WEBVTT 00:00:00.000 --> 00:00:02.430 The following content is provided under a Creative 00:00:02.430 --> 00:00:03.730 Commons license. 00:00:03.730 --> 00:00:06.060 Your support will help MIT OpenCourseWare 00:00:06.060 --> 00:00:10.090 continue to offer high-quality educational resources for free. 00:00:10.090 --> 00:00:12.690 To make a donation or to view additional materials 00:00:12.690 --> 00:00:16.560 from hundreds of MIT courses, visit MIT OpenCourseWare 00:00:16.560 --> 00:00:17.744 at ocw.mit.edu. 00:00:25.790 --> 00:00:27.860 ANNALISA L. WEIGEL: Tools and topics-- 00:00:27.860 --> 00:00:31.580 this is really the first part of two modules. 00:00:31.580 --> 00:00:33.852 The following module's going to be Six Sigma basics, 00:00:33.852 --> 00:00:35.810 and what I'm going to talk about in the quality 00:00:35.810 --> 00:00:37.400 tools and topics is going to flow 00:00:37.400 --> 00:00:39.650 in from the foundation for what Earl 00:00:39.650 --> 00:00:43.700 is going to talk to you about in the Six Sigma basics module. 00:00:43.700 --> 00:00:45.530 So at the end of this module, we expect 00:00:45.530 --> 00:00:47.570 you to be able to describe how quality 00:00:47.570 --> 00:00:49.760 is essential to lean in achieving customer 00:00:49.760 --> 00:00:51.050 satisfaction. 00:00:51.050 --> 00:00:54.290 And secondly, we want you to be able to use some basic quality 00:00:54.290 --> 00:00:56.570 tools, and we're going to spend about half 00:00:56.570 --> 00:01:00.620 of the lecture exploring these different kinds of tools. 00:01:00.620 --> 00:01:02.700 So why do we care about quality? 00:01:02.700 --> 00:01:05.930 Always have to start with that-- 00:01:05.930 --> 00:01:08.690 a lot of times the cost, of poor quality 00:01:08.690 --> 00:01:11.150 is directly measured in items like this. 00:01:11.150 --> 00:01:14.210 It's, how much scrap or rework do I have on a project? 00:01:14.210 --> 00:01:15.950 How many service calls do I have to make 00:01:15.950 --> 00:01:17.930 to fix something that's broken? 00:01:17.930 --> 00:01:20.870 What are the costs of fulfilling a warranty or other concessions 00:01:20.870 --> 00:01:23.090 that you have to make to customers when 00:01:23.090 --> 00:01:25.140 your product doesn't work? 00:01:25.140 --> 00:01:28.970 But it turns out that the hidden costs or the indirect costs 00:01:28.970 --> 00:01:31.610 of quality can be 2 to 3 times as large 00:01:31.610 --> 00:01:33.450 as your direct measured costs. 00:01:33.450 --> 00:01:36.020 So if you think the direct measured costs are large, 00:01:36.020 --> 00:01:39.050 really, what your organization is spending on poor quality 00:01:39.050 --> 00:01:41.690 is enormous, because it's much, much bigger than that. 00:01:41.690 --> 00:01:43.952 And they show up in terms of the excess inventory 00:01:43.952 --> 00:01:46.410 that you might need to keep around, because your processes, 00:01:46.410 --> 00:01:49.370 your quality is so poor over time for employees 00:01:49.370 --> 00:01:51.320 to deal with those poor quality issues-- 00:01:51.320 --> 00:01:53.210 all the non-value added and wasted 00:01:53.210 --> 00:01:56.570 steps, and queues, and delays that happen and cost money. 00:01:56.570 --> 00:01:59.570 And not the least bit is the reputation or image 00:01:59.570 --> 00:02:02.300 of the organization for having poor quality. 00:02:02.300 --> 00:02:07.400 And sometimes the cost of that is really quite immeasurable. 00:02:07.400 --> 00:02:08.642 So let's start with-- 00:02:08.642 --> 00:02:10.850 one of the common ways that people think about adding 00:02:10.850 --> 00:02:12.410 quality is with inspection. 00:02:12.410 --> 00:02:14.660 But we've already talked a little bit about inspection 00:02:14.660 --> 00:02:19.440 in the course, and how non-value added inspection really is. 00:02:19.440 --> 00:02:21.830 So if you're buying a car and you 00:02:21.830 --> 00:02:23.690 have lots of people who are inspecting it, 00:02:23.690 --> 00:02:27.370 does inspections make your car go faster? 00:02:27.370 --> 00:02:30.580 Does inspection make your car get better fuel economy? 00:02:30.580 --> 00:02:34.050 What else is important to you about a car? 00:02:34.050 --> 00:02:34.960 AUDIENCE: [INAUDIBLE] 00:02:34.960 --> 00:02:35.620 ANNALISA L. WEIGEL: Safety-- does 00:02:35.620 --> 00:02:36.912 inspection make your car safer? 00:02:39.430 --> 00:02:41.112 Some of you are nodding. 00:02:41.112 --> 00:02:42.820 OK, so some of your saying no, so there's 00:02:42.820 --> 00:02:46.090 some disagreement about if inspection increases safety. 00:02:46.090 --> 00:02:47.737 I think, after the next three slides, 00:02:47.737 --> 00:02:49.570 I will convince you that inspection does not 00:02:49.570 --> 00:02:51.220 make your car safer. 00:02:51.220 --> 00:02:53.285 Anything else you value in your car? 00:02:53.285 --> 00:02:54.160 AUDIENCE: [INAUDIBLE] 00:02:54.160 --> 00:02:55.535 ANNALISA L. WEIGEL: Gas mileage-- 00:02:55.535 --> 00:02:58.840 so does inspection make your gas mileage get better? 00:02:58.840 --> 00:03:01.630 Probably not-- the takeaway is inspection 00:03:01.630 --> 00:03:05.860 doesn't typically add to what you, the customer, value. 