1 00:00:01,070 --> 00:00:03,410 The following content is provided under a Creative 2 00:00:03,410 --> 00:00:04,830 Commons license. 3 00:00:04,830 --> 00:00:07,020 Your support will help MIT OpenCourseWare 4 00:00:07,020 --> 00:00:11,130 continue to offer high-quality educational resources for free. 5 00:00:11,130 --> 00:00:13,670 To make a donation or to view additional materials 6 00:00:13,670 --> 00:00:17,630 from hundreds of MIT courses, visit MIT OpenCourseWare 7 00:00:17,630 --> 00:00:18,510 at ocw.mit.edu. 8 00:00:23,160 --> 00:00:26,580 PROFESSOR: But today, we'll start with the customer 9 00:00:26,580 --> 00:00:31,320 information one, which I really will admit that probably 10 00:00:31,320 --> 00:00:33,660 some of you folks know more about this than I 11 00:00:33,660 --> 00:00:35,100 do at this point. 12 00:00:35,100 --> 00:00:36,930 But what I'm going to do is cover 13 00:00:36,930 --> 00:00:38,580 the trends that have been happening 14 00:00:38,580 --> 00:00:40,050 in the last couple of years. 15 00:00:40,050 --> 00:00:44,160 And then I'd like to get into a discussion, towards the end, 16 00:00:44,160 --> 00:00:46,140 about where we should go. 17 00:00:46,140 --> 00:00:50,030 Because where we're starting to do some research on this. 18 00:00:50,030 --> 00:00:54,000 Apar, who's usually in the class, who's 19 00:00:54,000 --> 00:00:56,400 doing some stuff here. 20 00:00:56,400 --> 00:01:01,170 And we can talk a little bit more about this. 21 00:01:01,170 --> 00:01:06,600 This is the outline essentially just saying where we've been 22 00:01:06,600 --> 00:01:11,040 and where we're going and some ideas, essentially-- 23 00:01:11,040 --> 00:01:14,400 both my personal ones and others'. 24 00:01:14,400 --> 00:01:17,280 This is actually a lecture that started -- 25 00:01:17,280 --> 00:01:21,270 The first version of this was done by Nigel Wilson. 26 00:01:21,270 --> 00:01:23,810 So I gotta give him a lot of credit for this. 27 00:01:23,810 --> 00:01:27,120 And then we've evolved it, as it went, 28 00:01:27,120 --> 00:01:30,550 over the last several years. 29 00:01:30,550 --> 00:01:37,740 So this is a spectrum of components of customer 30 00:01:37,740 --> 00:01:41,010 information-- or characteristics of customer information, 31 00:01:41,010 --> 00:01:42,720 really-- 32 00:01:42,720 --> 00:01:48,270 and how it's evolved from what transit agencies are providing 33 00:01:48,270 --> 00:01:51,510 to their customers. 34 00:01:51,510 --> 00:01:55,170 On the left is what was the old stuff. 35 00:01:55,170 --> 00:01:58,800 On the right is where it's going. 36 00:01:58,800 --> 00:02:02,140 And it's not all there yet. 37 00:02:02,140 --> 00:02:07,650 But generally, things have been moving, and fairly quickly 38 00:02:07,650 --> 00:02:11,100 in the last couple of years, from an operator 39 00:02:11,100 --> 00:02:12,960 view to a customer view. 40 00:02:12,960 --> 00:02:15,360 Operators used to just tell people 41 00:02:15,360 --> 00:02:25,769 what their schedule was and barely where the trains were. 42 00:02:25,769 --> 00:02:27,060 There were some maps out there. 43 00:02:27,060 --> 00:02:30,860 But way-finding was very, very bad. 44 00:02:30,860 --> 00:02:34,830 And so, essentially, you have to pick up these pieces 45 00:02:34,830 --> 00:02:36,870 of paper, which were static. 46 00:02:36,870 --> 00:02:42,270 And now it's moving to a more dynamic real time perspective. 47 00:02:42,270 --> 00:02:47,380 Everything was always pre-trip or pre-planning your trip. 48 00:02:47,380 --> 00:02:50,550 And now, there's some indication that things 49 00:02:50,550 --> 00:02:52,694 are getting more en route. 50 00:02:52,694 --> 00:02:55,110 You have these real time apps that we'll talk a little bit 51 00:02:55,110 --> 00:02:57,570 about. 52 00:02:57,570 --> 00:03:01,230 The last two are the ones that probably haven't moved quite as 53 00:03:01,230 --> 00:03:03,840 fast as some would have hoped. 54 00:03:06,374 --> 00:03:08,040 I think most of the customer information 55 00:03:08,040 --> 00:03:13,370 that's out there still treats everyone as a generic customer. 56 00:03:13,370 --> 00:03:15,750 It doesn't really see your particular needs, 57 00:03:15,750 --> 00:03:18,420 although there are things like service alerts 58 00:03:18,420 --> 00:03:20,820 that you sign up for very specific routes 59 00:03:20,820 --> 00:03:22,050 and things like that. 60 00:03:22,050 --> 00:03:27,030 But there's not too much specific to you. 61 00:03:27,030 --> 00:03:32,280 And then, generally, you've got to look still 62 00:03:32,280 --> 00:03:35,520 for the information and pull it up yourself. 63 00:03:35,520 --> 00:03:41,190 It's not coming to you without you taking any action, in most 64 00:03:41,190 --> 00:03:43,020 of these apps, although some of the apps 65 00:03:43,020 --> 00:03:46,820 are starting to move that way. 66 00:03:46,820 --> 00:03:49,249 So, yeah. 67 00:03:49,249 --> 00:03:51,290 AUDIENCE: The active systems and passive systems, 68 00:03:51,290 --> 00:03:53,918 is that supposed to be the other way around? 69 00:03:53,918 --> 00:03:55,370 What's with the active part? 70 00:03:55,370 --> 00:03:59,270 PROFESSOR: The active part is from the perspective 71 00:03:59,270 --> 00:03:59,960 of the customer. 72 00:03:59,960 --> 00:04:01,910 The customer has to actively seek out 73 00:04:01,910 --> 00:04:04,100 the information as opposed to it coming 74 00:04:04,100 --> 00:04:06,122 to him without doing anything. 75 00:04:06,122 --> 00:04:09,070 OK. 76 00:04:09,070 --> 00:04:09,570 Yeah. 77 00:04:09,570 --> 00:04:13,200 And anybody, you can interrupt, as well. 78 00:04:13,200 --> 00:04:16,620 I like the give and take. 79 00:04:16,620 --> 00:04:22,120 It's the technologies that allow these things to include things 80 00:04:22,120 --> 00:04:25,990 that you've pretty much talked about here, probably 81 00:04:25,990 --> 00:04:28,570 already, automatic vehicle location 82 00:04:28,570 --> 00:04:31,330 systems, automatic scheduling systems, 83 00:04:31,330 --> 00:04:37,780 so that the whole data of the mission of the transit agency 84 00:04:37,780 --> 00:04:41,030 is now in digital form. 85 00:04:41,030 --> 00:04:45,990 And more generally, is available to the public through what 86 00:04:45,990 --> 00:04:50,770 most [INAUDIBLE] used to call the Google Transit Data Feed. 87 00:04:50,770 --> 00:04:53,830 And now it's the General Transit Data Feed. 88 00:04:53,830 --> 00:04:58,050 Because, basically, everybody adopted what Google started. 89 00:04:58,050 --> 00:05:01,570 And it provides universal formats 90 00:05:01,570 --> 00:05:04,360 that all the transit agencies have basically 91 00:05:04,360 --> 00:05:07,930 converted their data to. 92 00:05:07,930 --> 00:05:11,470 GPS and Wi-Fi cell phones. 93 00:05:11,470 --> 00:05:14,530 GPS is used on the vehicles, generally, sometimes 94 00:05:14,530 --> 00:05:16,900 in the cell phones. 95 00:05:16,900 --> 00:05:19,780 But essentially, provides your location. 96 00:05:19,780 --> 00:05:22,780 And this is where the customer specific stuff 97 00:05:22,780 --> 00:05:23,590 has gotten better. 98 00:05:23,590 --> 00:05:26,590 Because that technology has allowed it. 99 00:05:26,590 --> 00:05:30,280 If you didn't have a phone that told everybody 100 00:05:30,280 --> 00:05:35,650 where you were you wouldn't be able to get information 101 00:05:35,650 --> 00:05:38,860 as good as you do now. 102 00:05:38,860 --> 00:05:43,900 AFC provides a database on your trip-making. 103 00:05:43,900 --> 00:05:46,940 And that's important for two reasons. 104 00:05:46,940 --> 00:05:53,410 One, it allows the agency to know where people are traveling 105 00:05:53,410 --> 00:05:56,110 much better than they used to. 106 00:05:56,110 --> 00:06:01,030 And it also allows them, if they so desire, 107 00:06:01,030 --> 00:06:05,620 to start to develop stuff for you based on your history. 108 00:06:05,620 --> 00:06:09,190 And this is a little bit controversial. 109 00:06:09,190 --> 00:06:11,320 We still have a lot of discussions 110 00:06:11,320 --> 00:06:15,880 with transit agencies about the privacy of the data 111 00:06:15,880 --> 00:06:20,770 and tracking you through there and your taps. 112 00:06:20,770 --> 00:06:26,920 But frankly, all the commercial people were doing this. 113 00:06:26,920 --> 00:06:31,270 And I think, the customers come to expect it. 114 00:06:31,270 --> 00:06:36,400 And the agencies have to get over this hump, 115 00:06:36,400 --> 00:06:40,300 do a privacy policy that says, we're 116 00:06:40,300 --> 00:06:42,320 going to use this data in these ways. 117 00:06:42,320 --> 00:06:45,220 And we're going to give people information 118 00:06:45,220 --> 00:06:47,770 based on their previous travel. 119 00:06:47,770 --> 00:06:50,870 I don't think too many people have done that yet, 120 00:06:50,870 --> 00:06:52,120 but they're thinking about it. 121 00:06:52,120 --> 00:06:55,540 And unfortunately, sometimes the government lawyers 122 00:06:55,540 --> 00:06:59,020 are more obstinate about doing things 123 00:06:59,020 --> 00:07:04,420 like this than the private sector lawyers who generally-- 124 00:07:04,420 --> 00:07:11,620 my biased observation is, the private lawyers 125 00:07:11,620 --> 00:07:13,630 finds a way to do with the company management 126 00:07:13,630 --> 00:07:14,350 wants them to do. 127 00:07:14,350 --> 00:07:17,699 The government lawyers finds a way 128 00:07:17,699 --> 00:07:19,240 to tell management they can't do what 129 00:07:19,240 --> 00:07:21,250 they want to do, in many cases. 130 00:07:21,250 --> 00:07:26,660 And it's unfortunate from that perspective. 131 00:07:26,660 --> 00:07:31,960 And then, obviously, wireless communications and the apps 132 00:07:31,960 --> 00:07:34,520 that have been developed. 133 00:07:34,520 --> 00:07:37,460 So before Google Transit, and this 134 00:07:37,460 --> 00:07:40,340 is probably ancient in your eyes, 135 00:07:40,340 --> 00:07:42,950 but there used to be a bunch of-- 136 00:07:42,950 --> 00:07:46,920 not a bunch-- about three or four or five-- 137 00:07:46,920 --> 00:07:51,870 software companies worldwide who did these things. 138 00:07:51,870 --> 00:07:56,580 And they were doing it with really old technology. 139 00:07:56,580 --> 00:07:59,220 And they were mainframe systems. 140 00:07:59,220 --> 00:08:01,620 And they collected literally, $1 million 141 00:08:01,620 --> 00:08:03,540 for each Transit Authority. 142 00:08:03,540 --> 00:08:11,400 And it was funny to see Google do this for free, essentially. 143 00:08:11,400 --> 00:08:14,610 Because these companies almost-- 144 00:08:14,610 --> 00:08:21,900 a lot of them just died because of the competition. 145 00:08:21,900 --> 00:08:26,460 So when they first did this, we were all amazed. 146 00:08:26,460 --> 00:08:29,040 Because a lot of the big transit authorities 147 00:08:29,040 --> 00:08:32,270 had invested lots and lots of money 148 00:08:32,270 --> 00:08:33,659 to the majority of planners. 