During this 1-hour intermediate webinar you’ll learn about some famous unintended consequences and how we can avoid them in our own worlds using FMEA: Failure Mode & Effects Analysis.
- Introduction to the FMEA
- What the role of risk is in process improvement
- How to address risk with the FMEA
- History of unintended consequences
Karlo: Welcome to GoLeanSixSigma.com’s webinar. Thanks for spending some quality time with us. Over 600 people have registered, and we are really excited that you are here. Lean and Six Sigma are the go to improvement methods used by leading organizations all over the world to minimize costs, maximize profits and develop better teams all while creating happier customers. Every month we craft webinars just for you, our global learner community! It simplifies the concept of tools of Lean and Six Sigma so that you can understand and apply them more easily, and be more successful. Today’s webinar is titled, “How to Avoid Unintended Project Consequences Using FMEA.”
I’m Karlo Tanjuakio, a managing partner at GoLeanSixSigma.com. Today’s presenter is also a managing partner at GoLeanSixSigma.com. My colleague the wonderfully talented and constantly knowledgeable Elisabeth Swan! Hey Elisabeth, how are you!
Elisabeth: Hey Karlo! How’s it going?
Karlo: Really good! Thanks! Are you excited about today’s webinar?
Elisabeth: I am totally excited about today’s webinar!
Karlo: Awesome, me too!
Our Expert: Elisabeth Swan
So Elisabeth is an Executive Advisor, Master Black Belt, Consultant, Coach and Trainer. For 25 years, Elisabeth has helped leading organizations like Amazon, Charles Schwab, Target, Volvo, Alberta Health Services, Starwood Hotels, and many others to successfully apply Lean Six Sigma to achieve their goals. Elisabeth lives in Cape Cod with her husband, and cute geriatric cat.
How to Interact
So here a are few housekeeping notes before we begin. During the webinar all attendees will be in “listen only mode.” At the end of this presentation we will have a questions and answer session, but please feel free to ask questions anytime by typing them into the “questions” area. We will ask you to participate in some polls as well. If we don’t answer all of your questions, we’ll be sure to post the answers, as well as share a recording of this webinar on our website, at GoLeanSixSigma.com.
Alright, I’m handing it over to you Elisabeth.
Elisabeth: Okay, let’s have our first interactive session and we are going to find out where everyone is from. We have, as Karlo said, hundreds of attendees from all over the world, so let’s see how early, or late people are up for this webinar. So, click on “ask a question”, and type in where you are dialing in from.
Karlo: Alright, we have California, Maryland, Virginia, Canada, Oregon, Washington, Romania, New Mexico, London, Germany, New York, Nigeria, India, Egypt and Calgary! So it looks like we have a really nice mix, thanks for joining us everyone!
Elisabeth: Yes! Thank you for staying up late, staying up early, whatever it took. I’m happy to be here with you guys from all over the world, thanks for joining.
Who Is GoLeanSixSigma.com?
Elisabeth: As Karlo mentioned, we have both been with GoLeanSixSigma.com since it’s inception. We founded GoLeanSixSigma.com on a few guiding principles. It’s our job to simplify complex concepts, and not to confuse anybody. The key to effective training should be practical, accessible and enjoyable. Why shouldn’t this be an enjoyable training? So we made it our mission to transform how people learn about Lean Six Sigma. We based all of our training at a case study that takes place in the Bahamas, at The Bahama Bistro (if we’re making it up, it might as well be in a nice place!). We want to make it easy for everyone, everywhere to build their problem solving muscles. So, we are going to continue to do that today for you with FMEA.
We’ve Helped People From…
Elisabeth: So luckily there is a lot of organizations out there that have the same view of how to deal with Lean Six Sigma that we do, so let’s take a look at some of the organizations we’ve worked with. Some of the services are online, healthcare, financial, manufacturing, state and government.
Six Sigma is about problem solving, and basically if you have an organization you have problems. So they need people that are good at problem solving, so we’re looking to strengthen and build problem solving muscles with Lean Six Sigma.
Elisabeth: So let’s take a look at our agenda for the day. We are going to look at the role of risk in process improvement. We are trying to make changes, but there are always risks when you change anything and that will take us to a tool called the FMEA (Failures Modes & Effects Analysis). Next we will talk about how to address risk with FMEA. Finally, we will give you some famous instances of unintended consequences that happened in the world. So let’s dive in!