00:03:05.860 --> 00:03:08.920 It may help to make sure that a bad product doesn't come out 00:03:08.920 --> 00:03:11.350 of a factory, out of a service to get to a customer, 00:03:11.350 --> 00:03:13.360 but inspection doesn't help you actually 00:03:13.360 --> 00:03:15.610 build in quality in the first time, 00:03:15.610 --> 00:03:17.780 because inspection is subject to a lot of problems. 00:03:17.780 --> 00:03:20.590 There's different varying skills of inspectors 00:03:20.590 --> 00:03:23.110 and their attention span at any given point in time. 00:03:23.110 --> 00:03:25.090 There's measurement capability of humans-- 00:03:25.090 --> 00:03:26.230 varies widely. 00:03:26.230 --> 00:03:27.730 And there's environmental factors 00:03:27.730 --> 00:03:31.300 as well that contribute to inspection having widely 00:03:31.300 --> 00:03:34.330 performing characteristics. 00:03:34.330 --> 00:03:35.170 Yes? 00:03:35.170 --> 00:03:38.950 AUDIENCE: [INAUDIBLE] inspection from the point 00:03:38.950 --> 00:03:42.920 of view of the owner of a car, where really, inspection 00:03:42.920 --> 00:03:49.060 is part of a system of the quality of the safety 00:03:49.060 --> 00:03:50.566 of our communities. 00:03:53.830 --> 00:03:56.440 We are stakeholders because we own a car 00:03:56.440 --> 00:03:58.870 and we have to have it inspected, but it's not-- 00:03:58.870 --> 00:04:00.310 it wasn't necessary there. 00:04:00.310 --> 00:04:00.977 The inspection-- 00:04:00.977 --> 00:04:02.435 ANNALISA L. WEIGEL: Are you talking 00:04:02.435 --> 00:04:04.360 about the state-mandated inspections? 00:04:04.360 --> 00:04:05.060 AUDIENCE: Yeah. 00:04:05.060 --> 00:04:07.620 ANNALISA L. WEIGEL: Oh, I'm sorry. 00:04:07.620 --> 00:04:08.620 I should have clarified. 00:04:08.620 --> 00:04:10.480 So I'm thinking about inspection like you're 00:04:10.480 --> 00:04:15.470 at the Ford or the GM plant, and they're manufacturing a car. 00:04:15.470 --> 00:04:15.970 Yep. 00:04:15.970 --> 00:04:18.550 So I'll note that for the future. 00:04:18.550 --> 00:04:21.640 Maybe that example's not the best one to take, but yes-- 00:04:21.640 --> 00:04:24.813 or think about a factory making a television. 00:04:24.813 --> 00:04:26.230 The fact that there are inspectors 00:04:26.230 --> 00:04:28.090 all along the assembly line doesn't 00:04:28.090 --> 00:04:30.400 mean you get a better picture quality on your TV, 00:04:30.400 --> 00:04:31.217 for example. 00:04:31.217 --> 00:04:32.800 Or it doesn't mean that it costs less. 00:04:32.800 --> 00:04:35.230 Cost might be a value that you have for your TV. 00:04:35.230 --> 00:04:37.235 Yeah, state inspections-- we'll talk 00:04:37.235 --> 00:04:38.360 about that in a little bit. 00:04:38.360 --> 00:04:40.120 And it turns out, when state inspections are done, 00:04:40.120 --> 00:04:42.310 they're actually not visually inspecting much. 00:04:42.310 --> 00:04:44.020 They're actually doing inspections 00:04:44.020 --> 00:04:47.410 that have instrumentations that are taking quantitative data. 00:04:47.410 --> 00:04:49.390 We do call that inspection, but what 00:04:49.390 --> 00:04:52.990 we mean here is really visual inspection, eyeballs 00:04:52.990 --> 00:04:54.850 on some kind of part of the process 00:04:54.850 --> 00:04:58.680 to make sure that the product is turned out right. 00:04:58.680 --> 00:05:00.670 So just to demonstrate to you some 00:05:00.670 --> 00:05:03.070 of the challenges involved with inspections, 00:05:03.070 --> 00:05:04.190 we have an exercise. 00:05:04.190 --> 00:05:05.950 So if you can take out your folders, 00:05:05.950 --> 00:05:09.430 there's an inspection exercise in there, 00:05:09.430 --> 00:05:12.250 and it's a bunch of prose. 00:05:12.250 --> 00:05:13.790 So it looks like this. 00:05:13.790 --> 00:05:15.880 It is entitled inspection exercise. 00:05:15.880 --> 00:05:17.590 So if you can take it out and turn 00:05:17.590 --> 00:05:19.690 it face down in front of you-- 00:05:19.690 --> 00:05:24.730 so take your text, flip it over face down so you don't see it, 00:05:24.730 --> 00:05:27.110 and take your pen in your hand, and get ready. 00:05:27.110 --> 00:05:31.360 So the exercise is going to be in two 30-second sessions. 00:05:31.360 --> 00:05:33.910 So your first task is to find all 00:05:33.910 --> 00:05:39.190 of the lowercase or uppercase Fs on a page of text. 00:05:39.190 --> 00:05:40.450 So you are an inspector. 00:05:40.450 --> 00:05:42.490 You are the F inspector. 00:05:42.490 --> 00:05:44.380 So take out that inspection exercise. 00:05:44.380 --> 00:05:46.070 Turn it face down. 00:05:46.070 --> 00:05:49.150 When I say go, you are going to circle all the lower 00:05:49.150 --> 00:05:51.820 and uppercase Fs that you find. 00:05:51.820 --> 00:05:55.750 On stop, I want you to turn your sheet over, OK? 00:05:55.750 --> 00:05:57.680 And we're going to pass it to the right. 00:05:57.680 --> 00:05:58.660 So everybody clear? 00:05:58.660 --> 00:06:01.630 I say go, you flip your page over, you circle-- 00:06:01.630 --> 00:06:03.010 it's important that you circle. 00:06:03.010 --> 00:06:05.200 Circle all the Fs, lower and uppercase. 00:06:05.200 --> 00:06:05.960 Ready? 00:06:05.960 --> 00:06:06.460 Go. 00:06:11.090 --> 00:06:12.950 And stop. 00:06:12.950 --> 00:06:14.960 Turn your paper over. 00:06:14.960 --> 00:06:18.680 Pass the sheet to the person on your right. 00:06:18.680 --> 00:06:21.380 Yeah, turn it face down. 00:06:21.380 --> 00:06:23.930 All right, this is the second inspection. 