149 00:08:33,659 --> 00:08:38,200 And they involved a tremendous amount of data preparation 150 00:08:38,200 --> 00:08:42,270 and a lot of customization for each city. 151 00:08:42,270 --> 00:08:48,750 And this was a breakthrough that nobody ever saw. 152 00:08:48,750 --> 00:08:50,730 Nobody quite understood why Google 153 00:08:50,730 --> 00:08:53,460 was doing it at the time. 154 00:08:53,460 --> 00:08:56,250 But it was welcomed. 155 00:08:56,250 --> 00:09:01,200 And there's some funny stories about how it was then adapted. 156 00:09:01,200 --> 00:09:03,620 Because a lot of the big transit authorities already 157 00:09:03,620 --> 00:09:05,370 had these trip planners that they invested 158 00:09:05,370 --> 00:09:07,190 these millions of dollars on. 159 00:09:07,190 --> 00:09:11,490 And they were loathe to give them up. 160 00:09:11,490 --> 00:09:15,450 And they were loathe to then change the input to the Google 161 00:09:15,450 --> 00:09:17,130 Trip input. 162 00:09:17,130 --> 00:09:19,680 Because all the formats of these other things 163 00:09:19,680 --> 00:09:21,180 were all proprietary. 164 00:09:21,180 --> 00:09:24,030 And they basically had to do a lot of work 165 00:09:24,030 --> 00:09:26,680 to get it into Google transit. 166 00:09:26,680 --> 00:09:31,420 But they got embarrassed into it by students, students 167 00:09:31,420 --> 00:09:34,000 even on the high school level, who 168 00:09:34,000 --> 00:09:37,600 really took their towns' schedules 169 00:09:37,600 --> 00:09:40,680 and put them into Google transit. 170 00:09:40,680 --> 00:09:45,640 And this was done in city after city, really. 171 00:09:45,640 --> 00:09:49,480 In some cases, in big cities, to show them up. 172 00:09:49,480 --> 00:09:51,520 And other cases, in little cities, 173 00:09:51,520 --> 00:09:54,040 because they just didn't have the staff to do it. 174 00:09:54,040 --> 00:09:58,302 So it was a great development. 175 00:09:58,302 --> 00:10:02,000 You've seen the apps that-- 176 00:10:02,000 --> 00:10:06,050 next vehicle arrival times, but on your phone, 177 00:10:06,050 --> 00:10:08,830 which you all probably use. 178 00:10:08,830 --> 00:10:10,554 There's many new apps. 179 00:10:10,554 --> 00:10:11,470 Some of them are good. 180 00:10:11,470 --> 00:10:12,636 Some of them aren't so good. 181 00:10:16,270 --> 00:10:17,640 Give the Google data out. 182 00:10:17,640 --> 00:10:21,010 And have other people write the apps. 183 00:10:21,010 --> 00:10:24,910 That was the chosen path, because they 184 00:10:24,910 --> 00:10:29,140 realize they couldn't do it as quickly as private people could 185 00:10:29,140 --> 00:10:30,760 do it. 186 00:10:30,760 --> 00:10:36,400 And they, essentially, let this go. 187 00:10:36,400 --> 00:10:40,780 My criticism of this approach-- 188 00:10:40,780 --> 00:10:45,130 I think in Boston we had between 35 and 45 transit 189 00:10:45,130 --> 00:10:49,570 apps, none of which were well supported 190 00:10:49,570 --> 00:10:51,850 for about two or three years. 191 00:10:51,850 --> 00:10:53,890 Nobody quite knew. 192 00:10:53,890 --> 00:10:55,985 The people who really cared-- 193 00:10:55,985 --> 00:10:58,360 there was a little bit of word-of-mouth that went around. 194 00:10:58,360 --> 00:11:02,030 There was the rating systems online and things like that. 195 00:11:02,030 --> 00:11:06,310 But when you ask around, there was at least a half dozen 196 00:11:06,310 --> 00:11:09,280 that were being used and nobody quite 197 00:11:09,280 --> 00:11:11,920 knew whether they were good or not. 198 00:11:11,920 --> 00:11:15,190 And basically, everybody had a different standard with respect 199 00:11:15,190 --> 00:11:18,850 to whether they would use them. 200 00:11:18,850 --> 00:11:23,570 And so, recently, they've changed. 201 00:11:23,570 --> 00:11:26,590 They've started to endorse apps. 202 00:11:26,590 --> 00:11:33,010 The MBTA this year endorsed Transit as their app. 203 00:11:33,010 --> 00:11:36,160 In a lot of cities, City Mapper has been become 204 00:11:36,160 --> 00:11:39,020 the standard for most people. 205 00:11:39,020 --> 00:11:42,400 And so, there's two or three that 206 00:11:42,400 --> 00:11:48,610 are reasonably good both in the design and in the information. 207 00:11:48,610 --> 00:11:53,230 But I'll talk about the good and the bad of these in a second 208 00:11:53,230 --> 00:11:56,571 here So I've put three-- 209 00:11:56,571 --> 00:11:57,070 yeah. 210 00:11:57,070 --> 00:11:58,870 Go ahead. 211 00:11:58,870 --> 00:12:04,820 AUDIENCE: So regarding these apps [INAUDIBLE] have you also 212 00:12:04,820 --> 00:12:10,266 looked at different levels of perspective of how people use 213 00:12:10,266 --> 00:12:15,290 the information, especially how features, about income levels-- 214 00:12:15,290 --> 00:12:17,580 PROFESSOR: Yeah. 215 00:12:17,580 --> 00:12:19,500 There's been very little. 216 00:12:19,500 --> 00:12:22,710 And there is some research going on. 217 00:12:22,710 --> 00:12:24,830 It's a real question. 218 00:12:24,830 --> 00:12:29,780 The research question is, have these changed people's habits? 219 00:12:29,780 --> 00:12:31,830 Has it changed their behavior? 220 00:12:31,830 --> 00:12:36,880 And there's not a lot of evidence. 221 00:12:36,880 --> 00:12:37,440 I don't know. 222 00:12:37,440 --> 00:12:39,790 Gabrielle, do you have any examples 223 00:12:39,790 --> 00:12:42,002 that you'd like to throw out? 224 00:12:42,002 --> 00:12:43,770 AUDIENCE: [INAUDIBLE] 225 00:12:43,770 --> 00:12:44,990 PROFESSOR: Yeah. 226 00:12:44,990 --> 00:12:48,780 Candace Brakewood, who was a graduate student here 227 00:12:48,780 --> 00:12:52,710 and then went on to get her PhD at Georgia Tech, 228 00:12:52,710 --> 00:12:56,280 has done some good research on this. 229 00:12:56,280 --> 00:12:58,065 But it's limited, in terms of the samples. 230 00:13:01,050 --> 00:13:03,030 And it's not conclusive with respect 231 00:13:03,030 --> 00:13:05,880 to across the spectrum of the use of the thing. 232 00:13:05,880 --> 00:13:08,760 AUDIENCE: I ask because not only because the process 233 00:13:08,760 --> 00:13:11,880 being used [INAUDIBLE] but because many 234 00:13:11,880 --> 00:13:15,970 of these points that we were just talking about 235 00:13:15,970 --> 00:13:19,210 are being emulated in the development world. 236 00:13:19,210 --> 00:13:21,700 And I think it's the way we should go. 237 00:13:21,700 --> 00:13:26,146 But if access to this data-- 238 00:13:26,146 --> 00:13:31,660 if people are not using this data in the developed world, 239 00:13:31,660 --> 00:13:35,080 then I also believe that policy is 240 00:13:35,080 --> 00:13:38,540 geared towards encouraging people or making this data 241 00:13:38,540 --> 00:13:42,970 more accessible [INAUDIBLE] people [INAUDIBLE] 242 00:13:42,970 --> 00:13:45,000 PROFESSOR: Yeah. 243 00:13:45,000 --> 00:13:48,470 And I think-- my feeling about this whole area 244 00:13:48,470 --> 00:13:52,760 is it's got more questions left than answers. 245 00:13:52,760 --> 00:13:58,370 There's not a definitive way, essentially. 246 00:13:58,370 --> 00:14:02,720 So I think we should we should talk about it as we go here. 247 00:14:02,720 --> 00:14:03,920 We'll see. 248 00:14:03,920 --> 00:14:10,280 So the basic aspects of these things you probably know. 249 00:14:12,860 --> 00:14:18,770 The best ones really depend on knowing your location. 250 00:14:18,770 --> 00:14:23,030 So again, you have to give the app permission 251 00:14:23,030 --> 00:14:26,660 to know your location through your smartphone. 252 00:14:26,660 --> 00:14:29,600 If you don't do that, they don't work well at all. 253 00:14:29,600 --> 00:14:31,370 Because, basically, then you have 254 00:14:31,370 --> 00:14:33,560 to tell them where you are. 255 00:14:33,560 --> 00:14:39,640 And most of them don't even have a good way to do that. 256 00:14:39,640 --> 00:14:44,600 The NextBus one, which was the earliest one, 257 00:14:44,600 --> 00:14:46,730 does have a drop-down menu, if you don't want 258 00:14:46,730 --> 00:14:48,050 to tell them your location. 259 00:14:48,050 --> 00:14:52,650 And you can just pick as you go down. 260 00:14:52,650 --> 00:14:55,030 But the other ones don't really offer that. 261 00:14:55,030 --> 00:14:57,140 You basically have to tell them the location. 262 00:14:57,140 --> 00:14:58,740 And it works well. 263 00:14:58,740 --> 00:15:00,500 It knows where you are right away. 264 00:15:00,500 --> 00:15:02,530 It looks at everything around you 265 00:15:02,530 --> 00:15:05,000 and it gives you the schedules, the next couple buses 266 00:15:05,000 --> 00:15:08,100 that are right around you. 267 00:15:08,100 --> 00:15:11,780 And so you can download any of these things 268 00:15:11,780 --> 00:15:13,790 and see how it works. 269 00:15:13,790 --> 00:15:16,640 I go between NextBus and Transit. 270 00:15:16,640 --> 00:15:18,260 I haven't used CityMapper that much. 271 00:15:18,260 --> 00:15:20,360 Although in Europe, almost everyone 272 00:15:20,360 --> 00:15:25,340 uses CityMapper much more than any of these other ones. 273 00:15:28,410 --> 00:15:32,570 The newest thing has been providing these other modes. 274 00:15:32,570 --> 00:15:38,330 So essentially on Transit now, you can order an Uber. 275 00:15:38,330 --> 00:15:45,500 And you could look at anyone who wants to give them 276 00:15:45,500 --> 00:15:48,900 data, essentially, of options. 277 00:15:48,900 --> 00:15:52,330 And so, I don't think Bridge is on there. 278 00:15:52,330 --> 00:15:55,710 But I've seen the car sharing things on there, too. 279 00:15:55,710 --> 00:15:58,790 So it becomes other mobility services as well as 280 00:15:58,790 --> 00:15:59,720 public transit. 281 00:15:59,720 --> 00:16:00,357 Yeah. 282 00:16:00,357 --> 00:16:03,505 AUDIENCE: I suspected that when those Uber services 283 00:16:03,505 --> 00:16:07,280 TNCs started popping up on the transit apps, 284 00:16:07,280 --> 00:16:10,055 I suspected that they might be paying 285 00:16:10,055 --> 00:16:11,760 those apps to have that option. 286 00:16:11,760 --> 00:16:16,100 Because those options didn't exist until not too long ago. 287 00:16:16,100 --> 00:16:18,830 And suddenly, you see this option. 288 00:16:18,830 --> 00:16:21,100 Is there any truth to that? 289 00:16:21,100 --> 00:16:23,320 PROFESSOR: Well, the first one that tried 290 00:16:23,320 --> 00:16:24,830 is a thing that's extinct now. 291 00:16:24,830 --> 00:16:26,350 It's called RideScout. 292 00:16:26,350 --> 00:16:28,910 RideScout was the first one who tried to put 293 00:16:28,910 --> 00:16:30,410 all these things together. 294 00:16:30,410 --> 00:16:34,790 And they actually got resistance at first from Uber and Lyft. 295 00:16:34,790 --> 00:16:38,540 Because at the beginning, Uber and Lyft 296 00:16:38,540 --> 00:16:42,260 felt like they had all the demand that they could handle. 