm and organize potential x’s <\/span> that impact the y. <\/span> This is the key step in diagnosing <\/span> the x’s that belong in y equals f of x. <\/span><\/p>\n

\n

Introduction to hypothesis testing<\/h3>\n
\n
\n

– If you’ve ever watched a crime or detective show, <\/span> you know that there’s usually a crime committed <\/span> first thing at the start of the show, <\/span> and what usually follows is that detectives pursue leads <\/span> to test various theories, <\/span> and maybe one theory is that the butler <\/span> committed the crime, <\/span> another theory is that business partner is a guilty one, <\/span> and so on. <\/span> Based on their experience and expertise, <\/span> the detectives will select the likely theories to pursue. <\/span> They work on each selected theory by gathering any evidence, <\/span> facts and data. <\/span> If there is insufficient evidence to support a theory, <\/span> they stop pursuing that theory. <\/span> However, if there is sufficient evidence <\/span> to support a theory, <\/span> they will make an arrest. <\/span> Theories are also called hypotheses. <\/span> This is an example of theory testing <\/span> or hypothesis testing. <\/span> So, what is hypothesis testing? <\/span> Hypothesis testing is sometimes called <\/span> the Scientific Method. <\/span> A theory or hypothesis is proposed, <\/span> then data or evidence is collected <\/span> to see if the theory is refuted or supported. <\/span> If the theory or hypothesis is not supported, <\/span> then it is considered disproven. <\/span> If the data supports it, <\/span> then the theory is validated. <\/span> In crime mysteries or detective shows, <\/span> if the evidence showed that the butler was out of town <\/span> at the time of the crime, <\/span> then there is no means and no opportunity, <\/span> then the theory that the butler did it is disproven. <\/span> In terms of a Six Sigma project, <\/span> the theories of potential axis are all listed <\/span> on the cause-effect diagram during the analyze phase. <\/span> Using the knowledge and experience of the project team <\/span> and subject matter experts, <\/span> the likely theories or X’s are selected <\/span> to be tested with data. <\/span> Here’s a CE diagram for pizza crust. <\/span> Let’s just say a number of theories was selected, <\/span> and one of them is oven temperature. <\/span> The theory is inconsistent oven temperature <\/span> causes variation in the quality of pizza crust. <\/span> To use the justice system or court of law analogy, <\/span> a person is innocent until proven guilty, <\/span> and the verdict is either guilty or not guilty, <\/span> never innocent. <\/span> Do we have enough evidence to convict oven temperature <\/span> of causing our pizza crust problem? <\/span> If there is insufficient evidence, <\/span> then we have to dismiss the hypothesis, <\/span> verdict not guilty. <\/span> However, if we find that there is a lot of variation <\/span> in oven temperatures among the various (mumbles), <\/span> and if we can establish a correlation and causation <\/span> with the variability in pizza crust quality, <\/span> then we have overwhelming evidence. <\/span> Since there is overwhelming evidence <\/span> that we cannot dismiss the hypothesis, <\/span> verdict guilty. <\/span> Oven temperature is a proven cause, <\/span> or key X in our pizza crust problem. <\/span><\/p>\n

\n
\n
<\/div>\n
\n

Data collection in the Analyze phase<\/h3>\n
\n
\n

– If you ever watched or read a tale of Sherlock Holmes, <\/span> you know that he does not just stumble upon evidence. <\/span> He has a rigorous plan <\/span> for gathering information and evidence. <\/span> He considers how he will collect it, <\/span> who he will talk to, <\/span> and where he should go. <\/span> He has a plan for data collection. <\/span> So, let’s imagine ourselves as Sherlock Holmes <\/span> and let’s discuss how to plan for data collection <\/span> to test hypotheses during the analyze phase <\/span> of a Six Sigma project. <\/span> Planning for data collection in the analyze phase starts <\/span> with the following questions. <\/span> What is the theory or hypothesis to be tested? <\/span> For example, you suspect that oven temperatures used <\/span> are not consistent from store to store, <\/span> or from chef to chef, <\/span> even within a same store. <\/span> What type of hypothesis test should be used? <\/span> For example, if we want to test <\/span> for differences among groups, <\/span> in this case, differences in average oven temperatures <\/span> between pizza stores or between the chefs in each store, <\/span> what data is needed? <\/span> In our example, it would be oven temperature by store, <\/span> by chef, during peak hours, <\/span> and off-peak slow hours, and so on. <\/span> How much data is needed to run the test <\/span> and draw a valid conclusion? <\/span> Do we need one data point or 100 data points? <\/span> From whom, when, and where? <\/span> From which persons should data be collected? <\/span> From when to when? <\/span> Last month, last year? <\/span> For how long? <\/span> A day, a week, or a month? <\/span> From where? <\/span> Which location? <\/span> Which database? <\/span> I recommend that a data collection plan be laid out <\/span> in a table with columns. <\/span> The column headings are the questions I listed. <\/span> Each selected X will be a row entry in that table. <\/span> The benefit in spending the time and effort <\/span> to plan for data collection cannot be emphasized enough. <\/span> Planning for data collection during the analyze phase <\/span> is important so that the project team knows exactly <\/span> which hypothesis are to be tested, <\/span> which data needs to be collected, <\/span> how much data to collect, from whom, <\/span> from when to when, and from where, <\/span> in order to validate the key X’s that belong <\/span> in Y = F of X. <\/span> In other words, you validate potential causes <\/span> or key X’s with facts and data. <\/span><\/p>\n