00:06:23.930 --> 00:06:26.480 So now on the paper, when you turn it over, 00:06:26.480 --> 00:06:29.090 you're going to mark Xs on all the other Fs 00:06:29.090 --> 00:06:33.650 you find that were not found the first time. 00:06:33.650 --> 00:06:34.220 So ready? 00:06:34.220 --> 00:06:35.087 You have 30 seconds. 00:06:35.087 --> 00:06:36.170 On your mark, get set, go. 00:06:43.880 --> 00:06:45.770 And stop. 00:06:45.770 --> 00:06:48.290 All right, now, take the page that you have, 00:06:48.290 --> 00:06:50.750 and I want you to count all of the Fs 00:06:50.750 --> 00:06:53.090 that you placed an X or a circle around. 00:06:53.090 --> 00:06:56.690 So tally up all the Fs that you found on the sheet-- both you 00:06:56.690 --> 00:06:59.600 and the person previous to you. 00:06:59.600 --> 00:07:02.840 And please make sure you tally separately the circles 00:07:02.840 --> 00:07:04.790 and the Xs. 00:07:04.790 --> 00:07:06.770 All right, when you have tallied your Xs, 00:07:06.770 --> 00:07:10.300 I'd like to get the number of Fs that were circled. 00:07:10.300 --> 00:07:12.050 And we're going to just go around the room 00:07:12.050 --> 00:07:14.330 and just shout out the number of circled Fs 00:07:14.330 --> 00:07:15.560 that you have in turn. 00:07:15.560 --> 00:07:17.910 I'm going to take them down here on the computer when 00:07:17.910 --> 00:07:18.410 you do this. 00:07:25.182 --> 00:07:25.890 And the next one? 00:07:25.890 --> 00:07:26.770 AUDIENCE: 31-- 00:07:26.770 --> 00:07:27.770 ANNALISA L. WEIGEL: 31-- 00:07:27.770 --> 00:07:28.790 oh, that's a record now. 00:07:28.790 --> 00:07:29.290 Go ahead. 00:07:29.290 --> 00:07:31.882 AUDIENCE: [INAUDIBLE] 00:07:31.882 --> 00:07:32.840 ANNALISA L. WEIGEL: 36? 00:07:32.840 --> 00:07:35.050 AUDIENCE: [INAUDIBLE] 00:07:35.200 --> 00:07:37.700 ANNALISA L. WEIGEL: So 36 is a data point you'd like to add? 00:07:37.700 --> 00:07:41.038 Well, that's a pretty high-quality inspector. 00:07:41.038 --> 00:07:42.080 That one definitely wins. 00:07:42.080 --> 00:07:43.920 All right, let's go on past two. 00:07:43.920 --> 00:07:47.480 So now what I'd like to know is the number of Xs 00:07:47.480 --> 00:07:48.823 that were caught. 00:07:48.823 --> 00:07:49.323 10? 00:07:49.323 --> 00:07:53.130 10-- OK, great. 00:07:53.130 --> 00:07:55.770 All right, so let's take a look. 00:07:55.770 --> 00:08:02.010 So this is a graph that shows us how many Fs we caught 00:08:02.010 --> 00:08:04.080 on our first pass on our second pass. 00:08:04.080 --> 00:08:08.340 The average is in the red. 00:08:08.340 --> 00:08:10.920 And then we also have the standard deviation 00:08:10.920 --> 00:08:13.020 that's shown, so you can see the kind of variation 00:08:13.020 --> 00:08:16.350 that we have between all of you highly-qualified inspectors 00:08:16.350 --> 00:08:16.947 in the room. 00:08:16.947 --> 00:08:19.530 There are still some significant variation in the number of Fs 00:08:19.530 --> 00:08:20.790 that we can catch. 00:08:20.790 --> 00:08:23.790 And then the blue column, which you see displayed in pass 2, 00:08:23.790 --> 00:08:26.250 is the projected average of Fs that 00:08:26.250 --> 00:08:30.570 might be caught based on the percentage of Fs that 00:08:30.570 --> 00:08:32.850 was caught on the first pass and the second pass. 00:08:32.850 --> 00:08:34.470 And we could keep calculating that, 00:08:34.470 --> 00:08:36.929 if we made third and fourth passes, and so on. 00:08:36.929 --> 00:08:39.179 AUDIENCE: Is that average the total number 00:08:39.179 --> 00:08:43.559 of Fs that would be caught within two passes-- so 00:08:43.559 --> 00:08:45.900 [INAUDIBLE]? 00:08:45.900 --> 00:08:46.747 Or is that-- 00:08:46.747 --> 00:08:48.330 ANNALISA L. WEIGEL: That's total, yes. 00:08:48.330 --> 00:08:50.640 So that's not this-- the number caught on the second. 00:08:50.640 --> 00:08:51.360 That's the total. 00:08:55.930 --> 00:09:01.000 All right, so now let's go ahead and look at our summaries here. 00:09:01.000 --> 00:09:05.650 So our inspector efficiency is running around 68%. 00:09:05.650 --> 00:09:08.170 All right, so let's take that inspector efficiency 00:09:08.170 --> 00:09:12.030 and then return back to our graphs here. 00:09:12.030 --> 00:09:14.200 So there's our inspection exercise. 00:09:14.200 --> 00:09:17.380 This is a graph that shows you the impact of inspector 00:09:17.380 --> 00:09:19.970 efficiency on escaped quality. 00:09:19.970 --> 00:09:25.360 So here on the y-axis, we have escape defects per million, 00:09:25.360 --> 00:09:28.510 and down here on the x-axis is the number 00:09:28.510 --> 00:09:30.710 of consecutive inspectors. 00:09:30.710 --> 00:09:32.440 So we had just did an example with two 00:09:32.440 --> 00:09:34.480 consecutive inspectors. 00:09:34.480 --> 00:09:36.670 And then here, our different color lines 00:09:36.670 --> 00:09:40.000 represent different levels of Inspector efficiency. 00:09:40.000 --> 00:09:44.110 So 70% is with the red line, so that's about the line 00:09:44.110 --> 00:09:46.130 that our class was on-- 00:09:46.130 --> 00:09:49.100 68% efficiency-- let's call it 70. 00:09:49.100 --> 00:09:51.160 So if we wanted to inspect-- 00:09:51.160 --> 00:09:53.290 visually inspect our product down 00:09:53.290 --> 00:09:59.040 to one escaped defect per million, 00:09:59.040 --> 00:10:05.270 we have to have more than 11 consecutive inspectors 00:10:05.270 --> 00:10:07.490 examining each product. 