297 00:16:42,260 --> 00:16:45,380 And they wanted people to go to their app. 298 00:16:45,380 --> 00:16:47,660 They didn't want to pay them. 299 00:16:47,660 --> 00:16:50,930 Now, I don't know. 300 00:16:50,930 --> 00:16:54,250 This Transit app clearly has it. 301 00:16:54,250 --> 00:16:56,630 And a lot of the other ones-- 302 00:16:56,630 --> 00:16:58,700 CityMapper also does. 303 00:16:58,700 --> 00:17:00,500 Yeah. 304 00:17:00,500 --> 00:17:03,200 So I think they are paying them. 305 00:17:03,200 --> 00:17:08,510 But that's the maturity of the TNCs as well. 306 00:17:08,510 --> 00:17:12,290 Because they're trying to get as much market share as they can. 307 00:17:12,290 --> 00:17:14,790 And also because Uber has so much money, 308 00:17:14,790 --> 00:17:17,560 they just throw it at everything in order 309 00:17:17,560 --> 00:17:19,530 to get as much as they want. 310 00:17:19,530 --> 00:17:23,089 Don't worry, I'll make other editorial comments about Uber 311 00:17:23,089 --> 00:17:23,869 as we go along. 312 00:17:23,869 --> 00:17:26,099 [LAUGHING] 313 00:17:26,099 --> 00:17:28,130 All right. 314 00:17:28,130 --> 00:17:35,330 So this is the possibilities. 315 00:17:35,330 --> 00:17:40,700 And the last is, can we attract more customers using these? 316 00:17:40,700 --> 00:17:44,470 And that's the open question. 317 00:17:44,470 --> 00:17:48,090 There's some indication that we can. 318 00:17:48,090 --> 00:17:50,820 I think, from a research side, we 319 00:17:50,820 --> 00:17:56,550 want to do more testing of pilot projects to see if we can. 320 00:17:56,550 --> 00:18:01,050 I believe the second point is really important. 321 00:18:01,050 --> 00:18:05,760 We need not only information, but integrated payment 322 00:18:05,760 --> 00:18:12,420 mechanisms on the phone so that you don't have a separate 323 00:18:12,420 --> 00:18:15,510 barrier to taking transit, which is, 324 00:18:15,510 --> 00:18:18,240 you gotta get a fare card or you gotta put money in the fare box 325 00:18:18,240 --> 00:18:19,680 or you gotta-- 326 00:18:19,680 --> 00:18:23,460 there's some transaction cost with the way 327 00:18:23,460 --> 00:18:25,530 fares are in Transit. 328 00:18:25,530 --> 00:18:28,950 If it's all on the phone and it's all together, 329 00:18:28,950 --> 00:18:32,280 it really makes it a lot easier, I think. 330 00:18:32,280 --> 00:18:39,690 And so, feel that if you put the two together, you get more. 331 00:18:39,690 --> 00:18:42,270 There aren't too many people that have done that. 332 00:18:42,270 --> 00:18:47,970 Chicago has the Ventra system that does, to a large degree, 333 00:18:47,970 --> 00:18:50,100 do this. 334 00:18:50,100 --> 00:18:53,610 They're only doing it for transit systems in Chicago. 335 00:18:53,610 --> 00:18:54,870 They're doing it for the CTA. 336 00:18:54,870 --> 00:18:57,000 But you can buy a metro ticket now. 337 00:18:57,000 --> 00:19:00,480 So it's actually multi operator in the Chicago area. 338 00:19:00,480 --> 00:19:01,910 It's the commuter rail. 339 00:19:01,910 --> 00:19:04,620 And you can also use it for PACE. 340 00:19:04,620 --> 00:19:08,880 So it's an account based system on an app. 341 00:19:08,880 --> 00:19:12,660 And then you can get information in terms of the services 342 00:19:12,660 --> 00:19:13,920 as well. 343 00:19:13,920 --> 00:19:16,650 So it's getting there. 344 00:19:16,650 --> 00:19:18,660 There aren't too many systems. 345 00:19:18,660 --> 00:19:23,335 Dallas uses information app that includes payment as well-- 346 00:19:23,335 --> 00:19:24,490 unlimited. 347 00:19:24,490 --> 00:19:27,460 There's a few of them out there. 348 00:19:27,460 --> 00:19:30,610 London's got the payment. 349 00:19:30,610 --> 00:19:34,070 And now they're doing the app for the information. 350 00:19:34,070 --> 00:19:39,600 So they were ahead in terms of different payment mechanisms. 351 00:19:43,460 --> 00:19:45,040 So it's getting there. 352 00:19:45,040 --> 00:19:48,340 The question really is, what do you need 353 00:19:48,340 --> 00:19:50,250 to get people's eyes on it? 354 00:19:50,250 --> 00:19:56,590 And right now, I believe it's mostly word-of-mouth. 355 00:19:56,590 --> 00:20:00,599 Those transit agencies spend very, very little on marketing. 356 00:20:00,599 --> 00:20:02,640 They used to do advertising and things like that. 357 00:20:02,640 --> 00:20:07,090 All that stuff's gone by the board with the operating 358 00:20:07,090 --> 00:20:08,710 resource constraints they've had. 359 00:20:08,710 --> 00:20:13,880 So it's like a lot of things. 360 00:20:13,880 --> 00:20:15,640 There's all these new things out there. 361 00:20:15,640 --> 00:20:18,340 But how do you get people to even look at it. 362 00:20:18,340 --> 00:20:23,050 And that's a big question that the big companies 363 00:20:23,050 --> 00:20:24,160 know how to do. 364 00:20:24,160 --> 00:20:27,520 The transit agencies clearly haven't figured that out yet. 365 00:20:31,884 --> 00:20:32,800 goes a little further. 366 00:20:35,970 --> 00:20:41,070 So this is an idea that when London 367 00:20:41,070 --> 00:20:44,820 was talking about integrated payment on a phone 368 00:20:44,820 --> 00:20:48,330 and using your credit card account to do it. 369 00:20:48,330 --> 00:20:50,490 George Kocher said, yeah, don't stop there. 370 00:20:50,490 --> 00:20:53,280 Do all the information, as well. 371 00:20:53,280 --> 00:20:59,910 And you can essentially just have the whole transaction 372 00:20:59,910 --> 00:21:02,910 with public transit and, potentially, other mobility 373 00:21:02,910 --> 00:21:05,220 services right in your pocket. 374 00:21:05,220 --> 00:21:09,000 The public sector has been wary of is 375 00:21:09,000 --> 00:21:14,460 this whole idea of tracking your movements and keeping it. 376 00:21:14,460 --> 00:21:18,750 And that has been a difficult thing 377 00:21:18,750 --> 00:21:20,250 to get over even in the research. 378 00:21:20,250 --> 00:21:22,680 We've tried to do little pilot projects. 379 00:21:22,680 --> 00:21:27,300 And the amount of bureaucracy you 380 00:21:27,300 --> 00:21:32,100 have to go through in order to essentially ask people if we 381 00:21:32,100 --> 00:21:34,320 can track their movements-- 382 00:21:34,320 --> 00:21:36,030 which, by the way, when you do ask them, 383 00:21:36,030 --> 00:21:38,280 and you tell them what you're doing it for, especially 384 00:21:38,280 --> 00:21:47,910 for transit, over 50% is our experience actually agree. 385 00:21:47,910 --> 00:21:54,240 So it's a large sample of the users agree to this. 386 00:21:54,240 --> 00:21:56,660 Because, essentially, they see the value 387 00:21:56,660 --> 00:22:00,040 at some of the information that's coming back. 388 00:22:00,040 --> 00:22:04,200 So the challenges that are left. 389 00:22:04,200 --> 00:22:07,680 There's still some agencies, when 390 00:22:07,680 --> 00:22:09,300 you go from city to city, that don't 391 00:22:09,300 --> 00:22:11,910 have a lot of good real time information. 392 00:22:11,910 --> 00:22:14,940 So the apps don't work very well. 393 00:22:14,940 --> 00:22:21,030 And for the life of me, it's just a matter of inertia, 394 00:22:21,030 --> 00:22:24,150 I believe, in these places. 395 00:22:24,150 --> 00:22:26,220 Although in some cases, they just 396 00:22:26,220 --> 00:22:29,340 don't have the technology staff. 397 00:22:29,340 --> 00:22:33,060 I mean, they're below water all the time, 398 00:22:33,060 --> 00:22:36,330 in terms of the projects they have on technology. 399 00:22:36,330 --> 00:22:40,020 And this is not their highest priority. 400 00:22:40,020 --> 00:22:42,720 But we still have a lot of cities that aren't 401 00:22:42,720 --> 00:22:44,197 releasing the real time data. 402 00:22:44,197 --> 00:22:45,780 So some of these apps just don't work. 403 00:22:50,028 --> 00:22:53,950 The standard format for the schedule data 404 00:22:53,950 --> 00:22:58,870 is pretty much established-- the general transit data feed. 405 00:22:58,870 --> 00:23:04,270 But the GTA [? available ?] time is not that well established. 406 00:23:04,270 --> 00:23:10,180 And most cities that provide real time data to the public 407 00:23:10,180 --> 00:23:11,830 do it this way. 408 00:23:11,830 --> 00:23:14,200 But they sometimes do it in a different way. 409 00:23:14,200 --> 00:23:16,930 So actually, MBTA has two formats. 410 00:23:16,930 --> 00:23:21,820 And it turns out that right now-- 411 00:23:21,820 --> 00:23:24,370 the GTA vest format, and you can correct me 412 00:23:24,370 --> 00:23:26,560 on this one, Gabrielle. 413 00:23:26,560 --> 00:23:30,070 But the GTA vest format requires the developers 414 00:23:30,070 --> 00:23:34,660 to download all of the data for all of the routes 415 00:23:34,660 --> 00:23:38,240 when they ask for it. 416 00:23:38,240 --> 00:23:40,240 It's setup for somebody who wants 417 00:23:40,240 --> 00:23:46,810 to pull every route, whereas in some cases, 418 00:23:46,810 --> 00:23:50,920 if you want to do this faster and you want to do it smarter, 419 00:23:50,920 --> 00:23:53,290 you don't want to keep pulling the whole system. 420 00:23:53,290 --> 00:23:56,050 Because you can't do it fast enough. 421 00:23:56,050 --> 00:23:59,380 And I think this is something we found when we were actually 422 00:23:59,380 --> 00:24:01,060 using it for a different purpose-- 423 00:24:01,060 --> 00:24:03,310 for an internal purpose. 424 00:24:03,310 --> 00:24:09,490 But we wanted to use their separate API to their database 425 00:24:09,490 --> 00:24:12,570 where this data resides. 426 00:24:12,570 --> 00:24:14,190 Is that-- 427 00:24:14,190 --> 00:24:17,450 AUDIENCE: It's mostly correct. 428 00:24:17,450 --> 00:24:25,070 One distinction [INAUDIBLE] [? format ?] [INAUDIBLE] data 429 00:24:25,070 --> 00:24:32,370 [INAUDIBLE] agencies like the [? MBTA ?] [INAUDIBLE] one 430 00:24:32,370 --> 00:24:36,390 example [INAUDIBLE] is the waiting time for the Green 431 00:24:36,390 --> 00:24:37,790 Line. 432 00:24:37,790 --> 00:24:41,699 So they've developed a custom API [INAUDIBLE].. 433 00:24:41,699 --> 00:24:42,740 PROFESSOR: Based on that. 434 00:24:45,570 --> 00:24:48,480 The biggest issue with all of these apps, 435 00:24:48,480 --> 00:24:52,800 really, is they don't work well in predicting 436 00:24:52,800 --> 00:24:56,280 when the bus is going to arrive unless there's 437 00:24:56,280 --> 00:24:58,140 no traffic out there. 438 00:24:58,140 --> 00:25:04,380 So when there are generally good conditions in terms of traffic, 439 00:25:04,380 --> 00:25:08,460 they're generally good at predicting the arrival. 440 00:25:08,460 --> 00:25:11,490 When there's disruptions and when there's 441 00:25:11,490 --> 00:25:14,970 any significant level of congestion, 442 00:25:14,970 --> 00:25:20,400 they always underestimate the time that you're going to wait. 443 00:25:20,400 --> 00:25:21,720 And you can see this. 444 00:25:21,720 --> 00:25:24,120 And people get very, very frustrated 445 00:25:24,120 --> 00:25:26,100 when it says it's going to arrive in 4 minutes, 446 00:25:26,100 --> 00:25:27,790 and it takes 8 or 9 minutes. 447 00:25:27,790 --> 00:25:28,500 Yeah. 448 00:25:28,500 --> 00:25:29,038 Yeah. 449 00:25:29,038 --> 00:25:32,014 AUDIENCE: Can you say something about when the code shows 450 00:25:32,014 --> 00:25:35,250 the vehicle [INAUDIBLE] 451 00:25:35,250 --> 00:25:35,940 PROFESSOR: Yeah. 452 00:25:35,940 --> 00:25:38,400 Well, the Transit app does that. 453 00:25:38,400 --> 00:25:43,110 You can push it and see the map and see where the bus is. 454 00:25:43,110 --> 00:25:45,240 And so, that helps. 