\n

How to analyze graphs and charts<\/h3>\n
\n
\n

– Customers are complaining about pizza crust. <\/span> The pizza chain is losing revenue. <\/span> A Six Sigma project team is now in the Analyze phase. <\/span> Potential X’s were generated and a few X’s <\/span> were selected to be tested and validated. <\/span> One potential X is that one or two locations <\/span> is causing all the crust problems. <\/span> Another potential X is that the crust problem is caused <\/span> by variation in oven temperatures. <\/span> We can use graphs and charts to help validate key X’s <\/span> during the Analyze phase. <\/span> For example, let’s address the first potential X <\/span> mentioned in this analysis. <\/span> Since the number of crust complaints is Discrete, <\/span> or Count Data, a bar chart or Pareto chart, is appropriate. <\/span> But if I want to prioritize the locations <\/span> based on a number of crust complaints, <\/span> then the Pareto chart is the most suitable graph. <\/span> Here is the Pareto chart of pizza crust complaints <\/span> across locations. <\/span> Looking at the Pareto chart, we can conclude that, yes, <\/span> the crust problem is indeed isolated to just one location. <\/span> We have proven one X. <\/span> Next, let’s address a second potential X, that the crust <\/span> problem is caused by variation in oven temperatures. <\/span> Since we now know that the crust problem is caused by <\/span> one location, the Six Sigma project team can focus there. <\/span> To test and validate this potential X, <\/span> the project team would need to plot the temperatures used <\/span> for each pizza, by each chef working in this location. <\/span> Which graph or chart can be used? <\/span> Since temperature is continuous data, <\/span> it is appropriate to use histograms, <\/span> dot plots, or box plots. <\/span> These graphs display the pattern of variation, <\/span> showing how spread out the temperatures are, <\/span> in addition to showing where they are centered. <\/span> Here is a graph of box plots showing oven temperatures <\/span> used by each chef at this location. <\/span> Looking at the graph, it is obvious as to what’s happening. <\/span> The variation in oven temperatures is caused by two chefs, <\/span> Chefs B and C. <\/span> Their oven temperatures are the most inconsistent. <\/span> We have proven another X. <\/span> We should follow up with Chefs B and C. <\/span> This is how you can use graphs and charts to analyze <\/span> and help prove potential causes and validate key X’s. <\/span><\/p>\n

\n
\n
<\/div>\n
\n

How to analyze process maps<\/h3>\n
\n
\n

– A process map is a graphical representation <\/span> showing the steps, activities, and decisions <\/span> involved in a process. <\/span> This process map was created earlier <\/span> during the measure phase, <\/span> so how do you analyze it? <\/span> Well, let’s talk about it. <\/span> I have some recommendations and questions you can ask <\/span> when you analyze a process map. <\/span> My first recommendation is to identify any bottlenecks. <\/span> Ask, where are the delays? <\/span> Where are the holdups? <\/span> Highlight those steps. <\/span> Then ask, why are there bottlenecks? <\/span> Is it because there’s too much work required <\/span> at those bottleneck steps? <\/span> Could it be the workload is not balanced among the steps? <\/span> Perhaps there is not enough people or resources assigned <\/span> to carry out those steps. <\/span> Identify any workarounds. <\/span> Perhaps the official procedure is too unwieldy. <\/span> People bypass bureaucracy and take shortcuts. <\/span> Question why are there so many steps. <\/span> Ideally, the perfect process has only one step. <\/span> Snap, and it’s done. <\/span> Mission accomplished. <\/span> Examine each step and ask, <\/span> why is it needed? <\/span> What value does it add? <\/span> Identify the steps that really add value <\/span> to the final outcome or end result. <\/span> These are called Value-Added steps or VA for short. <\/span> If they don’t add value, <\/span> highlight them as Non-Value Added steps, <\/span> or NVA for short. <\/span> If there are steps that are necessary <\/span> to get to the Value-Added steps, <\/span> for example, booting up the computer <\/span> and logging into the system, <\/span> classify them as Value-Enabling steps, or VE for short. <\/span> They do not add value, but at the same time, <\/span> they are necessary to enable you <\/span> to get to the Value-Added steps. <\/span> Zero in on the decision diamonds. <\/span> These diamonds may be inspections, <\/span> checks, or evaluation decisions. <\/span> How often does it fail to meet the decision criteria? <\/span> If significant, ask why. <\/span> Look at the rework loops <\/span> where work has to be sent back to be redone <\/span> because it failed an inspection. <\/span> Ask how long are the rework loops? <\/span> I’ve seen long rework loops in many processes, <\/span> and the following scenario is a pet peeve of mine. <\/span> You may have seen this in your own processes. <\/span> Say we have a 100-step process, <\/span> and there is an inspection step that takes place in step 99, <\/span> and if it fails that inspection, <\/span> the product has to go all the way back <\/span> to step two to be redone. <\/span> That’s a rework loop that is 98 steps long. <\/span> Shorten the rework loop. <\/span> Why can’t the inspection take place right after step two <\/span> or better yet, implement self-checking, <\/span> or even better yet, mistake-proof what happens in step two. <\/span> So there you have it. <\/span> Some tips and questions for analyzing process maps. <\/span><\/p>\n

 <\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

Steps in the Analyze phase – Have you ever wondered why problems that were “supposedly solved” a few months ago resurface? Why are we having this problem again? Didn’t Mike and his team do this project a few months back? All too often, problems that were “solved” were not really solved because the true root ..<\/p>\n

<\/div>\n

Read more<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[42],"tags":[],"class_list":["post-1781","post","type-post","status-publish","format-standard","hentry","category-six-sigma-foundations"],"_links":{"self":[{"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=\/wp\/v2\/posts\/1781","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1781"}],"version-history":[{"count":3,"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=\/wp\/v2\/posts\/1781\/revisions"}],"predecessor-version":[{"id":1784,"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=\/wp\/v2\/posts\/1781\/revisions\/1784"}],"wp:attachment":[{"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1781"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1781"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wiki.thomasandsofia.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1781"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}