00:10:07.490 --> 00:10:09.170 Do you start to see how that adds up 00:10:09.170 --> 00:10:13.723 to a lot of time for maybe not the biggest payoff? 00:10:13.723 --> 00:10:15.640 You need to have lots and lots of inspections, 00:10:15.640 --> 00:10:17.800 and what did we say about inspection? 00:10:17.800 --> 00:10:19.100 It's non-value added. 00:10:19.100 --> 00:10:21.040 It's really a waste in the process 00:10:21.040 --> 00:10:22.930 to do visual inspection. 00:10:22.930 --> 00:10:25.900 You want to be able to build in your quality up front. 00:10:25.900 --> 00:10:28.750 You want to have consistent processes that 00:10:28.750 --> 00:10:30.610 deliver the kind of quality you want, 00:10:30.610 --> 00:10:34.220 and not try to inspect it in later. 00:10:34.220 --> 00:10:36.760 So that's just a motivation for why 00:10:36.760 --> 00:10:38.800 we need to think about quality tools, 00:10:38.800 --> 00:10:40.660 and building them into all the processes 00:10:40.660 --> 00:10:44.380 that we use to accomplish our tasks. 00:10:44.380 --> 00:10:46.000 And lean uses a number of them. 00:10:46.000 --> 00:10:48.940 They came out of the foundation of total quality management, 00:10:48.940 --> 00:10:51.640 which became really big in the 1980s, 00:10:51.640 --> 00:10:53.320 and also emerged from Japan. 00:10:53.320 --> 00:10:55.240 And TQM, as it was called, was built 00:10:55.240 --> 00:10:57.740 on the notion of customer satisfaction, 00:10:57.740 --> 00:10:59.950 and we achieved that through employee involvement 00:10:59.950 --> 00:11:01.630 and continuous improvement. 00:11:01.630 --> 00:11:04.930 And then those were all enabled by focusing on some problem 00:11:04.930 --> 00:11:08.740 solving tools, designing processes, purchasing 00:11:08.740 --> 00:11:11.478 efficiently, and thinking about your product service design, 00:11:11.478 --> 00:11:14.020 as well as benchmarking-- going out and looking at how you're 00:11:14.020 --> 00:11:16.780 doing relative to your peers. 00:11:16.780 --> 00:11:19.180 But lean, on the concepts we've looked at in class, 00:11:19.180 --> 00:11:21.550 are very much related to all of these. 00:11:21.550 --> 00:11:24.220 We talked about value stream mapping analysis 00:11:24.220 --> 00:11:26.718 as really analyzing the process design part of it. 00:11:26.718 --> 00:11:29.260 We talked, at least yesterday in the enterprise [INAUDIBLE],, 00:11:29.260 --> 00:11:31.840 about lean supply chains as being the way 00:11:31.840 --> 00:11:34.180 that we're looking at purchasing-- and lean product 00:11:34.180 --> 00:11:36.190 development or lean health care deployment 00:11:36.190 --> 00:11:39.010 as a focus on the product and service design. 00:11:39.010 --> 00:11:40.510 And today, in the module here, we're 00:11:40.510 --> 00:11:43.570 going to talk about seven very common quality tools that 00:11:43.570 --> 00:11:47.930 help you address the problem of troublesome quality. 00:11:47.930 --> 00:11:50.280 So the seven basic quality tools are these. 00:11:50.280 --> 00:11:51.740 We'll start with flowcharts. 00:11:51.740 --> 00:11:54.230 Then there's some cause and effect diagrams-- 00:11:54.230 --> 00:11:56.540 example of which is shown here on the chart. 00:11:56.540 --> 00:12:00.200 Then there's check sheets or checklists, histograms, Pareto 00:12:00.200 --> 00:12:02.120 charts, scatter diagrams. 00:12:02.120 --> 00:12:04.580 And then control charts we won't cover in this module, 00:12:04.580 --> 00:12:07.520 but Earl will be covering it in the next module on Six Sigma 00:12:07.520 --> 00:12:09.370 basics. 00:12:09.370 --> 00:12:11.470 So stepping right into our first tool, 00:12:11.470 --> 00:12:14.380 flowcharts, examples of these are 00:12:14.380 --> 00:12:16.360 the process maps that we've already done, 00:12:16.360 --> 00:12:17.560 the value stream maps. 00:12:17.560 --> 00:12:20.740 A flow chart is which activities are taking place in sequence, 00:12:20.740 --> 00:12:23.740 and the inputs of one flow into the process become 00:12:23.740 --> 00:12:27.370 outputs, which are inputs into another part of the flow chart. 00:12:27.370 --> 00:12:31.270 You also see these commonly as describing software program 00:12:31.270 --> 00:12:33.850 flows, and sometimes [INAUDIBLE] algorithms 00:12:33.850 --> 00:12:37.330 are also described with a process and a flow chart. 00:12:37.330 --> 00:12:39.435 So why are these a quality tool? 00:12:39.435 --> 00:12:40.810 There are a number of reasons why 00:12:40.810 --> 00:12:43.240 we think about these is helping to improve the product 00:12:43.240 --> 00:12:44.770 or service quality. 00:12:44.770 --> 00:12:47.020 This kind of visual description that a flow chart 00:12:47.020 --> 00:12:49.660 gives you really helps to improve comprehension 00:12:49.660 --> 00:12:52.120 among a large group of people, because it's laying out 00:12:52.120 --> 00:12:55.600 there visually those steps of the process takes. 00:12:55.600 --> 00:12:58.720 And in addition, it becomes a flow, 00:12:58.720 --> 00:13:00.190 and it helps us make sure that we 00:13:00.190 --> 00:13:01.930 are doing each step of the process that's 00:13:01.930 --> 00:13:03.670 required in the order it's required 00:13:03.670 --> 00:13:04.750 and when it's required. 