Where Does Risk Come In?
Elisabeth: The concept that we’re working with, is that when you make a change there are ripple effects. Some of them are foreseeable, but some of them are not. It takes concerted efforts to think about, “Are those ripples?” and how to plan for them. Can we prevent them? And if not, what else can we do about them? So what are the ripple effects and where does risk come in?
Elisabeth: So this is one of those famous examples of unintended consequences. In 1915, this was due to the disaster of the the Titanic. The government passed something called the Federal Seaman’s Act, and that required the Eastland, (a Great Lakes passenger steamer) to retrofit the upper decks with a complete set of lifeboats (now as you remember, especially with the movie The Titanic, they didn’t have enough lifeboats for everyone). Now the rule is, you have to provide enough lifeboats for everyone. So they retrofitted the Eastland, but the problem was that the Eastland was already top heavy, so the extra lifeboats made it dangerously top heavy. On a maiden voyage the passengers congregated on the portside and the Eastland rolled over and sank! It was only 20 feet from the wharf on the south bank of the Chicago River. That was just threes years after the wreck of the Titanic. So looking at, “How could they have thought about what could go wrong with the extra lifeboats?”
Types of Risk
Elisabeth: There are different types of risks and in this case it was the risk to bodily harm people. You also have the risk of money loss to the company, and the customers. The classic case of risk is when people put out two for one offers, and demand goes so high the company can’t keep up with it. The demand out strips the supply, and people get irritated that they didn’t get the product they were offered. Now there is a risk of harming a reputation, which is the underlying risk to all of this. A risk now that we are dealing with in the modern ages is the risk of hacking. Financial services deal with this a lot. They are constantly trying to balance, how to easily make things for you the customer to get access to your accounts to do what you need to do, with as few steps as possible. But if they make it easy for you, they are also making it easy for hackers. There are people working 24 hours a day to figure out what’s the latest threshold, where are the loopholes, where’s the hole in the code and then manipulate it. Another, is risk of safety. For example the Galaxy Phone that just came out, and the battery overheats causing it to explode. So how could have they have foreseen that problem? They clearly made some changes to the battery and the result was too much risk. Next, is the risk of simply aggravating people. Is there something that you’re doing going to cause potential aggravation to a customer? Finally, all of this can hurt the reputation. So we are going to look at how do we deal with risk.
Cause for Alarm
I’m going to take you to a personal story, and we are going to use this story to bring you to this tool, the FMEA. So back in 2008, I traveled a lot for work. I was on the road training with different organizations and consulting. I was staying at a Westin, one of the nicer Starwood properties and when I got to my room they had replaced all of the alarm clocks with this very slick looking docking station. It was sheer gunmetal grey, very clean surface I couldn’t really see buttons on it. I loved it, but I didn’t know how to use it. I have got into the habit of requesting a wake up call and setting the clock. One of them would always go wrong, I would not get a wake up call, or the clock wouldn’t work, so I always did both. So now I wanted to make sure that I could set it, Since I couldn’t figure it out, I called the front desk, and asked how to set these new alarm clocks. They said, “well they are new, and we aren’t really sure, you should check with housekeeping”. So I checked with housekeeping and couple of people came in and they tried to help me, and they couldn’t figure it out either. So they called someone from maintenance and they came over with 15 different keys on his belt, and he took one look at the whole situation, and he said he would be right back, then he came back and must have gone into the storage somewhere because he had an old alarm clock, the type that I have been used to in forever. And that solved the problem. But then I thought, what about the other guests? Did they know how to work this new gadget? Did they try and do what I did by calling the front desk? And call maintenance? Occupy housekeeping for half an hour? This is the kind of thinking behind Failure Modes & Effects Analysis. What could go wrong?
A Gift From NASA
Elisabeth: We got FMEA from NASA. If you think about it, they really have to do a lot. They have to use risk assessment upfront, because they cannot use trial and error. They couldn’t launch astronauts, and test if they could get them back home safely. Instead, they had to basically think through all the possibilities. Now if you think about Apollo 13, (or saw the movie) there was a lot of back up systems, and even when the backup systems went wrong they could still figure out a way to handle the malfunction. Thinking ahead comes from NASA. Now let’s take a look at what they developed.