455 00:25:45,240 --> 00:25:48,210 Because when you get to know your route 456 00:25:48,210 --> 00:25:52,050 and you get to know how long it generally takes when-- 457 00:25:52,050 --> 00:25:55,590 I take the route 1 bus towards Harvard. 458 00:25:55,590 --> 00:25:56,895 So I look out over the bridge. 459 00:25:56,895 --> 00:26:00,000 And if the bridge is backed up and it says 3 minutes, 460 00:26:00,000 --> 00:26:04,260 I know it's really about 8 to 10 minutes. 461 00:26:04,260 --> 00:26:05,220 So you're right. 462 00:26:05,220 --> 00:26:08,680 It does help when it does that. 463 00:26:08,680 --> 00:26:14,820 But frankly, for everyone, that's another button. 464 00:26:14,820 --> 00:26:18,060 You have to scroll through the map to see if it's there. 465 00:26:18,060 --> 00:26:19,770 The vehicles don't always show up. 466 00:26:19,770 --> 00:26:23,010 Because that vehicle hadn't been pulled in the last-- 467 00:26:23,010 --> 00:26:26,710 whatever-- minutes or something. 468 00:26:26,710 --> 00:26:28,650 And there's a lag time. 469 00:26:28,650 --> 00:26:32,770 There's a lag time with the update on this thing. 470 00:26:32,770 --> 00:26:34,800 There's a student in electrical engineering 471 00:26:34,800 --> 00:26:39,420 who wants to do his master's thesis on throwing out-- 472 00:26:39,420 --> 00:26:41,040 most of these predictions in Boston 473 00:26:41,040 --> 00:26:44,400 are based on the NextBus black box algorithm. 474 00:26:44,400 --> 00:26:47,110 And we don't know exactly what's in there. 475 00:26:47,110 --> 00:26:51,570 It's mostly based on historical information about travel times 476 00:26:51,570 --> 00:26:52,530 on the route. 477 00:26:52,530 --> 00:26:54,450 So the way the algorithm generally works 478 00:26:54,450 --> 00:26:58,230 is historical travel times on the route by time-of-day. 479 00:26:58,230 --> 00:27:01,620 And then it puts in a little bit of the last one or two buses. 480 00:27:01,620 --> 00:27:04,890 But we don't think it puts in enough of that. 481 00:27:04,890 --> 00:27:09,690 So there is a student in electrical engineering 482 00:27:09,690 --> 00:27:13,400 who wants to perhaps do a master's program. 483 00:27:13,400 --> 00:27:19,320 Because he is very frustrated at the MIT shuttle predictions, 484 00:27:19,320 --> 00:27:23,430 which are always not very accurate. 485 00:27:23,430 --> 00:27:26,880 And he wants to throw out the NextBus predictions 486 00:27:26,880 --> 00:27:29,800 and put in the homegrown prediction. 487 00:27:29,800 --> 00:27:32,100 And so we'll see how that works out. 488 00:27:32,100 --> 00:27:34,300 AUDIENCE: I guess, the next level for these apps 489 00:27:34,300 --> 00:27:40,050 is to pull data from Waze or some live data 490 00:27:40,050 --> 00:27:42,460 I think it's the big problem which is how much 491 00:27:42,460 --> 00:27:45,795 longer it actually will take the bus to get from A to B. 492 00:27:45,795 --> 00:27:49,050 And then there's the advantage, thinking of Waze, 493 00:27:49,050 --> 00:27:51,530 of having that they're linking a lot 494 00:27:51,530 --> 00:27:54,770 more vehicles and people are transmitting this, 495 00:27:54,770 --> 00:27:59,060 well then integrate expected Transit travel 496 00:27:59,060 --> 00:28:01,490 times and the actual travel time on the road. 497 00:28:01,490 --> 00:28:02,650 We could integrate that. 498 00:28:02,650 --> 00:28:04,970 And it would be a very powerful tool. 499 00:28:04,970 --> 00:28:05,760 PROFESSOR: Yeah. 500 00:28:05,760 --> 00:28:07,650 That's exactly what he is going to do. 501 00:28:07,650 --> 00:28:12,890 He's going to use the Google real time traffic. 502 00:28:12,890 --> 00:28:13,770 And then, so, yeah. 503 00:28:13,770 --> 00:28:16,755 AUDIENCE: [INAUDIBLE] when you the [INAUDIBLE] transit 504 00:28:16,755 --> 00:28:18,030 [INAUDIBLE] 505 00:28:18,030 --> 00:28:19,560 PROFESSOR: It's using their-- 506 00:28:19,560 --> 00:28:27,720 AUDIENCE: [INAUDIBLE] 507 00:28:27,720 --> 00:28:30,630 PROFESSOR: I think on Google Transit, 508 00:28:30,630 --> 00:28:34,040 there's a little indicator whether it's real time or not 509 00:28:34,040 --> 00:28:35,000 real time. 510 00:28:35,000 --> 00:28:39,530 And if that's bolded, you know it's real time. 511 00:28:39,530 --> 00:28:41,540 And it's based on something. 512 00:28:41,540 --> 00:28:43,900 But again, those companies-- 513 00:28:43,900 --> 00:28:45,740 it's just like NextBus. 514 00:28:45,740 --> 00:28:47,210 They have their own algorithm. 515 00:28:47,210 --> 00:28:49,730 And the question really is-- 516 00:28:49,730 --> 00:28:51,830 I don't know who the people in the traffic field 517 00:28:51,830 --> 00:28:55,100 are, whether they're looking at these predictions. 518 00:28:55,100 --> 00:29:00,450 But I know if you look at the Bluetooth signs 519 00:29:00,450 --> 00:29:04,120 on the traffic side, there's a huge lag effect. 520 00:29:04,120 --> 00:29:09,780 So when traffic is building up, you never get enough-- 521 00:29:09,780 --> 00:29:11,940 the Bluetooth prediction is too low. 522 00:29:11,940 --> 00:29:16,350 When traffic is coming down, the Bluetooth projection 523 00:29:16,350 --> 00:29:17,850 is too high. 524 00:29:17,850 --> 00:29:20,940 And I look at this all the time on the highways. 525 00:29:20,940 --> 00:29:23,980 Because I gotta go back and forth to Cape Cod. 526 00:29:23,980 --> 00:29:30,030 And it's amazing to see, it is better than the Transit 527 00:29:30,030 --> 00:29:31,500 predictions, I can tell you that. 528 00:29:31,500 --> 00:29:33,333 The other issue with the Transit predictions 529 00:29:33,333 --> 00:29:37,600 is they've got issues that the Traffic doesn't have. 530 00:29:37,600 --> 00:29:40,470 They got dwell time at the stops. 531 00:29:40,470 --> 00:29:44,610 And all it takes is one wheelchair at Central Square 532 00:29:44,610 --> 00:29:48,870 to be loaded and that thing is off by 2 or 3 minutes, just 533 00:29:48,870 --> 00:29:49,950 like that. 534 00:29:49,950 --> 00:29:53,400 Because it takes 2 or 3 minutes to load a wheelchair 535 00:29:53,400 --> 00:29:54,783 and secure the wheelchair. 536 00:29:57,510 --> 00:30:00,690 And then there's sometimes a crowd coming out of the subway. 537 00:30:00,690 --> 00:30:05,280 And depending on the driver, they wave 538 00:30:05,280 --> 00:30:08,580 on the people who they think have passes. 539 00:30:08,580 --> 00:30:12,180 Other drivers, they tell him to get on the back door. 540 00:30:12,180 --> 00:30:14,910 There's other drivers say, no, you gotta-- 541 00:30:14,910 --> 00:30:17,760 and then other drivers insist you get behind the yellow line. 542 00:30:17,760 --> 00:30:20,160 And they'll sit there at the stop for awhile. 543 00:30:20,160 --> 00:30:22,230 So there's all this driver behavior, 544 00:30:22,230 --> 00:30:28,110 not to speak of the difference in pulling in and out of stops 545 00:30:28,110 --> 00:30:32,130 among drivers, which you don't have in the general traffic 546 00:30:32,130 --> 00:30:33,250 flow. 547 00:30:33,250 --> 00:30:34,950 So you got the general traffic, you 548 00:30:34,950 --> 00:30:38,770 got the dwell times for things like handicap vehicles. 549 00:30:38,770 --> 00:30:40,500 So it's a huge problem. 550 00:30:40,500 --> 00:30:45,620 It's not it's not a simple algorithm to solve. 551 00:30:45,620 --> 00:30:50,410 It's one of those things that we should be looking to do better. 552 00:30:54,631 --> 00:30:58,160 In major disruptions, it really becomes 553 00:30:58,160 --> 00:31:02,160 an issue of how much do you tell the public, 554 00:31:02,160 --> 00:31:03,620 let's do something else. 555 00:31:03,620 --> 00:31:06,690 This bus is not coming in any reasonable amount of time. 556 00:31:06,690 --> 00:31:09,120 You can walk there a lot faster. 557 00:31:09,120 --> 00:31:12,800 And all the transit agencies have been 558 00:31:12,800 --> 00:31:14,690 reluctant to figure that out. 559 00:31:14,690 --> 00:31:19,310 Although London is a lot closer to thinking about, 560 00:31:19,310 --> 00:31:21,500 how do we tell people about disruptions? 561 00:31:21,500 --> 00:31:27,110 And how much do we divert them from their normal path? 562 00:31:27,110 --> 00:31:30,630 But there's not a lot of real good research 563 00:31:30,630 --> 00:31:32,220 out there about how to do that. 564 00:31:34,890 --> 00:31:37,605 There is an issue with the transit agencies 565 00:31:37,605 --> 00:31:39,730 that, again, it's because it's a government agency. 566 00:31:39,730 --> 00:31:42,540 And this is probably more in the US than anywhere else. 567 00:31:42,540 --> 00:31:46,230 But because of the Americans with Disabilities Act, 568 00:31:46,230 --> 00:31:48,150 certain types of information can't 569 00:31:48,150 --> 00:31:53,430 be given out to the public by the transit agency 570 00:31:53,430 --> 00:31:55,740 unless it's fully accessible. 571 00:31:55,740 --> 00:32:00,070 And so, one of the reasons we don't have more bus 572 00:32:00,070 --> 00:32:05,020 stop signs is because small bus stop signs on small monitors 573 00:32:05,020 --> 00:32:07,960 are not accessible under the terms of the Americans 574 00:32:07,960 --> 00:32:09,430 with Disabilities Act. 575 00:32:09,430 --> 00:32:14,820 So they have to spend a lot more money doing accessible signs. 576 00:32:14,820 --> 00:32:20,692 And I would argue that this is something that's-- 577 00:32:20,692 --> 00:32:28,780 you can buy every disabled user an accessible smartphone app 578 00:32:28,780 --> 00:32:32,620 built for that cheaper than making 579 00:32:32,620 --> 00:32:36,370 every piece of information accessible. 580 00:32:36,370 --> 00:32:42,270 But there's been no movement on this. 581 00:32:42,270 --> 00:32:44,950 That's just an accepted fact. 582 00:32:44,950 --> 00:32:47,250 What they've done to get around this a little bit is 583 00:32:47,250 --> 00:32:51,390 they say to the cities and towns or somebody else-- 584 00:32:51,390 --> 00:32:54,060 the federal transit agency has their interpretation 585 00:32:54,060 --> 00:32:56,760 of an accessible sign, and other people 586 00:32:56,760 --> 00:32:59,970 have different interpretations. 587 00:32:59,970 --> 00:33:03,240 So there is some movement allowing. 588 00:33:03,240 --> 00:33:06,030 Like these private transit screens 589 00:33:06,030 --> 00:33:11,220 you see, they have to be paid for with non-federal money, 590 00:33:11,220 --> 00:33:12,960 essentially. 591 00:33:12,960 --> 00:33:15,990 And they're paid for, usually, by private people 592 00:33:15,990 --> 00:33:19,710 or by the city under some non-federal grant. 593 00:33:19,710 --> 00:33:20,210 Yeah. 594 00:33:20,210 --> 00:33:22,168 AUDIENCE: Do they have to work with the NextBus 595 00:33:22,168 --> 00:33:24,540 commission [INAUDIBLE] 596 00:33:24,540 --> 00:33:26,490 PROFESSOR: Have to have an audible thing 597 00:33:26,490 --> 00:33:29,250 or a Braille thing or something like that. 598 00:33:29,250 --> 00:33:31,860 Yeah. 599 00:33:31,860 --> 00:33:35,760 And finally, this idea of private ride sharing. 