00:13:04.750 --> 00:13:07.360 So it becomes that tool for us to make sure that we've 00:13:07.360 --> 00:13:08.890 done the right thing. 00:13:08.890 --> 00:13:11.860 It also helps to tie our outputs to the inputs 00:13:11.860 --> 00:13:14.140 of different steps in the flow chart, 00:13:14.140 --> 00:13:16.420 and it can also assist us with data collection 00:13:16.420 --> 00:13:19.420 by helping us understand where in the process flow 00:13:19.420 --> 00:13:22.040 we need to collect and add data. 00:13:22.040 --> 00:13:25.360 The second basic quality tool are cause and effect diagrams. 00:13:25.360 --> 00:13:27.820 These are also called fishbone diagrams, 00:13:27.820 --> 00:13:31.330 or Ishikawa diagrams after the gentleman, Mr. Ishikawa, 00:13:31.330 --> 00:13:33.730 who invented these. 00:13:33.730 --> 00:13:36.040 And they look a lot like a fishbone. 00:13:36.040 --> 00:13:38.890 And it's a tool for root cause analysis, 00:13:38.890 --> 00:13:42.430 and it's often supported by using the five Ys exercise. 00:13:42.430 --> 00:13:44.920 And this is really to help you with brainstorming 00:13:44.920 --> 00:13:47.500 all the potential causes or things 00:13:47.500 --> 00:13:49.310 that might affect a problem. 00:13:49.310 --> 00:13:51.550 So we think about the problem as being represented 00:13:51.550 --> 00:13:55.840 on this horizontal line here, and then there are usually 00:13:55.840 --> 00:13:59.290 six categories of causes to help with your brainstorming 00:13:59.290 --> 00:14:00.830 to try to fill those in. 00:14:00.830 --> 00:14:04.240 So you want to think about causes related to materials, 00:14:04.240 --> 00:14:08.380 or causes related to personnel, or to measurements, 00:14:08.380 --> 00:14:11.380 or to methods, or to the environment-- 00:14:11.380 --> 00:14:13.090 or external factors-- 00:14:13.090 --> 00:14:14.980 or to machines. 00:14:14.980 --> 00:14:18.010 Now, not every discipline uses these exact same six, 00:14:18.010 --> 00:14:19.960 but they're a great starting point for most. 00:14:19.960 --> 00:14:22.540 And some organizations and disciplines 00:14:22.540 --> 00:14:25.690 have determined their own particular veining structure 00:14:25.690 --> 00:14:28.300 that they like for their fishbone diagrams. 00:14:28.300 --> 00:14:30.730 And then, as you see the examples I'll show you next, 00:14:30.730 --> 00:14:32.663 there are some primary causes that 00:14:32.663 --> 00:14:34.330 relate to materials, which are typically 00:14:34.330 --> 00:14:37.450 shown on the horizontal axes feeding into these spines. 00:14:37.450 --> 00:14:39.490 And then there are sometimes secondary causes 00:14:39.490 --> 00:14:41.630 that feed into a primary cause. 00:14:41.630 --> 00:14:42.880 So let me show you an example. 00:14:42.880 --> 00:14:47.000 I think that's easier way to understand how these are used. 00:14:47.000 --> 00:14:50.080 So this is an example from the Institute for Health Care 00:14:50.080 --> 00:14:51.850 Improvement, and it's a cause and effect 00:14:51.850 --> 00:14:56.260 diagram about why tests take so long to get results. 00:14:56.260 --> 00:14:58.840 And they've used the categories for the main causes 00:14:58.840 --> 00:15:02.950 of people, environment, equipment, methods, 00:15:02.950 --> 00:15:05.570 and materials. 00:15:05.570 --> 00:15:09.280 And so if we wanted to ask, well, why might materials make 00:15:09.280 --> 00:15:11.740 test times take so long, it might 00:15:11.740 --> 00:15:15.460 be because there are lab supplies as part of materials, 00:15:15.460 --> 00:15:17.890 and those lab supplies-- this is a secondary cause 00:15:17.890 --> 00:15:20.800 illustration may be unavailable or they may 00:15:20.800 --> 00:15:24.520 be spoiled and no longer good. 00:15:24.520 --> 00:15:29.290 We may think about the methods of why tests take too long. 00:15:29.290 --> 00:15:31.790 So what is it about the method that I'm doing with the test? 00:15:31.790 --> 00:15:33.670 Maybe there are too many people involved. 00:15:33.670 --> 00:15:37.490 Maybe there are unnecessary steps in my method. 00:15:37.490 --> 00:15:40.570 So it's not a definitive answer. 00:15:40.570 --> 00:15:42.220 This is the brainstorming tool that you 00:15:42.220 --> 00:15:45.250 use to get all of your thoughts flowing about, 00:15:45.250 --> 00:15:48.220 what might be all the potential causes that contribute 00:15:48.220 --> 00:15:51.260 to a particular problem? 00:15:51.260 --> 00:15:53.510 All right, another tool are check sheets, 00:15:53.510 --> 00:15:56.510 or a simplified version of these are checklists. 00:15:56.510 --> 00:15:58.430 But a check sheet is just a structured tool 00:15:58.430 --> 00:15:59.850 for collecting data. 00:15:59.850 --> 00:16:02.780 So we'll put up for you this purely hypothetical example 00:16:02.780 --> 00:16:05.630 of keeping track of the instances of quality problems 00:16:05.630 --> 00:16:07.470 in our lean academy courses. 00:16:07.470 --> 00:16:11.120 So we've identified these various problems here, 00:16:11.120 --> 00:16:13.760 on the column under problem, and then 00:16:13.760 --> 00:16:16.040 we're trying to keep track of all the instances we see 00:16:16.040 --> 00:16:18.260 on Monday, on Tuesday, on Wednesday, 00:16:18.260 --> 00:16:20.510 and we're totaling them for each of the days 00:16:20.510 --> 00:16:22.830 and for each of the instances. 