Elisabeth: So this is the FMEA, it’s a big spreadsheet with a lot of columns. It has words on it like severity, RPN and detection. People often ask me, “What’s the FEMA?”, instead of FMEA. FEMA is the organization which provided relief during Hurricane Katrina, and that’s not right. So let’s break this down, and go through the example I just gave you to learn about the FMEA.
Elisabeth: The first thing we are going to ask is, “What are we discussing here?” We are going to discuss an improvement, change and step. In this case, we are going to discuss a new crazy alarm clock that they had at the Westin Hotel.
The first column says, “What is that step, what is that feature?” In this case, it is the new alarm clock docking station.
Next question, “What could go wrong?” This is the idea of a failure mode, it just means, “How could it fail?”, or even simpler, “What could go wrong?”. This means I could possibly fail to set the alarm to wake up, because I don’t know how to work the alarm.
So, “How bad would that be?” This is a question of the severity of the the failure. Now in this case it is, not getting up and attending my webinar and I don’t get to my class. This will cause me some issues.
How Bad Would That Be?
Elisabeth: Now you will go to the scale. The FMEA always has scales, and the first one we deal with is Severity Scale, “How bad will it be?” It is a scale of 1 to 10, with 1 being the least problematic, and 10 is the worst. In this case this it would be a 10, because it would expose me to loss, harm, disruption and there would be no warning. There would be no way for me to know that the alarm clock was not set correctly. The severity scale really depends on your process. Different processes have different issues that would be a 10, so your scale would differ.
Adapting the Scale
Elisabeth: Let’s think about what your scales might be? What would a 10 be in your process? It could be a violation to a government regulation, expose people to harm, a risk of financial loss, expose private information or it could just aggravate customers.
QUICK POLL: What might a “10” be in your process?
Elisabeth: I will launch a poll and you guys will tell me what a 10 would be in your process.
Karlo: We are at 67% now, and when we reach 80% we will close the polls and share the results.
Alright! We are at 80%, and I have closed the polls. It looks like the highest percentage is, “failure exposes people to bodily harm” at 33%, next, “risk of financial loss” at 22%, followed by “potential to violate a government regulation” at 18%, then we have “could aggravate the customer” at 15%, finally “may expose private information” at 13%.
Elisabeth: Okay in some of these there might be a combo, it may not just be one or the other. Meaning more than one thing that may have happened. My advice is you should look online, there a lot of examples of polls out there and find the one that works for you. Make sure you adapt the poll, especially the severity poll which will need to be adapted to your process.
Rating the Severity
Elisabeth: Now back to the FMEA. We have inputted the first three columns, and so far we have a new alarm clock, I failed to set up new alarm clock, I failed to wake up and on the severity scale this is a 10. So this is what we have so far.
Now the next part of the FMEA is, “What’s the cause?”. Talking about this failure mode (and there will always be more than one failure mode, but we are just going to go with one right now), why would that happen? We are looking at the reason for that failure and in this case it was, I did not understand the new technology. So that is the reason, that is the potential cause and that gives us a sense of “How often would that happen?”. The idea is if you can think about what the cause of that failure would be, then you can work out how often it might happen.
How Often Would it Happen?
Elisabeth: Now, we are going to another scale called the occurrence Scale. The occurrence scale doesn’t generally need to be as adapted as the severity. You can use either a time period- “once a day” is a 10, and “once every plus years” is a 1. If you have a production situation, it might be per item failure rate- a 10 is “greater than is equal to 1 in 2” (so 1 in 2 items have this particular failure), and then down at a 1 is “1 in 6,000”. So this scale is a little more universal, and easier to use. So work out the scale that you want to adapt.
Elisabeth: So now let’s think about what you guys think. Take a look at the scale, and at the situation that was happening with this new alarm clock. What do you think the occurrence would be in this particular situation?
QUICK POLL: How often would that happen?
Elisabeth: We are going to launch another poll, and I want you guys to have a voice in this one, because this is what you would do as a team! You will basically go to the specific part of the scale based on what part of the FMEA you were working on, and use that scale as a reference point to rate it on a scale of 1 to 10, or whatever you think.