600 00:33:35,760 --> 00:33:39,990 And so, we have all these systems 601 00:33:39,990 --> 00:33:43,980 and private operators that are being incorporated here. 602 00:33:43,980 --> 00:33:48,750 But why not-- you can find me. 603 00:33:48,750 --> 00:33:56,310 If I want to give you a ride from MIT to west Cambridge, 604 00:33:56,310 --> 00:33:58,170 why can't I get in that app? 605 00:33:58,170 --> 00:34:02,580 And then you could pay me a few bucks for that ride. 606 00:34:02,580 --> 00:34:07,890 And there's a few people who are developing mostly web 607 00:34:07,890 --> 00:34:09,900 applications right now, but are starting 608 00:34:09,900 --> 00:34:12,780 to develop the phone apps to do private ride 609 00:34:12,780 --> 00:34:15,630 sharing in this, as well. 610 00:34:15,630 --> 00:34:20,699 The Ubers of the world don't like these things at all. 611 00:34:20,699 --> 00:34:24,179 And I don't really have any good examples of people doing it 612 00:34:24,179 --> 00:34:25,409 on a wide-scale basis. 613 00:34:25,409 --> 00:34:28,170 But MIT is very interested in doing this 614 00:34:28,170 --> 00:34:31,120 within their mobile app. 615 00:34:31,120 --> 00:34:31,620 So-- 616 00:34:31,620 --> 00:34:32,245 AUDIENCE: Yeah. 617 00:34:32,245 --> 00:34:33,660 If I can give you an example. 618 00:34:33,660 --> 00:34:36,989 In Israel, where I come from, Uber and Lyft 619 00:34:36,989 --> 00:34:38,389 have not been allowed in Israel. 620 00:34:38,389 --> 00:34:40,170 But they have allowed Google. 621 00:34:40,170 --> 00:34:43,739 And Google started privately developing consultant areas 622 00:34:43,739 --> 00:34:46,550 where people use exactly these services where 623 00:34:46,550 --> 00:34:48,480 if a private individual says, I'm 624 00:34:48,480 --> 00:34:52,070 going from a suburb to the beach right now at this time of day. 625 00:34:52,070 --> 00:34:55,396 If you want to join me and pay me a few bucks for the ride. 626 00:34:55,396 --> 00:34:58,764 So Google actually has started a pilot [INAUDIBLE].. 627 00:34:58,764 --> 00:34:59,430 PROFESSOR: Yeah. 628 00:34:59,430 --> 00:35:03,660 So I would argue that this might not 629 00:35:03,660 --> 00:35:06,600 be the best thing for Transit if we got everybody sharing 630 00:35:06,600 --> 00:35:07,530 rides all the time. 631 00:35:07,530 --> 00:35:10,000 But for the metropolitan area and urban area, 632 00:35:10,000 --> 00:35:11,360 it's probably a great thing. 633 00:35:11,360 --> 00:35:13,860 And it's better than having Uber and Lyft drivers 634 00:35:13,860 --> 00:35:17,490 who, when they're not full, running around empty. 635 00:35:17,490 --> 00:35:20,880 Because they're basically a purposeful trip 636 00:35:20,880 --> 00:35:23,655 that's being made, as opposed to just being a business. 637 00:35:28,040 --> 00:35:32,280 So these are potential research questions. 638 00:35:32,280 --> 00:35:37,220 I think I've talked about the arrival predictions. 639 00:35:37,220 --> 00:35:42,970 The second one is important. 640 00:35:42,970 --> 00:35:44,720 We do a lot of research in the Transit lab 641 00:35:44,720 --> 00:35:49,020 about this using the real time information, 642 00:35:49,020 --> 00:35:52,820 both from their vehicles and from customers. 643 00:35:52,820 --> 00:35:56,570 Much less, thus far, from customers in a real time 644 00:35:56,570 --> 00:35:58,070 framework. 645 00:35:58,070 --> 00:36:02,780 But if we have that kind of information coming 646 00:36:02,780 --> 00:36:06,020 in all the time, how do they make adjustments 647 00:36:06,020 --> 00:36:07,770 to the operating plan? 648 00:36:07,770 --> 00:36:15,960 So for example, the simplest one of these is on ball game days. 649 00:36:15,960 --> 00:36:21,230 They store trains outside of Kenmore Square 650 00:36:21,230 --> 00:36:23,420 and outside of the Boston Garden. 651 00:36:23,420 --> 00:36:26,360 And then they only release them when the crowd shows up 652 00:36:26,360 --> 00:36:27,740 on the platform. 653 00:36:27,740 --> 00:36:31,095 But it's still very low tech. 654 00:36:31,095 --> 00:36:32,390 There's a guy down there. 655 00:36:32,390 --> 00:36:33,490 And he says, send one in. 656 00:36:33,490 --> 00:36:37,290 [LAUGHING] 657 00:36:37,290 --> 00:36:41,100 But if you think about that on a smaller scale, 658 00:36:41,100 --> 00:36:48,890 if a school gets out, and it's on the Riverside line on the D 659 00:36:48,890 --> 00:36:54,950 branch, instead of on the even headway at 3:48, 660 00:36:54,950 --> 00:36:57,110 you dispatch it a little bit early. 661 00:36:57,110 --> 00:36:59,510 Because you knew the school just got out. 662 00:36:59,510 --> 00:37:04,730 And you will pick up that school load right away. 663 00:37:04,730 --> 00:37:07,910 And then the next bus can pick up 664 00:37:07,910 --> 00:37:09,770 all the rest of the passengers. 665 00:37:09,770 --> 00:37:12,470 Or maybe they'll put two in a row or something like that. 666 00:37:12,470 --> 00:37:17,000 This is good research. 667 00:37:17,000 --> 00:37:21,200 It's still in its infancy, I would suggest, at this point. 668 00:37:24,160 --> 00:37:28,990 The travel behavior question is paramount. 669 00:37:28,990 --> 00:37:33,740 And as I said, I don't know of really good research on this. 670 00:37:33,740 --> 00:37:40,930 And the bad part about it is I think any research that 671 00:37:40,930 --> 00:37:46,480 relies on surveys, revealed behavior, or stated behavior 672 00:37:46,480 --> 00:37:49,120 is just no good anymore. 673 00:37:49,120 --> 00:37:50,820 It may have been good in the past 674 00:37:50,820 --> 00:37:54,840 or they may have been lying to us about how good it was. 675 00:37:54,840 --> 00:38:00,280 But right now, we need something that looks at the real data. 676 00:38:00,280 --> 00:38:04,000 And so, this is where the tracking data is key. 677 00:38:04,000 --> 00:38:06,526 And it's starting to be done. 678 00:38:06,526 --> 00:38:11,680 AUDIENCE: [INAUDIBLE] I worked in industry data before this. 679 00:38:11,680 --> 00:38:14,416 Most of the issues we had in the private sector 680 00:38:14,416 --> 00:38:19,000 were overcome by making the data fully anonymous. 681 00:38:19,000 --> 00:38:23,110 Is that still not the case with these Transit systems? 682 00:38:23,110 --> 00:38:26,200 Couldn't you just have instead of me being Scott Foster, 683 00:38:26,200 --> 00:38:28,740 I'd be number 12379. 684 00:38:28,740 --> 00:38:30,960 PROFESSOR: Yeah. 685 00:38:30,960 --> 00:38:34,360 It is clearly doable. 686 00:38:34,360 --> 00:38:35,540 And it is done. 687 00:38:35,540 --> 00:38:39,010 We hashed the names and the serial numbers 688 00:38:39,010 --> 00:38:41,350 and all that stuff. 689 00:38:41,350 --> 00:38:46,030 But-- I just had a conversation about this with a public agency 690 00:38:46,030 --> 00:38:48,730 person this morning-- 691 00:38:48,730 --> 00:38:53,012 they get very concerned about small numbers. 692 00:38:53,012 --> 00:38:54,970 And on Transit, there's a lot of small numbers. 693 00:38:54,970 --> 00:38:59,080 If you're actually looking at OD movements from A 694 00:38:59,080 --> 00:39:05,210 to B in the early morning, there may be only a couple 695 00:39:05,210 --> 00:39:06,340 of people doing that. 696 00:39:06,340 --> 00:39:09,740 And so then they get nervous about doing that. 697 00:39:09,740 --> 00:39:11,890 And I would argue that those are the very things 698 00:39:11,890 --> 00:39:12,880 we want to learn about. 699 00:39:12,880 --> 00:39:16,280 Because you look at the bus. 700 00:39:16,280 --> 00:39:19,090 And you know it's full at 8:00 in the morning. 701 00:39:19,090 --> 00:39:23,550 But where should that bus be, maybe, at 6:00 in the morning? 702 00:39:23,550 --> 00:39:26,330 It maybe should be somewhere very different. 703 00:39:26,330 --> 00:39:28,960 And we have to use this automated data to do that. 704 00:39:28,960 --> 00:39:32,500 And we'd like to track who who's using it 705 00:39:32,500 --> 00:39:33,610 and how they're using it. 706 00:39:33,610 --> 00:39:36,130 Because they might be using it very, very differently. 707 00:39:36,130 --> 00:39:39,700 I don't know if you've noticed this advocacy group, Transit 708 00:39:39,700 --> 00:39:45,190 Matters in the Boston area, but they have this proposal 709 00:39:45,190 --> 00:39:46,600 for all night service. 710 00:39:46,600 --> 00:39:49,900 And it's different than the late night service 711 00:39:49,900 --> 00:39:55,537 proposal in that it's looking at service for workers early 712 00:39:55,537 --> 00:39:57,370 in the morning and at the airport and things 713 00:39:57,370 --> 00:39:58,411 like-- this guy knows it. 714 00:39:58,411 --> 00:39:59,560 Yeah. 715 00:39:59,560 --> 00:40:00,110 He did. 716 00:40:00,110 --> 00:40:00,700 Yeah. 717 00:40:00,700 --> 00:40:01,199 OK. 718 00:40:05,500 --> 00:40:11,770 But anyway, it's just, that makes sense. 719 00:40:11,770 --> 00:40:14,800 But what you don't have is good data 720 00:40:14,800 --> 00:40:17,410 about who's using it where. 721 00:40:17,410 --> 00:40:22,690 And there's not enough out there to really measure it, 722 00:40:22,690 --> 00:40:24,010 in that particular case. 723 00:40:24,010 --> 00:40:26,530 Because there would be induced demand. 724 00:40:26,530 --> 00:40:31,275 AUDIENCE: So we can guess at where people are going just 725 00:40:31,275 --> 00:40:36,260 by where [INAUDIBLE] can't get the regularity when you get 726 00:40:36,260 --> 00:40:38,900 the small [INAUDIBLE] And the private sector 727 00:40:38,900 --> 00:40:40,020 is the same thing. 728 00:40:40,020 --> 00:40:42,070 You have to make sure you don't come 729 00:40:42,070 --> 00:40:44,500 close to identifying an individual-- 730 00:40:44,500 --> 00:40:45,695 customer level data. 731 00:40:45,695 --> 00:40:48,470 So it's always aggregated in some way 732 00:40:48,470 --> 00:40:50,440 to make sure that you couldn't figure it out. 733 00:40:50,440 --> 00:40:51,981 You have all the people maybe getting 734 00:40:51,981 --> 00:40:56,329 on [INAUDIBLE] between 5:00 AM and 6:00 AM knowing 735 00:40:56,329 --> 00:40:57,370 that they got on at 5:05. 736 00:40:57,370 --> 00:40:58,036 PROFESSOR: Yeah. 737 00:40:58,036 --> 00:41:04,030 And there's a whole spectrum here of what is acceptable 738 00:41:04,030 --> 00:41:05,720 and what is not. 739 00:41:05,720 --> 00:41:08,320 I guess, I just argue, and I've said-- 740 00:41:08,320 --> 00:41:13,030 I had a TRB talk about five years ago 741 00:41:13,030 --> 00:41:20,960 about how the phone knows when I'm near a pizza spot. 742 00:41:20,960 --> 00:41:23,860 And it sends me an ad. 743 00:41:23,860 --> 00:41:26,150 And I said, you know, and if you actually 744 00:41:26,150 --> 00:41:30,740 look at the terms of use of Google on those phones, 745 00:41:30,740 --> 00:41:33,620 it's like they can do anything with that. 746 00:41:33,620 --> 00:41:36,500 They can look at your phone call. 747 00:41:36,500 --> 00:41:38,390 If you actually read those things, 748 00:41:38,390 --> 00:41:43,370 they use every piece of data that they have accessible 749 00:41:43,370 --> 00:41:46,250 almost any way they want. 750 00:41:46,250 --> 00:41:49,130 And if you say you decline, then you're 751 00:41:49,130 --> 00:41:52,270 location stuff doesn't work ever. 752 00:41:52,270 --> 00:41:54,680 So it's a mess. 753 00:41:54,680 --> 00:41:57,890 And if these folks can get you to do it, 754 00:41:57,890 --> 00:42:00,170 I say, well, the private sector is doing it. 755 00:42:00,170 --> 00:42:05,600 There isn't many bad things in the public sector. 