00:16:22.830 --> 00:16:24.560 So this helps us know what kind of data 00:16:24.560 --> 00:16:27.440 to collect so we can monitor the quality of the process 00:16:27.440 --> 00:16:30.470 and understand where we might want to make some improvements. 00:16:30.470 --> 00:16:33.620 So I said a simplified version of this is just a checklist. 00:16:33.620 --> 00:16:36.230 Pilots often use these when they're getting ready to fly. 00:16:36.230 --> 00:16:38.510 It's a structured list of exactly what you're 00:16:38.510 --> 00:16:41.570 supposed to be doing, and each time you accomplish the item, 00:16:41.570 --> 00:16:43.820 you check that you've done it, and then 00:16:43.820 --> 00:16:46.460 you place another check on the other activity. 00:16:46.460 --> 00:16:50.030 When my first child was born and I found myself 00:16:50.030 --> 00:16:52.610 in a sleep-deprived state, like most parents do, 00:16:52.610 --> 00:16:55.260 my husband and I were trying to get out the door with our son, 00:16:55.260 --> 00:16:56.630 and we kept forgetting things. 00:16:56.630 --> 00:16:58.160 And of course, we were tired and we really 00:16:58.160 --> 00:17:00.380 didn't have the mental capacity to remember everything, 00:17:00.380 --> 00:17:02.463 but both being trained in lean, we said, you know, 00:17:02.463 --> 00:17:04.160 we need a checklist. 00:17:04.160 --> 00:17:07.260 And we made a checklist of all the items he needed-- diapers, 00:17:07.260 --> 00:17:10.369 wipes, toys, sippy cups, food-- whatever-- 00:17:10.369 --> 00:17:12.079 and we kept them right next to our door, 00:17:12.079 --> 00:17:14.300 and every time we left, we picked up a checklist-- 00:17:14.300 --> 00:17:15.482 being brain-dead parents-- 00:17:15.482 --> 00:17:17.190 and all I had to do was read it, and say, 00:17:17.190 --> 00:17:18.319 did I have it in my bag? 00:17:18.319 --> 00:17:19.550 And I checked it off. 00:17:19.550 --> 00:17:23.630 And after that, we were happier, happier families-- 00:17:23.630 --> 00:17:25.640 so yet another instance of how lean in the home 00:17:25.640 --> 00:17:28.460 is very helpful. 00:17:28.460 --> 00:17:30.900 All right, next tool is a histogram. 00:17:30.900 --> 00:17:32.390 So histograms are typically showing 00:17:32.390 --> 00:17:35.500 the frequency of occurrence of a particular activity or item. 00:17:35.500 --> 00:17:38.000 And from this, you can start to understand a little bit more 00:17:38.000 --> 00:17:40.460 about the frequency distribution that you're seeing-- 00:17:40.460 --> 00:17:42.650 is this a normally distributed event, 00:17:42.650 --> 00:17:45.270 is it randomly distributed, and so on. 00:17:45.270 --> 00:17:48.620 So we can also use it to see how a distribution is spread out-- 00:17:48.620 --> 00:17:50.870 whether it's very concentrated or very spread. 00:17:50.870 --> 00:17:55.550 We can also use histograms to see changes over time. 00:17:55.550 --> 00:17:58.920 We can look at inputs and outputs in a quantitative way, 00:17:58.920 --> 00:18:00.770 and we can also use histograms to help 00:18:00.770 --> 00:18:02.510 us compare what we need to be doing 00:18:02.510 --> 00:18:05.690 to what the customer requires us to be doing. 00:18:05.690 --> 00:18:08.300 And here's where we get to the sweet part of the lecture, 00:18:08.300 --> 00:18:09.440 or about midway through. 00:18:09.440 --> 00:18:11.930 And we're going to do an exercise involving some M&M's. 00:18:11.930 --> 00:18:14.630 And I'll ask Q if you can start to distribute the packages 00:18:14.630 --> 00:18:15.670 of M&Ms. 00:18:15.670 --> 00:18:16.837 AUDIENCE: It's already done. 00:18:16.837 --> 00:18:18.628 ANNALISA L. WEIGEL: Oh, excellent-- they're 00:18:18.628 --> 00:18:19.374 here-- great. 00:18:19.374 --> 00:18:22.100 AUDIENCE: I hope nobody's actually eaten them already. 00:18:22.100 --> 00:18:23.645 [INTERPOSING VOICES] 00:18:26.518 --> 00:18:29.060 ANNALISA L. WEIGEL: All right, so what we'd like you to do is 00:18:29.060 --> 00:18:35.240 your bag of M&Ms, but don't eat them yet. 00:18:35.240 --> 00:18:38.090 We want you to count the number of M&Ms in your bag, 00:18:38.090 --> 00:18:40.860 and we'd like you to do that by color. 00:18:40.860 --> 00:18:43.160 So arrange all the reds, and the blues, 00:18:43.160 --> 00:18:46.610 and the greens, and the yellows, and the browns all 00:18:46.610 --> 00:18:49.700 in different categories, and then count them up. 00:18:49.700 --> 00:18:52.400 And what you want to do then is take the data sheet 00:18:52.400 --> 00:18:53.630 that is on your table-- 00:18:53.630 --> 00:18:55.040 did everybody find the data sheet 00:18:55.040 --> 00:18:58.790 underneath your M&M packages? 00:18:58.790 --> 00:19:02.150 Once you have your M&M bag tallied, please 00:19:02.150 --> 00:19:05.810 fill in the data on the data sheet. 00:19:05.810 --> 00:19:06.740 So go to it. 00:19:12.920 --> 00:19:15.350 Once your table has filled in that sheet, 00:19:15.350 --> 00:19:16.790 please bring it up from. 00:19:23.310 --> 00:19:25.980 All right, if you can direct your attention up to the front 00:19:25.980 --> 00:19:33.360 of the class, so this is a histogram of the number of M&Ms 00:19:33.360 --> 00:19:35.850 that are in all of your bags. 00:19:35.850 --> 00:19:39.570 So we have the quantity of M&Ms that were found in your bags, 00:19:39.570 --> 00:19:42.930 and we have the frequency of occurrence of those particular 00:19:42.930 --> 00:19:44.910 M&Ms in your bag. 00:19:44.910 --> 00:19:48.150 So let me also ask you to look around at your table 00:19:48.150 --> 00:19:53.020 and look at the color distributions that you see. 00:19:53.020 --> 00:19:54.970 What do you hazard about those? 