Karlo: Aright! I have closed the polls and once per day is the highest at 52%, followed by once every 3-4 days at 16%, every 3 months is 13%, once per week at 11% and once per month at 8%.
Elisabeth: Okay so more than half the group agreed with me, it would happen more than once a day. Back in 2008, you would have more than one person a day having this problem.
Elisabeth: So now we have occurrence at a 10, because it is going to happen more than once a day. I think the occupancy of the hotel will have an impact, because there was over 400 rooms so the odds are pretty good.
Would We Know In Time?
Elisabeth: The next question is, “Would we know in time?”. This section is looking at whether you would have time to react to this before the impact of the customer. Is there anyway the process could alert someone in charge to do something before the customer was impacted? In this situation we will have to say no, because there was no way to know that the alarm was mis-set. You can’t have housekeeping coming by the room knocking on the door checking if you set your alarm. So we do not have a way to access if customers have made this mistake.
Elisabeth: If I go to the Detection Scale, on a scale of 1 to 10 where 1 is “almost certain”, current controls are almost certain to detect the failure mode and the reliable detection controls are known with similar processes. Or a 10 “almost impossible”, there is no known control available to detect failure mode. In this case, the failure is setting the alarm clock wrong, and there is no way to tell. This gets an automatic 10.
Bottom Line That For Me
Elisabeth: So now we have a detection of 10, which brings us to a Risk Priority Number. The Risk Priority Number is multiplying severity (which was a 10), by occurrence (which was a 10), by detection (which was a 10). The Risk Priority Number is 1,000, and that is the most it can possibly be. The role of a Risk Priority Number is to prioritize the risk.
Now we are just looking at one failure mode, and one potential failure effect. A failure mode might have a lot of different effects, but we are just looking at one. The FMEA becomes very busy with a lot of more roles, assessing different steps, and looking at all risks in terms of how bad they are. Once you multiple severity, occurrence and detection whats the number you got? You’re going to address the higher numbers first, and that’s the basic goal.
Elisabeth: “What are you going to do about it?” Now that we know what the risk is, how are we going to accommodate or do something about that?
The team brainstorms and they come up with recommended actions to reducing either the occurrence of the cause, or how to improve detection, or if possible reduce the severity- The staff will give instructions and leave the instructions in the room. Okay, problem solved! Well, who is going to do that?- All the housekeeping managers are responsible of making this happen. Finally, was the actions taken? We want to see when it was actually done and when- The directions were created, copied and distributed on September 15 so we are all good.
Elisabeth: The far right of the FMEA shows you made these changes, and fixed this process.
What is the severity? Well it’s still just as bad if people set this alarm clock wrong, and they don’t get up. Guests are irritated, so it’s just as bad. How about occurrence? Now that we have occurrence reduced, our scale shows it happens once a month. Over 400 rooms and once a month someone’s not going to read the instructions, and they’re going to mis-set the alarm clock. So we have dropped the occurrence down to a 7. What about detection? This is still a 10, because there is no way to know if guests set their alarm clock wrong. Now we do a recalculation of the Risk Priority Number, and we get a 700. It’s lower but it’s still not great, because there is still risk involved.
If I use another Lean Six Sigma tool the 5 Whys to understand why the guests were not understanding, I would say that the main clientele were a little older and not quite comfortable with technology. If you had your teenage nephew on this he would have nailed it right off the bat! He would have been loving it, playing music and everything’s great. But in my case I wasn’t there.
I read about another hotel chain the Marriott, and they also introduced a very hip alarm clock docking station. Within a month they switched them all back to the standard old alarm clock, and I’m guessing for similar reasons. Their solution was to just get rid of the change all together, which is another option.
Elisabeth: Let’s go back and see where we are. We have the whole line filled in with our FMEA. We have completed one failure mode, one failure effect and we worked that all the way through to the right. We could have had more than one action, and I mentioned we can have a lot more failure modes, and a lot more failure effects. But I am trying to make it simple to teach you the concept.
Elisabeth: Now we’ve done the FMEA, so let’s think about the risks in the process. Once we unearth them, and determine what could go wrong, how do we do something about that? I want to you to think about that for a while, because I want to get some input from you.