756 00:42:05,600 --> 00:42:08,290 People think, oh, they'll arrest you or something. 757 00:42:08,290 --> 00:42:11,745 And maybe, in this case, maybe these days they will. 758 00:42:11,745 --> 00:42:12,286 I don't know. 759 00:42:12,286 --> 00:42:17,210 [LAUGHING] But in the old days, when I was talking about this, 760 00:42:17,210 --> 00:42:18,590 I said, you know, come on. 761 00:42:18,590 --> 00:42:22,270 Let the government do its job in the sense of planning services. 762 00:42:22,270 --> 00:42:23,130 Yeah. 763 00:42:23,130 --> 00:42:31,770 So one thing that comes out of this, 764 00:42:31,770 --> 00:42:39,050 and this is customer information and marketing combined, 765 00:42:39,050 --> 00:42:42,140 if we have a lot of this automated fare data, 766 00:42:42,140 --> 00:42:45,440 on a voluntary basis, we can create customer panels. 767 00:42:45,440 --> 00:42:50,450 So you can ask people, we'd like to help plan things better. 768 00:42:50,450 --> 00:42:53,120 Let's put these panels together. 769 00:42:53,120 --> 00:42:56,704 We're going to use your data for two months. 770 00:42:56,704 --> 00:42:59,120 And we want to ask you some questions about your attitudes 771 00:42:59,120 --> 00:42:59,619 and things. 772 00:42:59,619 --> 00:43:04,640 And this is done a lot in the private sector. 773 00:43:04,640 --> 00:43:07,100 When we've done it, a lot-- 774 00:43:07,100 --> 00:43:11,210 anywhere between 35% and 50% have said, yes. 775 00:43:11,210 --> 00:43:15,740 You can essentially do this on an ongoing basis. 776 00:43:15,740 --> 00:43:18,020 And you can create a very large panel. 777 00:43:18,020 --> 00:43:23,050 And you can randomly select them for other little surveys. 778 00:43:23,050 --> 00:43:26,660 And then you could see if they react with respect 779 00:43:26,660 --> 00:43:30,960 to their particular travel behavior. 780 00:43:30,960 --> 00:43:34,580 And so, I think this is the way we learn 781 00:43:34,580 --> 00:43:36,920 the travel behavior impacts. 782 00:43:36,920 --> 00:43:38,780 But you gotta get a big panel. 783 00:43:38,780 --> 00:43:41,690 And so far, the transit agencies have 784 00:43:41,690 --> 00:43:43,430 been a little worried about this, 785 00:43:43,430 --> 00:43:45,080 again, from the privacy perspective. 786 00:43:45,080 --> 00:43:48,470 I say, if people agree, there's no privacy issue. 787 00:43:51,120 --> 00:43:53,510 And then you can continually enrich this. 788 00:43:53,510 --> 00:43:56,160 You could keep recruiting people for this 789 00:43:56,160 --> 00:43:59,040 so that there is a lot of churn in the transit ridership-- 790 00:43:59,040 --> 00:44:03,720 that people don't stay forever riding the bus. 791 00:44:03,720 --> 00:44:06,890 Because they move or they get a new job or they buy a car. 792 00:44:12,890 --> 00:44:18,400 The thing you could then move to is pilot programs. 793 00:44:18,400 --> 00:44:23,320 And you could select people on these panels 794 00:44:23,320 --> 00:44:25,020 to be offered new things. 795 00:44:25,020 --> 00:44:26,800 And so this thing that we're trying 796 00:44:26,800 --> 00:44:32,620 to get started with the CTA is to do mobility as a service. 797 00:44:32,620 --> 00:44:34,600 So they have this Ventra app. 798 00:44:34,600 --> 00:44:37,690 It works really, really well for transit. 799 00:44:37,690 --> 00:44:42,970 Well, why not put bike sharing, car sharing, potentially TNCs 800 00:44:42,970 --> 00:44:44,590 on this app? 801 00:44:44,590 --> 00:44:48,310 And do it in a way that the transit agency 802 00:44:48,310 --> 00:44:52,790 dictates who gets paid and how much they get paid. 803 00:44:52,790 --> 00:44:55,120 And this is a big sell. 804 00:44:55,120 --> 00:44:58,570 Because they see Ubers and Lyfts and the car-sharing people 805 00:44:58,570 --> 00:45:02,410 as the evil empire. 806 00:45:02,410 --> 00:45:07,630 But what we say is, do it on a pilot basis, 807 00:45:07,630 --> 00:45:10,180 because you can get these individual panels. 808 00:45:10,180 --> 00:45:14,260 You could do it on a general public basis, 809 00:45:14,260 --> 00:45:15,730 if they volunteer. 810 00:45:15,730 --> 00:45:18,950 Or you could do it through employers, 811 00:45:18,950 --> 00:45:22,480 which, then, is another way transit, 812 00:45:22,480 --> 00:45:31,280 I think, has underutilized their market strength, in a sense. 813 00:45:31,280 --> 00:45:37,190 Employers, in large cities, have a motivation 814 00:45:37,190 --> 00:45:39,260 to promote transit. 815 00:45:39,260 --> 00:45:41,900 Because if they don't promote transit, 816 00:45:41,900 --> 00:45:44,570 they have to make sure that their employees can find 817 00:45:44,570 --> 00:45:46,421 parking near their offices. 818 00:45:46,421 --> 00:45:48,170 So they either have to provide the parking 819 00:45:48,170 --> 00:45:51,140 or they have to provide a parking subsidy. 820 00:45:51,140 --> 00:45:53,570 As many of you know, we have been doing a lot here 821 00:45:53,570 --> 00:45:57,110 in MIT with this and trying to change the behavior 822 00:45:57,110 --> 00:46:03,290 by essentially giving a free transit pass to every employer 823 00:46:03,290 --> 00:46:08,210 and charging daily for parking as opposed to what 824 00:46:08,210 --> 00:46:09,830 used to be a Either, Or. 825 00:46:09,830 --> 00:46:14,960 You pick transit or you pick parking. 826 00:46:14,960 --> 00:46:18,630 You can do all these things to learn. 827 00:46:18,630 --> 00:46:22,170 But what it all depends on is this mobile data 828 00:46:22,170 --> 00:46:28,500 and other automated data sources to tie it all together. 829 00:46:28,500 --> 00:46:31,700 I think one of the issues that this will also help 830 00:46:31,700 --> 00:46:34,790 is you can get feedback from people 831 00:46:34,790 --> 00:46:39,370 about what they want to know when they want to know it. 832 00:46:39,370 --> 00:46:42,240 So you can't do that in a fully-automated fashion. 833 00:46:42,240 --> 00:46:44,630 So these panels are the only way that you 834 00:46:44,630 --> 00:46:49,430 can get to a large representative group of people 835 00:46:49,430 --> 00:46:53,360 and quiz them about when they had bad service 836 00:46:53,360 --> 00:46:57,360 and what they did and what kind of information 837 00:46:57,360 --> 00:46:58,880 that you need to get to them. 838 00:47:01,830 --> 00:47:06,600 So the next four slides are from Montreal. 839 00:47:06,600 --> 00:47:09,750 And Isabelle Trottier presented them a couple of years 840 00:47:09,750 --> 00:47:12,630 ago when she came down here to what we call the Transit 841 00:47:12,630 --> 00:47:14,430 Leaders Roundtable. 842 00:47:14,430 --> 00:47:18,420 And they are a little bit ahead of the curve 843 00:47:18,420 --> 00:47:22,260 in terms of providing mobility as a service. 844 00:47:22,260 --> 00:47:24,260 And I'll just give you some examples. 845 00:47:24,260 --> 00:47:28,680 This is mostly oriented towards biking 846 00:47:28,680 --> 00:47:31,680 in Montreal and bike sharing. 847 00:47:31,680 --> 00:47:35,070 But I picked it, because I have an ax 848 00:47:35,070 --> 00:47:36,330 to grind about the biking. 849 00:47:36,330 --> 00:47:40,700 And so, I'm not quite as bad as Fred. 850 00:47:40,700 --> 00:47:42,350 But I have a good example here. 851 00:47:42,350 --> 00:47:48,930 So the bicycles are growing in popularity. 852 00:47:48,930 --> 00:47:52,020 I don't know exactly what passport mobility is. 853 00:47:52,020 --> 00:47:54,100 I don't remember what it is. 854 00:47:54,100 --> 00:47:57,690 But it's essentially a way to buy things. 855 00:47:57,690 --> 00:48:01,802 I think it's an account that you can use. 856 00:48:01,802 --> 00:48:08,024 AUDIENCE: [INAUDIBLE] 857 00:48:08,024 --> 00:48:08,690 PROFESSOR: Yeah. 858 00:48:11,860 --> 00:48:15,210 So the reason I picked the bicycle here is this 859 00:48:15,210 --> 00:48:18,450 is a place that has been very successful at combining bikes 860 00:48:18,450 --> 00:48:20,940 and buses in the same lane, which we're 861 00:48:20,940 --> 00:48:24,360 having a huge fight here in the Boston area about, 862 00:48:24,360 --> 00:48:27,280 whether it's safe or not. 863 00:48:27,280 --> 00:48:33,750 And they have done and recently have been doing studies of this 864 00:48:33,750 --> 00:48:37,620 and looking at where it works and where it doesn't work. 865 00:48:37,620 --> 00:48:41,550 And they've brought the bicycle people in to see the data 866 00:48:41,550 --> 00:48:43,050 and do it transparently. 867 00:48:43,050 --> 00:48:45,150 And I argue, we have to do this in Boston 868 00:48:45,150 --> 00:48:49,590 or else the MBTA is dead. 869 00:48:49,590 --> 00:48:51,450 There's a bunch of examples. 870 00:48:51,450 --> 00:48:53,850 The conclusion drawn from the first pilot project 871 00:48:53,850 --> 00:48:57,520 was it's both possible and even desirable. 872 00:48:57,520 --> 00:49:01,080 It's safe, and it's functional. 873 00:49:01,080 --> 00:49:03,570 The cyclists like it. 874 00:49:03,570 --> 00:49:06,990 The bus performance-- they used to have the lane by themselves. 875 00:49:06,990 --> 00:49:08,070 Now they don't. 876 00:49:08,070 --> 00:49:10,890 It's not gone down. 877 00:49:10,890 --> 00:49:14,680 And they're deploying it now, on a case by case basis, 878 00:49:14,680 --> 00:49:18,360 depending a little bit on the layout 879 00:49:18,360 --> 00:49:21,300 of the streets and the lanes. 880 00:49:21,300 --> 00:49:26,940 And so my little commercial is we need to do this for mobility 881 00:49:26,940 --> 00:49:29,610 in the Boston area. 882 00:49:29,610 --> 00:49:31,880 So my final slide in this whole area 883 00:49:31,880 --> 00:49:37,850 is Lead to a Re-definition of Public Transit. 884 00:49:37,850 --> 00:49:42,680 And I think that we're almost there, at this point. 885 00:49:42,680 --> 00:49:46,220 All these mobility services are being-- 886 00:49:46,220 --> 00:49:48,980 the same people are using all of these mobility services. 887 00:49:48,980 --> 00:49:52,050 There are now the private car users. 888 00:49:52,050 --> 00:49:54,950 And there are people using mobility services. 889 00:49:54,950 --> 00:49:59,600 And they're using a combination of them. 890 00:49:59,600 --> 00:50:03,380 I did talk two years ago when they 891 00:50:03,380 --> 00:50:09,440 did the first study of users in the New York City area of Uber 892 00:50:09,440 --> 00:50:12,230 and taxis and public transit. 893 00:50:12,230 --> 00:50:15,620 And at the time, the Uber fare in Manhattan 894 00:50:15,620 --> 00:50:22,740 averaged about $20, which was still pretty high. 895 00:50:22,740 --> 00:50:28,520 And the analysis said that if it got down to around $10, 896 00:50:28,520 --> 00:50:34,130 on average, and people used 1/3 third of their trips-- 897 00:50:34,130 --> 00:50:35,870 all their trip making on Uber-- 898 00:50:35,870 --> 00:50:40,520 and 2/3 on transit, you would actually 899 00:50:40,520 --> 00:50:47,330 spend less on that combination than buying a car even 900 00:50:47,330 --> 00:50:52,470 in the outer Boroughs of New York. 901 00:50:52,470 --> 00:50:54,920 So what's happened in the meantime 902 00:50:54,920 --> 00:50:57,170 is that Uber fares have come down. 903 00:50:57,170 --> 00:51:00,620 And they're probably at that level, as an average, now. 904 00:51:00,620 --> 00:51:04,520 So that means if you're economically savvy, 905 00:51:04,520 --> 00:51:06,980 you're crazy to be operating a car. 906 00:51:06,980 --> 00:51:09,701 Because, generally, for the longer trips 907 00:51:09,701 --> 00:51:10,700 you can take the subway. 