00:19:54.970 --> 00:19:57.990 What do you learn from looking around at the distributions? 00:19:57.990 --> 00:20:00.725 AUDIENCE: [INAUDIBLE] 00:20:00.725 --> 00:20:02.850 ANNALISA L. WEIGEL: People are shaking their heads. 00:20:02.850 --> 00:20:03.683 What does that mean? 00:20:03.683 --> 00:20:06.856 [INTERPOSING VOICES] 00:20:07.630 --> 00:20:08.410 Not many reds? 00:20:08.410 --> 00:20:10.840 So maybe red is expensive to produce. 00:20:10.840 --> 00:20:11.590 All right. 00:20:11.590 --> 00:20:12.736 What else? 00:20:12.736 --> 00:20:15.892 AUDIENCE: [INAUDIBLE] 00:20:15.892 --> 00:20:18.225 ANNALISA L. WEIGEL: Do you see an even number of colors? 00:20:18.225 --> 00:20:19.513 AUDIENCE: No. 00:20:19.513 --> 00:20:21.180 ANNALISA L. WEIGEL: Not within your bag? 00:20:21.180 --> 00:20:23.230 Even if you took a big sample on your table? 00:20:23.230 --> 00:20:24.060 AUDIENCE: No. 00:20:24.060 --> 00:20:25.200 ANNALISA L. WEIGEL: Still not the same? 00:20:25.200 --> 00:20:26.100 AUDIENCE: [INAUDIBLE] 00:20:26.100 --> 00:20:28.558 ANNALISA L. WEIGEL: And then, what about the number of M&Ms 00:20:28.558 --> 00:20:30.390 that are in a bag-- 00:20:30.390 --> 00:20:31.627 the same, variable? 00:20:31.627 --> 00:20:33.880 [INTERPOSING VOICES] 00:20:34.530 --> 00:20:36.730 ANNALISA L. WEIGEL: Kind of variable-- 00:20:36.730 --> 00:20:40.800 so if we're thinking about quality and we're thinking 00:20:40.800 --> 00:20:45.780 about the M&M attribute that you think Mars company-- 00:20:45.780 --> 00:20:48.793 which is [INAUDIBLE] the company that makes M&Ms-- 00:20:48.793 --> 00:20:50.460 what attribute do you think that they're 00:20:50.460 --> 00:20:54.418 applying quality principles to control the variation of? 00:20:54.418 --> 00:20:56.460 So do you think they're controlling the variation 00:20:56.460 --> 00:20:59.280 of the total M&Ms in a bag? 00:20:59.280 --> 00:21:00.900 Not with that big a variation-- how 00:21:00.900 --> 00:21:02.910 about the color distribution? 00:21:02.910 --> 00:21:07.950 Not that-- so is it something else? 00:21:07.950 --> 00:21:09.460 I won't accept that as an answer. 00:21:09.460 --> 00:21:11.140 What do you think it might be? 00:21:11.140 --> 00:21:11.910 AUDIENCE: Weight-- 00:21:11.910 --> 00:21:12.480 ANNALISA L. WEIGEL: Weight-- 00:21:12.480 --> 00:21:13.605 what does say in your bags? 00:21:17.850 --> 00:21:19.470 AUDIENCE: 1.69 ounces-- 00:21:19.470 --> 00:21:22.410 ANNALISA L. WEIGEL: 1.69 ounces-- 00:21:22.410 --> 00:21:24.630 yeah, and that's exactly what they're controlling. 00:21:24.630 --> 00:21:26.760 They're not only saying, you get about an ounce or about 00:21:26.760 --> 00:21:27.593 an ounce and a half. 00:21:27.593 --> 00:21:31.740 They're saying, you get exactly 1.69 ounces of M&Ms 00:21:31.740 --> 00:21:32.520 in your bag. 00:21:32.520 --> 00:21:36.150 So that's a variable that they're controlling for. 00:21:36.150 --> 00:21:37.830 So we can use a histogram to learn 00:21:37.830 --> 00:21:41.730 a lot about the variations and about the distributions 00:21:41.730 --> 00:21:45.540 that we can see in our process, in our data. 00:21:45.540 --> 00:21:48.540 All right, so let's continue with another example, a Pareto 00:21:48.540 --> 00:21:49.410 chart. 00:21:49.410 --> 00:21:50.480 [INTERPOSING VOICES] 00:21:50.480 --> 00:21:52.170 AUDIENCE: Can we now have it? 00:21:52.170 --> 00:21:55.370 ANNALISA L. WEIGEL: Yes, you may now eat. 00:21:55.370 --> 00:21:56.870 I forgot to give you the go-ahead. 00:21:56.870 --> 00:22:01.490 Yes, munch away on your M&Ms. We want the sugar level to raise 00:22:01.490 --> 00:22:04.190 up so that you stay active and functioning for the rest 00:22:04.190 --> 00:22:05.760 of the afternoon. 00:22:05.760 --> 00:22:08.480 So just to give you maybe a more serious example of a Pareto 00:22:08.480 --> 00:22:12.140 chart, these are causes leading to emergency department 00:22:12.140 --> 00:22:12.950 admissions. 00:22:12.950 --> 00:22:14.310 So we start here. 00:22:14.310 --> 00:22:17.060 And this is called a Pareto, because we have the tallest 00:22:17.060 --> 00:22:19.040 bar starting in the front, and working 00:22:19.040 --> 00:22:20.900 our way down to lower bars. 00:22:20.900 --> 00:22:23.150 And what Pareto charts do are call our attention 00:22:23.150 --> 00:22:24.950 to the things that are happening the most. 00:22:24.950 --> 00:22:26.900 So what this chart says is pneumonia, 00:22:26.900 --> 00:22:28.760 in this particular emergency department, 00:22:28.760 --> 00:22:30.920 is a leading cause of admissions, 00:22:30.920 --> 00:22:34.040 and accounts for almost 25% of those admissions. 00:22:34.040 --> 00:22:37.170 And UTIs are next, urinary tract infections. 00:22:37.170 --> 00:22:39.680 So if you wanted to think about addressing 00:22:39.680 --> 00:22:41.300 some of the causes of admissions, 00:22:41.300 --> 00:22:44.000 then you might really look at these leading categories. 