I’m going to ask you a question. I want you to imagine that you are in charge of fire safety for a small hotel chain. What kind of things are you going to be concerned with? I want you to go into the question area, and type what kinds of things you are going to be concerned with as someone who is in charge of fire safety in a small hotel chain.
Okay, we have: staff training, fire alarms, exit plans, flammable Items in the room, number of exits, death and casualty insurance, fire extinguishers, sprinklers and evacuation plans.A few of you have stated you will be looking at infrastructure.
There are two ways to go, and one is contingency plans. Contingency means the failure has happened, and in this case the fire. What is the contingency plan once the fire has happened to lessen the impact fire, and reduce the amount of damage it might do? Basically contain the fire. And the response I received from you guys were correct: fire alarms, exit plans and all of these answers will help you with your contingency plan. But what this means is that we are planning for fire. We are assuming that a fire is going to happen.
The second way to go is preventive actions. And the responses I received from you guys were: staff training, infrastructure, is the building built with flammable materials? No smoking. Other things we can do to prevent fire. Storage of cleaning liquids should not be near in term of kitchen and flames.
So contingency plans, mean we are planning for things to go wrong, and how do we contain it. But preventive actions, are how do we prevent it from happening in the first place? How do we prevent the error from turning into the failure?
Elisabeth: That’s what I want to push now: How do we stop the failure in the first place? This brings us to another technique called Mistake-Proofing, and you’ve probably also heard this called Poka Yoke. Not to be confused with doing the hokey pokey! This means “How can we stop people from making mistakes?”, and most mistakes are human made. We are trying to basically protect people from themselves.
People suffer from forgetfulness, lack of concentration, have a misunderstanding and jump to a conclusion, have misidentification with poor labeling, poor visual management, inadequate skills and willful errors by taking short cuts. I’m constantly amazed, because my husband will tape the safety feature on the handle lever to the lawnmower, which makes it run. Now if you tape that safety lever and you come across a failure, you’re unable to quick release the safety lever because it’s taped! This is a human error. We also have inadvertent errors for example, clicking send instead of spell check, slowness, driving with a delayed reaction, lack of standards and inconsistent methods (this is why we work towards standard work. Even surprised errors for example, when training pilots to practice being stalled in a plane simulator, which is such a surprised rare error, because they are not ready for it, so they have to practice. Unintentional errors, may be sabotage, but this is rare.
Types of Poka Yokes
Elisabeth: Mistake-proofing comes in two flavors: Hard Poka Yokes and Soft Poka Yokes. Hard Poka Yokes, prevent people from making the error which leads to the failure. You can’t have a defect, because they are unable to do it. Soft Poka Yokes are warnings, they warn people that this could cause a problem and people can choose to do it anyway.
The alarm that goes off when you don’t fasten your seatbelt in the car, you can ignore it and drive unbuckled. Or the light that goes no and tells you that you are low on gas, you can ignore it and run out of gas. These are examples of Soft Poka Yokes. But you cannot take the keys out of the car, until it is in park. Or you cannot fit a diesel nozzle into an unleaded fuel tank. These are examples of a Hard Poka Yokes, because it will prevent you from making that error.
A topical issue now is texting while driving, which is clearly a huge risk that has caused a lot of accidents. The failure mode on a car accident would be a severity of 10, because it is bodily harm. This is happening more and more, because there is no way to warn people. If you’re texting while driving you could cause an accident, and there is nothing to give you that warning. A recent article pointed out that cellphones makers have the ability to block people from texting while driving. I think Apple in particular has a patent for technology that will prevent texting while driving, but it has not deployed it yet but that could be coming soon.
Elisabeth: Let’s come back to unintended consequences that are public, historic and current, how did we deal with them, finally just reflect on how they could have been prevented or not.
Wi-Fi Kiosks in NYC
Elisabeth: Public unintended consequences. This is very recent, and you may have heard about it in the news, it’s the New wi-fi kiosks in NYC. About 400 of these kiosks were installed where you can power up a gadget, and get free wi-fi. This was the mayor’s effort to create better access for everyone.