908 00:51:10,700 --> 00:51:14,180 And for the shorter trips you take Uber. 909 00:51:14,180 --> 00:51:22,500 And the question is, has it changed behavior? 910 00:51:22,500 --> 00:51:27,140 The New York City people say, there's more cars in Manhattan. 911 00:51:27,140 --> 00:51:29,780 So that indicates there are more people using Uber. 912 00:51:29,780 --> 00:51:32,180 Because they're not using more private cars in Manhattan. 913 00:51:32,180 --> 00:51:36,030 Because they have no place to park. 914 00:51:36,030 --> 00:51:41,240 So this is a thing that transit is 915 00:51:41,240 --> 00:51:45,020 extremely worried about but it's doing very little. 916 00:51:45,020 --> 00:51:52,880 And the question is, can we control this in any way? 917 00:51:52,880 --> 00:51:54,860 And I would argue, the only way they 918 00:51:54,860 --> 00:51:59,060 can do it is by having a joint deal where 919 00:51:59,060 --> 00:52:04,910 they give information and payment together on an app. 920 00:52:04,910 --> 00:52:08,354 And they give you unlimited rides on transit. 921 00:52:08,354 --> 00:52:09,770 And they give you a certain number 922 00:52:09,770 --> 00:52:12,380 of rides on these other modes. 923 00:52:12,380 --> 00:52:18,710 And basically, it's a good economic deal for everyone. 924 00:52:18,710 --> 00:52:21,040 And you market it like that. 925 00:52:21,040 --> 00:52:26,180 Because frankly, transit has been cutting back 926 00:52:26,180 --> 00:52:28,700 the really bad off-peak services. 927 00:52:28,700 --> 00:52:34,250 Because they can't afford to subsidize it at $10 a ride. 928 00:52:34,250 --> 00:52:38,210 And if they only subsidize-- 929 00:52:38,210 --> 00:52:42,590 for example, you might give them a certain number of Uber trips 930 00:52:42,590 --> 00:52:44,510 that are just to and from rail stations. 931 00:52:44,510 --> 00:52:47,420 Because they can do this geofencing thing. 932 00:52:47,420 --> 00:52:54,440 Or you might give a subsidy for Uber trips 933 00:52:54,440 --> 00:52:59,260 up to a certain amount per trip for a certain number of trips. 934 00:52:59,260 --> 00:53:02,330 There's a whole bunch of bundling that you can do here. 935 00:53:02,330 --> 00:53:05,480 And what I'm saying is that we gotta 936 00:53:05,480 --> 00:53:10,100 do this in a series of controlled experiments 937 00:53:10,100 --> 00:53:14,520 and learn what people want and what people use and learn, 938 00:53:14,520 --> 00:53:19,220 frankly, whether we're going to end up being just high 939 00:53:19,220 --> 00:53:23,150 frequency corridor routes left. 940 00:53:23,150 --> 00:53:27,590 And there is a real debate about this. 941 00:53:27,590 --> 00:53:31,580 And there's a real debate about whether this 942 00:53:31,580 --> 00:53:37,014 is where we're going to end up or whether Uber and Lyft are 943 00:53:37,014 --> 00:53:38,180 going to go out of business. 944 00:53:38,180 --> 00:53:41,960 Because they, right now, have unsustainable fares. 945 00:53:41,960 --> 00:53:46,040 That's what the transit people hope and bet on. 946 00:53:46,040 --> 00:53:48,840 But I'm not sure that's going to happen or not. 947 00:53:48,840 --> 00:53:50,460 So, yeah. 948 00:53:50,460 --> 00:53:52,572 AUDIENCE: That's what's really interesting to me. 949 00:53:52,572 --> 00:53:55,084 I definitely agree these are going to have phenomenal 950 00:53:55,084 --> 00:53:55,584 impact.. 951 00:53:55,584 --> 00:53:58,790 But in London, they are investing huge amounts 952 00:53:58,790 --> 00:54:01,832 of money for Crossrail and Crossrail 2, which 953 00:54:01,832 --> 00:54:04,260 are high frequency routes. 954 00:54:04,260 --> 00:54:07,530 But in cities where the road layout is not 955 00:54:07,530 --> 00:54:14,470 a grid they have to scale larger, [INAUDIBLE] 956 00:54:14,470 --> 00:54:15,650 PROFESSOR: Yeah. 957 00:54:15,650 --> 00:54:20,060 And London's doing it largely on the economic theory 958 00:54:20,060 --> 00:54:21,530 of agglomeration-- 959 00:54:21,530 --> 00:54:26,210 that they have such a density of employment opportunities 960 00:54:26,210 --> 00:54:30,120 and other activities that are altogether, 961 00:54:30,120 --> 00:54:33,620 they should have an economic benefit to the thing. 962 00:54:33,620 --> 00:54:36,500 Frankly, every developed country, 963 00:54:36,500 --> 00:54:39,710 every almost developing country in the world 964 00:54:39,710 --> 00:54:45,350 is doing more investment in transit than the United States. 965 00:54:45,350 --> 00:54:49,700 And Fred would argue that we should just 966 00:54:49,700 --> 00:54:52,730 forget about all this stuff and just build more rail 967 00:54:52,730 --> 00:54:56,260 lines in certain places. 968 00:54:56,260 --> 00:54:58,970 I think that, in many cities, we're 969 00:54:58,970 --> 00:55:00,860 beyond a little bit of that. 970 00:55:00,860 --> 00:55:04,400 Because the sprawl is so great. 971 00:55:04,400 --> 00:55:04,910 It's right. 972 00:55:04,910 --> 00:55:08,570 In the big old cities, the densities-- 973 00:55:08,570 --> 00:55:12,240 if we want to increase the density in Boston or in New 974 00:55:12,240 --> 00:55:16,620 York or Philadelphia, yes, you have to have rail. 975 00:55:16,620 --> 00:55:19,750 Because essentially, you can't increase the density anymore. 976 00:55:19,750 --> 00:55:21,570 Because there's no place to put the cars. 977 00:55:21,570 --> 00:55:24,630 But in some of these other cities that are spread out 978 00:55:24,630 --> 00:55:29,250 more, it's just not as clear of what you should 979 00:55:29,250 --> 00:55:31,890 do in terms of investment. 980 00:55:31,890 --> 00:55:32,403 Yeah. 981 00:55:32,403 --> 00:55:33,236 AUDIENCE: I'm sorry. 982 00:55:33,236 --> 00:55:35,210 You said that transit authorities 983 00:55:35,210 --> 00:55:37,670 are worried about this. 984 00:55:37,670 --> 00:55:41,560 [INAUDIBLE] 985 00:55:41,560 --> 00:55:46,630 PROFESSOR: Nationally, for the last decade, 986 00:55:46,630 --> 00:55:50,370 transit ridership has gone up steadily. 987 00:55:50,370 --> 00:55:52,510 And everybody was really happy. 988 00:55:52,510 --> 00:55:55,750 The last two to three years, bus ridership-- not 989 00:55:55,750 --> 00:55:58,510 rail ridership-- bus ridership has gone down. 990 00:55:58,510 --> 00:56:00,280 And all the large cities. 991 00:56:00,280 --> 00:56:03,180 And everybody's debating about what's the cause of that. 992 00:56:03,180 --> 00:56:05,350 Is it just the gas prices? 993 00:56:05,350 --> 00:56:07,750 Is it Uber and Lyft? 994 00:56:07,750 --> 00:56:09,430 What is it? 995 00:56:09,430 --> 00:56:11,570 And they're worried. 996 00:56:11,570 --> 00:56:12,070 Yeah. 997 00:56:12,070 --> 00:56:18,714 AUDIENCE: [INAUDIBLE] 998 00:56:18,714 --> 00:56:19,630 PROFESSOR: Nationally? 999 00:56:19,630 --> 00:56:20,129 Yeah. 1000 00:56:23,380 --> 00:56:27,254 There's a big debate about why did D.C. has gone down. 1001 00:56:27,254 --> 00:56:28,920 Because they've had all these disruption 1002 00:56:28,920 --> 00:56:31,440 because of the safety and because 1003 00:56:31,440 --> 00:56:34,410 of the reconstruction of things. 1004 00:56:34,410 --> 00:56:36,210 So D.C. is a special case. 1005 00:56:36,210 --> 00:56:38,740 But New York flattened out. 1006 00:56:38,740 --> 00:56:39,510 That's right. 1007 00:56:39,510 --> 00:56:41,590 And New York was going up. 1008 00:56:41,590 --> 00:56:47,250 And nobody can argue it's the economy, because the economy 1009 00:56:47,250 --> 00:56:48,690 has been growing. 1010 00:56:52,094 --> 00:56:52,594 Yeah. 1011 00:56:52,594 --> 00:57:00,070 AUDIENCE: [INAUDIBLE] 1012 00:57:00,070 --> 00:57:00,976 PROFESSOR: Go ahead. 1013 00:57:00,976 --> 00:57:03,252 [LAUGHING] He knows this better than I do. 1014 00:57:03,252 --> 00:57:06,912 AUDIENCE: [INAUDIBLE] 1015 00:57:06,912 --> 00:57:08,620 PROFESSOR: They're skeptical in Montreal? 1016 00:57:08,620 --> 00:57:09,115 AUDIENCE: Can't decide it. 1017 00:57:09,115 --> 00:57:10,706 There is skepticism in Montreal. 1018 00:57:13,570 --> 00:57:16,240 In theory it should work. 1019 00:57:16,240 --> 00:57:19,346 Because buses and some high schools average the same speed. 1020 00:57:24,670 --> 00:57:27,920 In practice, the cyclist goes 15 miles an hour. 1021 00:57:27,920 --> 00:57:30,150 The bus goes zero/thirty zero/thirty 1022 00:57:30,150 --> 00:57:33,320 So there might be a leapfrog leaping out anyway. 1023 00:57:33,320 --> 00:57:38,310 I think what the issue is that cyclists 1024 00:57:38,310 --> 00:57:40,710 feel that they might get squeezed [? there. ?] 1025 00:57:40,710 --> 00:57:41,560 There's two issues. 1026 00:57:41,560 --> 00:57:43,210 One is that if you just have-- 1027 00:57:43,210 --> 00:57:46,140 I think for really short segments, these makes sense. 1028 00:57:46,140 --> 00:57:52,080 For [INAUDIBLE] you're probably thinking of [INAUDIBLE].. 1029 00:57:52,080 --> 00:57:54,090 A couple hundred yards with a huge bottleneck. 1030 00:57:54,090 --> 00:57:57,360 And frankly, if a bus had to follow a cyclist 1031 00:57:57,360 --> 00:57:59,340 at 8 miles an hour, [INAUDIBLE] 1032 00:57:59,340 --> 00:57:59,890 PROFESSOR: It wouldn't matter. 1033 00:57:59,890 --> 00:58:02,223 AUDIENCE: It would still be much faster than a car going 1034 00:58:02,223 --> 00:58:03,480 two or three lights. 1035 00:58:03,480 --> 00:58:09,110 But in longer corridors, especially where 1036 00:58:09,110 --> 00:58:12,120 you have the bus pulling from the curve and back, 1037 00:58:12,120 --> 00:58:14,330 it becomes more of an issue. 1038 00:58:14,330 --> 00:58:16,350 There's the question of whether you could build 1039 00:58:16,350 --> 00:58:17,680 some kind of floating bus stop. 1040 00:58:17,680 --> 00:58:19,430 So you'd have separation at the bus stops. 1041 00:58:19,430 --> 00:58:22,120 And the vehicles would never have to cross paths. 1042 00:58:22,120 --> 00:58:23,930 And the buses could pass the cyclists. 1043 00:58:23,930 --> 00:58:26,690 And the cyclists could pass the buses where needed. 1044 00:58:26,690 --> 00:58:31,337 But that's more, I think, cities sort of see this as a panacea 1045 00:58:31,337 --> 00:58:33,420 where they can lay down some [INAUDIBLE] for that. 1046 00:58:33,420 --> 00:58:35,037 And it's not a problem [INAUDIBLE] 1047 00:58:35,037 --> 00:58:36,620 mention that in [INAUDIBLE] in Madison 1048 00:58:36,620 --> 00:58:38,830 they have shared bus bike lanes. 1049 00:58:38,830 --> 00:58:40,510 And they work pretty well. 1050 00:58:43,100 --> 00:58:44,620 And I don't disagree. 1051 00:58:44,620 --> 00:58:46,690 But those are also some five lane wide streets 1052 00:58:46,690 --> 00:58:51,224 where they can provide a really wide facility. 