00:22:44.000 --> 00:22:47.150 So you might not want to think so much about cellulitis 00:22:47.150 --> 00:22:49.700 or hypertension, because those aren't your leading folks who 00:22:49.700 --> 00:22:50.880 are coming into the ER-- 00:22:50.880 --> 00:22:51.380 ED. 00:22:51.380 --> 00:22:53.570 Maybe you want to think about these areas. 00:22:53.570 --> 00:22:56.090 And similarly, this line up here tells you 00:22:56.090 --> 00:23:00.950 the cumulative percentage, as you add in each of the columns 00:23:00.950 --> 00:23:02.570 as you go along your Pareto chart. 00:23:02.570 --> 00:23:07.970 So your pneumonia and UTI account for a total percentage 00:23:07.970 --> 00:23:14.500 of about 75% of your cases. 00:23:14.500 --> 00:23:17.440 Just another example from the aerospace world-- this 00:23:17.440 --> 00:23:19.360 is a similar kind of Pareto chart, 00:23:19.360 --> 00:23:22.690 but done for the root cause of various discrepancies 00:23:22.690 --> 00:23:23.355 on satellites. 00:23:23.355 --> 00:23:24.730 This actually comes from research 00:23:24.730 --> 00:23:26.440 that I did as a master's student. 00:23:26.440 --> 00:23:29.470 And we were trying to understand exactly what this kind of chart 00:23:29.470 --> 00:23:31.480 tells us, which is, if I wanted to eliminate 00:23:31.480 --> 00:23:34.300 the waste in my testing, where do I start? 00:23:34.300 --> 00:23:36.670 And we found that the biggest problems involved 00:23:36.670 --> 00:23:38.050 employees or operators. 00:23:38.050 --> 00:23:39.970 So there may be issues of training or skills 00:23:39.970 --> 00:23:43.600 here, and that was the non-satellite-related causes. 00:23:43.600 --> 00:23:46.210 And then we called a category of satellite-related causes, 00:23:46.210 --> 00:23:49.060 and it turned out that this basic design of the satellite 00:23:49.060 --> 00:23:51.710 was at fault most of the time. 00:23:51.710 --> 00:23:54.160 So if you wanted to try to eliminate some of those causes, 00:23:54.160 --> 00:23:56.260 you might start with perhaps training, 00:23:56.260 --> 00:23:58.240 and then thinking about design that 00:23:58.240 --> 00:24:01.360 might ensure more robust performance under the testing 00:24:01.360 --> 00:24:04.090 conditions. 00:24:04.090 --> 00:24:06.510 Next quality tool, scatter diagrams-- 00:24:06.510 --> 00:24:10.530 these are just plots of x, y pairs of numerical data. 00:24:10.530 --> 00:24:12.660 And what we can see from here is whether there's 00:24:12.660 --> 00:24:15.108 any kind of correlation or pattern-- 00:24:15.108 --> 00:24:17.400 because correlation and pattern are not the same thing. 00:24:17.400 --> 00:24:18.942 Correlation indicates whether they're 00:24:18.942 --> 00:24:21.690 following some kind of straight line, like they are here. 00:24:21.690 --> 00:24:24.460 So this type of data is correlated data. 00:24:24.460 --> 00:24:27.450 We can say that x varies with y in some relationship 00:24:27.450 --> 00:24:29.910 that we can estimate from the scatterplot. 00:24:29.910 --> 00:24:31.440 Or we can look at scatterplots here 00:24:31.440 --> 00:24:33.930 and say, well, there's really not a terribly strong 00:24:33.930 --> 00:24:36.780 correlation between our x and our y-axis. 00:24:36.780 --> 00:24:38.620 The patterns are not very strong. 00:24:38.620 --> 00:24:41.700 It's a very good starting point for root cause analysis. 00:24:41.700 --> 00:24:43.860 If two variables are related, you 00:24:43.860 --> 00:24:46.260 want to think about them in terms of which one 00:24:46.260 --> 00:24:48.000 might be causing the other one. 00:24:48.000 --> 00:24:50.730 Keep in mind, correlation is not causation, 00:24:50.730 --> 00:24:54.390 but there's some kind of link between those two variables. 00:24:54.390 --> 00:24:56.940 All right, just to sum up and try to get ourselves back 00:24:56.940 --> 00:24:59.612 on schedule, so first thing we hope 00:24:59.612 --> 00:25:01.320 to prove to you-- that inspection is just 00:25:01.320 --> 00:25:04.560 a very ineffective way to produce high-quality products 00:25:04.560 --> 00:25:05.198 and services. 00:25:05.198 --> 00:25:06.990 Sometimes you're going to have inspections. 00:25:06.990 --> 00:25:09.157 Sometimes you're going to do it, but it's not really 00:25:09.157 --> 00:25:11.280 the most effective way. 00:25:11.280 --> 00:25:13.213 We then talked about total quality management 00:25:13.213 --> 00:25:14.880 and lean thinking, and tried to show you 00:25:14.880 --> 00:25:17.160 how they're closely related to each other, 00:25:17.160 --> 00:25:20.370 and that they both utilize some simple structured 00:25:20.370 --> 00:25:24.420 quantitative and qualitative tools to help achieve quality 00:25:24.420 --> 00:25:27.480 as a result of applying the different processes. 00:25:27.480 --> 00:25:29.400 We covered seven-- well, actually, 00:25:29.400 --> 00:25:31.670 covered six basic quality tools in this module. 00:25:31.670 --> 00:25:34.060 We'll cover one more in the next module. 00:25:34.060 --> 00:25:37.110 But we want you to remember that capable people are 00:25:37.110 --> 00:25:40.210 the most important factor in achieving quality. 00:25:40.210 --> 00:25:42.960 So it's not just having a better testing machine, 00:25:42.960 --> 00:25:45.310 or having a better design, or other factors, 00:25:45.310 --> 00:25:47.900 but it's capable people.