Now the unintended consequence was that people started treating these kiosks like personal living rooms. People brought chairs, tables, couches, ordered food, alcohol. The kiosks began to become party centers with a lot of noise and people drinking into the night. People were making this a place to live, if they did not have a home. There was a lot of complaints. So they disabled the wi-fi as of September 16, and now they are trying to figure out what to do. Could they have planned for this? Could they have thought ahead of what the risks were of providing free public wi-fi in the middle of a sidewalk? What could go wrong? I get the feeling they did not do a lot of risk assessment, and they had a plan for 8,000 so this is a big deal and cost a lot of money. This is not a small mistake to make, and could’ve used an FMEA to basically prevent issues with the wi-fi.
Elisabeth: Going back in time, you’re all familiar with the 1919 National Prohibition Act, which lead only two people being able to buy alcohol in public spaces. It also lead to a huge underground network of backroom distilleries, illicit supply chains, vast quantities of illegal alcohol that was all facilitated by increasing organized crime. This was the beginning of the mob, and completely an unintended consequence. The immense profits was basically an enabled expansion of criminal enterprise, and precipitated such an increase of violence that President Roosevelt had to repeal the act in 1933. Organized crime had already created deep lasting roots and excellent supply chains for themselves. Unlike the Kiosks they could not repeal the new underground mob structure.
Wolves in Yellowstone
Elisabeth: This is positive, and we call this “a happy accident”. If you are not aware of this, there is a fabulous link with a great video showing how this happened, and we’ll share that link with you later.
This is an issue with Yellowstone National Park, in the United States. In 1995 the deer population had no predators, and it had grown so big that they wiped-out almost all the vegetation leaving completely barren valleys in Yellowstone. They came to the idea of reintroducing a very limited number of wolves back into Yellowstone, the wolves had been there years ago but were removed over time. Just the presence of the wolves (so this isn’t wolves actually harming a deer) the deer avoided the valleys, and kept to the woods not going through the open valleys.
An immediate growth of vegetation began, trees alone grew to five times their height in three years. So think about that! Really massive change! It also created a habitat for mice, rabbits, birds, beautiful hawks, golden eagles and bald eagles. All these returning animals, with a profoundly improved ecosystem.
The biggest prize that they didn’t plan on was the change in the vegetation reducing the erosion with the river. Before the river had been meandering, because there was nothing to stop it from changing course. The river was broadening into a large flat area, into sort of a swamp. But now the river had to find edges, because the vegetation could keep the erosion at bay, making the path of the rivers change. It changed the physical geography of Yellowstone park, which was completely unexpected! As you can tell, all these effects were very positive. We’ll leave you that video, it’s very incredible to see. So this was an unintended consequence, but a great one! So they are not all bad.
Another great one was with Starwood Hotels. They did something called the heavenly bed. They didn’t expect that the guests would love the heavenly bed, which had super cushiony mattress, extra pads, extra fluffy pillows, beautiful quilts. All about comfort! Guests wanted it, so so they had to suddenly figure out how to create a supply chain to supply guests who wanted to order the heavenly bed. Again a good thing, and they did not expect it at first!
Templates and Tools
Elisabeth: Let’s look at the two tools we just went over. The FMEA, is a spreadsheet and you can download the spreadsheet for free from our website with examples of how to use it. Keep in mind this great simple idea of Poka Yoke, or mistake proofing which is a beautiful antidote to the risk that come up in a process, and some of these are very simple. I would say, always go to what’s the preventative measure before you think about contingency plans. So don’t plan for fire, think about how to prevent it before you do the contingency plan. This is a powerful pair if you can think through your risk, and prevent them from happening. Those preventive measures go back into your process, your new project plan and they become part of your implementation plan of how you’re going to run your process.
Today We Covered
Elisabeth: We talked about the role of risk in process improvement, and ripple effects you’re up against when you change things, and how to plan for it. The FMEA is the way to plan for those ripple effects. Then, how do you address that risk? Well, we come up with preventative measures, and contingency plans. Finally, of course we gave you some entertaining examples of unintended consequences in our lives around us.
Elisabeth: We want you to take a moment and ask questions. So what questions do you have for me? I will answer as many as I can on this webinar, then whatever I don’t answer will be answered after the webinar. We’ll post all the answers along with this webinar, slides and recording. Everything will be up on our site, and you will have access to it all!
Elisabeth: While you guys are submitting your questions, here’s a reminder of what we have for you guys. You can learn more about the FMEA, Green Belt Training & Certification and Black Belt Training & Certification. You can get the free Yellow Belt Training.
Elisabeth: The next webinar is Intro to Lean Six Sigma. Thats Tracy O’Rouke, my fabulous colleague that will be running that, so take a us back to a little high level webinar but a great primer on Lean, the combo of Lean and Six Sigma
Elisabeth: Another thing that just came out is a podcast which is great. This is Jerry Wright, he is president and chairman of the board of the Association of Manufacturing Excellence. It’s a great interview, some great advice, great perspective of the Lean Six Sigma world from the manufacturing side
Elisabeth: So Karlo, what questions do we have?
Karlo: So we have a few questions in, and just as a reminder if you have any questions you can enter them into the questions area on the right side.
The first questions comes from Jason. In a process with about 100 steps and interdepartmental hand-offs, does each step have an FMEA done?
Elisabeth: That is a great question! You bring up another point, which is you can do an FMEA on an existing process, which we call the “as is process”. You could just do some risk assessment on the process as it stands. We are not doing a project, but just an FMEA. If you have 100 steps with interdepartmental hand-offs, then I would look for the areas that either involve moments of truth, so where in the process does it interact with the customer. The way the customer is giving information, getting a phone call, submitting an application things like that. Also you mentioned handoffs, so those are a great potential of what we call “dropped balls”, and any handoffs are potential failures. People don’t get the information they needed, don’t understand what’s been given and don’t follow through with what the other department asks for. As the organization gets bigger they get more silo, and that means people stop talking to each other, they don’t really know what the other department does. We get into a habit of sort of “throwing stuff over the wall”, like “oh it’s yours now.” So the FMEA is a great way to address those hand-offs. You have to do a little triage, you don’t want to do every single step, you want to brainstorm with the team, “hey what are the potentials for pitfalls? Let’s address those first!” Now people generally have a good sense of where stuff isn’t handled appropriately between the different departments. So great question, thank you Jason.
Karlo: Thanks Jason. So our next question is, some people are suggesting a simpler scale on five levels instead of ten, what would you suggest?
Elisabeth: I am a firm believer that you make the tool work for you! I’ve definitely seen a lot of five step scales, and if that works for the process I’d say go for it! Especially if you make the use of the tool easy, then people want to do it with you. It’s a great tool for online if you teams that are split up, and they are not colocated running everything by conference call. The FMEA is a great tool to use on a conference call, because you can see exactly what’s being entered, you can pop up the scales whether it’s a five point or a ten point as you’re doing it. So I’m all in favor of ease of use, and if a five point scale works, go for it!
Karlo: Awesome! The next question comes from Ria. Should there be more than one problem in a particular project? Are these problems to be analyzed individually using the FMEA?
Elisabeth: Yes. You can use one FMEA, but you want to address each step, feature and change individually. So if you see opportunities with a couple different things you want to accept them separately, because each one of them could have more than one failure mode. Also each failure mode can have more than one effect. We did a very simple one today, I kept you at one line and that’s easy, but the reality is that these things have a host of potential problems and potential risks. Yes, address them separately.
Karlo: Great! Our next question is from Bryce. At which step during the DMAIC is it most important to introduce the FMEA?
Elisabeth: Good question, thank you! In the DMAIC process the classic step when you are using the FMEA is the improve phase. The improve step is when you decided on the implementation plan, you have improvement ideas, selected all the solutions you want and now you’re getting close to implementing, or you’re getting close to a pilot. We want FMEA before pilot, and definitely before implementation. It’s guaranteed in the improve phase, but potentially earlier on in the process, you can do it in the analyze phase to look at existing risks. This might give you ideas of what to change, and you can also use it as an analysis on an “as is project”. You get some idea that this is actually a really risky step. Is there another way to do this? Do we have to do it this way? Does it have to be this complicated? Or does this have to give us so much exposure in this particular process? So absolutely improve phase, potentially analyze phase. Great question, thank you!
Karlo: Great! The next question is from George. If we are talking about improving the quality of service provided, should FMEA focus on what the supporter needs to provide and what the customer receives separately? Or both at the same time?
Elisabeth: Definitely both. You want to look at both ends. The surface itself, the quality and then how you get there. It’s a really nice way to think about it, at those two angles. So yes both.
Karlo: Thanks! So we have about ten minutes left, and just as a reminder if you have any questions please ask them in the questions area.
The next question comes from Maggie. Is there an ideal number of lines in an FMEA sheet for it to be of a real value?
Elisabeth: No. It is one of those classic “it depends”, so it really depends on what your looking at. If you think about different processes in the automobile industry with all the parts that go into a car, both mechanical and electronical systems. They are just massive! When they are thinking about changing the electronics or feature in a car, they have to do an FMEA on every single feature and electronic change. So it really depends on the process.
A simpler process might be, we took away this inspection step. So what could go wrong if we don’t have someone reviewing applications before we enter them? You could quickly go through the application, and identify where it could go wrong. So it really depends on the process. and how complicated or complex the process is. Thats a good question!
Karlo: Thanks Elisabeth! The next question, is it possible to capture all unintended consequences?
Elisabeth: That’s a really good question. I will have to say I doubt it, because I think you can do a better job, and some people do a cursory job. It’s a lot of work, and I’m not going to pretend that an FMEA doesn’t require some dedicated focus. You want to do it with a team, so it’s requiring you to get together for discussion. You will find pretty quickly that you end up doing a triage of sorts. Just look at the major things we changed, and let’s think them through carefully. Its work and effort, but I think there’s only so much time people have. Those of you who have done process improvement work, I’m sure you often get frustrated with leadership or managers that are asking, “When are you going to be done? When are you going to improve this process? What are the results?” And it was given to you yesterday, so you’re trying to do the best job you can. You want this process to be perfect, but we often look at the time to get that perfection. We are always in pursuit of perfection so we might do multiple rounds. The thing about FMEA, is that you can do a first past, and you can do it again just like any other tool. You can continually come back to it, and say, “Alright we’ve tuned it up, let’s do it again.” So think of it as not a one time thing, but a never ending thing.
Karlo: Great! Our next question is, can you suggest more good ways to practice FMEA scenarios from the service industry?
Elisabeth: The best way to practice is to do it! Get together with people inside of a similar process, and brainstorm together. If you can be collocated then get a conference room, and sit down to discuss this step of the process. One thing I do with a classroom is I ask, “In your wildest dreams, how would you like to improve this process? What do you think would be the best way to really fix this? If money wasn’t an issue, if sky was the limit, you could break the bounds of time and space, then what would change? Think about this change, and what would be the ripple effects, and how could that go wrong? This would give you a nice opportunity to really expand, because if you make a change that big you probably have a really rich opportunity to think about what would happen, and what the risks would be. So if you want to practice, maybe this would be a nice avenue to take. Maybe something you can do together with other people in the same process. If the same areas of the process bother you, then find out what will be the huge change, and what could go wrong. Good question!
Karlo: Thanks Elisabeth! Our last question comes from Bryce. I notice that when using the FMEA, our team tends to rabbit hole quite a bit. Any ideas on how to limit that?
Elisabeth: You can pick the top issues and steps that you want to deal with, and then basically set a time limit. For example, you have a one hour meeting with four issues so each issue gets 15 minutes. Set someone as the timekeeper to raise everybody’s heads up to say, “Okay we’re rabbit holing here! Is it worth it? We have four other issues to address, so have we dealt with enough on this one? Do you want to expand the time on this one (which means one of them drops out, because we only have a max amount of time)? Does this mean we are now planning another whole meeting?”
When people start seeing their calendars fill up with meetings, and realizing they’re not going to be able to look at particular risks that also need to be addressed, you get people scaling it back. They begin saying, “Okay we’ve done enough on this, and we need to move to the follow up actions to see what can be done? What are the ways to fix this? Let’s move into fix phase as opposed to analysis phase.” We’re talking about classic analysis paralysis, so I think that setting a timekeeper, and setting a time frame would be helpful. Good questions!
Karlo: Awesome! That wraps up our webinar! Thanks for joining us today, and we hope you enjoyed it as much as we did, and hope you found it helpful.
If you have any feedback please share in the poll that will come up at the end of the webinar. We always design webinars based on your feedback, so if you have any suggestions please share them with us in the comments, or send us an email. Thanks everyone! Thanks Elisabeth! On behalf of our whole team here at GoLeanSixSigma.com we’re so happy you were able to join us!