1053 00:58:51,224 --> 00:58:51,890 PROFESSOR: Yeah. 1054 00:58:51,890 --> 00:58:54,920 It does depend on the width of the lane. 1055 00:58:54,920 --> 00:58:59,270 I agree [INAUDIBLE] agree that we could do a lot with queue 1056 00:58:59,270 --> 00:59:00,110 jump lanes-- 1057 00:59:00,110 --> 00:59:03,290 not fully dedicated lanes. 1058 00:59:03,290 --> 00:59:08,240 We can't give up a full lane all the way up and down Mass Ave. 1059 00:59:08,240 --> 00:59:10,730 But in front of MIT, we can. 1060 00:59:10,730 --> 00:59:14,480 From the river to Albany Street, we can. 1061 00:59:14,480 --> 00:59:18,290 Because there's no commercial activity. 1062 00:59:18,290 --> 00:59:23,825 We have 17 parking spaces out here in front of 77 Mass Ave. 1063 00:59:23,825 --> 00:59:24,710 17 spaces. 1064 00:59:24,710 --> 00:59:27,740 You can put them on Amherst Alley really easily 1065 00:59:27,740 --> 00:59:29,510 and have all this space that we could 1066 00:59:29,510 --> 00:59:31,340 devote for bikes and buses. 1067 00:59:31,340 --> 00:59:35,780 So what I would argue is, we need queue jump lanes. 1068 00:59:35,780 --> 00:59:37,910 A lot of people talk about-- 1069 00:59:37,910 --> 00:59:41,100 did Peter Firth talk about TSP yet? 1070 00:59:41,100 --> 00:59:41,600 OK. 1071 00:59:41,600 --> 00:59:42,170 He's coming. 1072 00:59:42,170 --> 00:59:45,860 Peter's going to come and talk about Transit Signal Priority. 1073 00:59:45,860 --> 00:59:48,290 And listen carefully. 1074 00:59:48,290 --> 00:59:52,760 Because transit signal priority provides very, very small 1075 00:59:52,760 --> 00:59:58,310 improvements, especially if the bus is stuck behind a car. 1076 00:59:58,310 --> 01:00:04,970 And so, the question is, transit signal priority 1077 01:00:04,970 --> 01:00:09,260 works well when at the congested intersections 1078 01:00:09,260 --> 01:00:12,530 we have a queue jump lane that the bus can get around 1079 01:00:12,530 --> 01:00:17,030 the long line of cars that might be there in the rush hour. 1080 01:00:17,030 --> 01:00:19,360 And so we need both. 1081 01:00:19,360 --> 01:00:23,180 And what I would argue is that in those small segments, 1082 01:00:23,180 --> 01:00:26,628 the buses can share it with bikes. 1083 01:00:26,628 --> 01:00:30,090 AUDIENCE: [INAUDIBLE] 1084 01:00:30,090 --> 01:00:32,600 PROFESSOR: Enforcement. 1085 01:00:32,600 --> 01:00:33,210 Yeah. 1086 01:00:33,210 --> 01:00:34,210 I mean that's the only-- 1087 01:00:34,210 --> 01:00:38,570 AUDIENCE: In San Francisco [INAUDIBLE] 1088 01:00:38,570 --> 01:00:40,830 driver literally busted out camera 1089 01:00:40,830 --> 01:00:42,400 in front of driver [INAUDIBLE] 1090 01:00:42,400 --> 01:00:43,750 PROFESSOR: They can push. 1091 01:00:43,750 --> 01:00:44,250 Yeah. 1092 01:00:44,250 --> 01:00:45,964 AUDIENCE: Take pictures [INAUDIBLE] 1093 01:00:45,964 --> 01:00:46,630 PROFESSOR: Yeah. 1094 01:00:46,630 --> 01:00:48,040 I think a lot of people do that. 1095 01:00:48,040 --> 01:00:48,850 Yeah. 1096 01:00:48,850 --> 01:00:51,550 I think London does the same thing. 1097 01:00:51,550 --> 01:00:53,280 London has a lot of these. 1098 01:00:53,280 --> 01:00:54,406 And they-- 1099 01:00:54,406 --> 01:00:56,114 AUDIENCE: [INAUDIBLE] Bus lanes and there 1100 01:00:56,114 --> 01:00:57,330 are also taxi lanes [INAUDIBLE] 1101 01:00:57,330 --> 01:00:58,954 PROFESSOR: And they're also bike lanes. 1102 01:00:58,954 --> 01:01:03,430 And in London, it looks scary to me. 1103 01:01:03,430 --> 01:01:05,420 Because they go so fast. 1104 01:01:05,420 --> 01:01:09,010 But around here, as Ari says, you 1105 01:01:09,010 --> 01:01:13,150 know if we get seven or eight miles an hour out of a bus, 1106 01:01:13,150 --> 01:01:14,560 we're happy. 1107 01:01:14,560 --> 01:01:16,740 Because they're only average four or five. 1108 01:01:16,740 --> 01:01:17,240 [LAUGHING] 1109 01:01:17,240 --> 01:01:19,700 AUDIENCE: That seven or eight miles an hour 1110 01:01:19,700 --> 01:01:22,226 is enough to get to a light that's actually 1111 01:01:22,226 --> 01:01:25,802 held green for ten seconds so if that light changes [INAUDIBLE].. 1112 01:01:30,430 --> 01:01:32,394 AUDIENCE: I don't see why that problem is 1113 01:01:32,394 --> 01:01:33,955 unique to the facility. 1114 01:01:33,955 --> 01:01:39,210 Because even if the bus is sharing a lane of traffic 1115 01:01:39,210 --> 01:01:40,984 and the bike lane is not a bike lane, 1116 01:01:40,984 --> 01:01:42,442 there's still [INAUDIBLE] conflict. 1117 01:01:44,540 --> 01:01:46,010 AUDIENCE: That's certainly true. 1118 01:01:46,010 --> 01:01:49,020 I think what the issue is that cyclists-- 1119 01:01:49,020 --> 01:01:50,910 some cyclists can be very special. 1120 01:01:50,910 --> 01:01:53,529 And I'm not talking about anyone who works at the City 1121 01:01:53,529 --> 01:01:54,112 of Cambridge.. 1122 01:01:57,030 --> 01:02:01,140 There are some cycling advocates who would rather-- 1123 01:02:01,140 --> 01:02:03,020 there are some cyclists and cycling advocates 1124 01:02:03,020 --> 01:02:06,130 who don't understand that there were transit issues 1125 01:02:06,130 --> 01:02:08,320 before traffic [INAUDIBLE]. 1126 01:02:08,320 --> 01:02:13,970 And there needs to be a bit of a detente between the cyclists 1127 01:02:13,970 --> 01:02:15,204 and the transit users. 1128 01:02:15,204 --> 01:02:17,120 The problem-- I think Frank pointed this out-- 1129 01:02:17,120 --> 01:02:19,244 is that the cyclists are meant for [INAUDIBLE] more 1130 01:02:19,244 --> 01:02:22,990 organizers than cyclists who are put in mortal danger every time 1131 01:02:22,990 --> 01:02:23,990 they're going somewhere. 1132 01:02:23,990 --> 01:02:27,400 And transit users are just like [INAUDIBLE].. 1133 01:02:27,400 --> 01:02:31,100 And there's a difference between those two groups. 1134 01:02:31,100 --> 01:02:35,930 And I know that this past November, I 1135 01:02:35,930 --> 01:02:40,130 think in the past fall I'm on the Cambridge Cycling Committee 1136 01:02:40,130 --> 01:02:42,650 and I try to be a voice of reason 1137 01:02:42,650 --> 01:02:45,100 for transit as well on those bicycle committees. 1138 01:02:45,100 --> 01:02:45,650 [INAUDIBLE] 1139 01:02:45,650 --> 01:02:47,150 PROFESSOR: The only voice of reason. 1140 01:02:47,150 --> 01:02:47,650 No. 1141 01:02:47,650 --> 01:02:48,670 [LAUGHING] 1142 01:02:48,670 --> 01:02:51,610 AUDIENCE: [INAUDIBLE] sometimes there 1143 01:02:51,610 --> 01:02:58,310 can be [INAUDIBLE] bus stops on Mass Ave. 1144 01:02:58,310 --> 01:03:01,430 PROFESSOR: Because 40% of the people on Mass Ave in the rush 1145 01:03:01,430 --> 01:03:02,180 hour are on buses. 1146 01:03:04,510 --> 01:03:05,010 Yeah. 1147 01:03:05,010 --> 01:03:08,327 AUDIENCE: And 50% of them are on bikes. 1148 01:03:08,327 --> 01:03:09,410 PROFESSOR: It's even more. 1149 01:03:09,410 --> 01:03:09,910 Yeah. 1150 01:03:09,910 --> 01:03:11,210 It's almost 50%. 1151 01:03:11,210 --> 01:03:12,680 Yeah. 1152 01:03:12,680 --> 01:03:14,844 And a lot of the arterials around here, 1153 01:03:14,844 --> 01:03:15,760 that's the percentage. 1154 01:03:15,760 --> 01:03:19,310 It's between 35% and 50% of the people are on buses. 1155 01:03:19,310 --> 01:03:21,893 AUDIENCE: [INAUDIBLE] The bike doesn't take up that much room. 1156 01:03:26,910 --> 01:03:32,297 So the issue, I think, winds up being that-- 1157 01:03:32,297 --> 01:03:33,630 PROFESSOR: We're off topic here. 1158 01:03:33,630 --> 01:03:34,510 [LAUGHING] 1159 01:03:34,510 --> 01:03:36,052 AUDIENCE: We had a nice conversation. 1160 01:03:36,052 --> 01:03:38,093 I think when you get people in the same room they 1161 01:03:38,093 --> 01:03:40,320 start to understand that there's a give and take. 1162 01:03:40,320 --> 01:03:41,300 [INAUDIBLE] and I said, if you're 1163 01:03:41,300 --> 01:03:42,990 doing planning an isolation where you talk 1164 01:03:42,990 --> 01:03:45,130 to bike people and they, say, we want X and you talk to transit 1165 01:03:45,130 --> 01:03:47,640 people and they say they want Y, the easiest thing 1166 01:03:47,640 --> 01:03:48,348 is to do nothing. 1167 01:03:48,348 --> 01:03:51,300 [INAUDIBLE] 1168 01:03:51,300 --> 01:03:53,040 PROFESSOR: Yeah. 1169 01:03:53,040 --> 01:03:57,210 I think the bus lane actually gets-- 1170 01:03:57,210 --> 01:03:59,880 although, these days I'm not sure-- 1171 01:03:59,880 --> 01:04:05,130 but I would say that the bus lanes get more adverse reaction 1172 01:04:05,130 --> 01:04:07,560 from the drivers than the bike lanes do. 1173 01:04:07,560 --> 01:04:11,490 Because they don't seem to take away driving lanes. 1174 01:04:11,490 --> 01:04:14,010 Whereas a bus lane seems to either take 1175 01:04:14,010 --> 01:04:16,770 a driving or a parking lane away. 1176 01:04:16,770 --> 01:04:21,300 And so, it gets more opposition from the drivers. 1177 01:04:21,300 --> 01:04:24,990 But there's a quote. 1178 01:04:24,990 --> 01:04:27,580 I forget the quote now. 1179 01:04:27,580 --> 01:04:32,110 But basically, urbanologists-- people-- 1180 01:04:32,110 --> 01:04:34,440 have started to realize that we're 1181 01:04:34,440 --> 01:04:37,920 spending tons and tons of money doing bike facilities 1182 01:04:37,920 --> 01:04:39,360 in these urban areas. 1183 01:04:39,360 --> 01:04:42,570 And we're spending nothing on transit facilities. 1184 01:04:42,570 --> 01:04:46,320 And that's a long-term disaster. 1185 01:04:46,320 --> 01:04:50,760 Because essentially, there is a limit 1186 01:04:50,760 --> 01:04:54,780 on the amount of bike use, I think, in any of these places. 1187 01:04:54,780 --> 01:04:56,700 Because there's lots of people. 1188 01:04:56,700 --> 01:04:57,780 I used to bike. 1189 01:04:57,780 --> 01:04:59,580 I live four miles away. 1190 01:04:59,580 --> 01:05:00,240 I used to bike. 1191 01:05:00,240 --> 01:05:01,350 I'm getting old. 1192 01:05:01,350 --> 01:05:06,851 So I'm losing a little bit of my perception and stuff like that. 1193 01:05:06,851 --> 01:05:08,100 I had a couple of close calls. 1194 01:05:08,100 --> 01:05:09,330 I stopped. 1195 01:05:09,330 --> 01:05:10,920 I basically just stopped. 1196 01:05:10,920 --> 01:05:12,600 I said, I'll bike when I go to Cape Cod. 1197 01:05:12,600 --> 01:05:14,490 I'm not going to bike in these streets. 1198 01:05:14,490 --> 01:05:16,410 Mainly I was afraid of the other bikers 1199 01:05:16,410 --> 01:05:21,670 then I was as opposed to being afraid of the cars. 1200 01:05:21,670 --> 01:05:25,700 And there's lots of people that just can't. 1201 01:05:25,700 --> 01:05:27,630 And a lot of those people want to take buses. 1202 01:05:27,630 --> 01:05:31,590 But they also don't want to be held up in congestion streets 1203 01:05:31,590 --> 01:05:32,860 all the time. 1204 01:05:32,860 --> 01:05:35,530 So can we switch? 1205 01:05:35,530 --> 01:05:37,120 I don't know your computer. 1206 01:05:37,120 --> 01:05:38,700 Can we switch to the other one? 1207 01:05:38,700 --> 01:05:41,640 We'll just get started on the Performance Measurement. 1208 01:05:41,640 --> 01:05:45,750 And then we'll go into the next one. 1209 01:05:51,195 --> 01:05:52,690 Great. 1210 01:05:52,690 --> 01:05:55,770 The reading, I think, is important. 1211 01:05:58,890 --> 01:06:01,290 It's a dry topic. 1212 01:06:01,290 --> 01:06:05,340 It's how transit authorities evaluate themselves 1213 01:06:05,340 --> 01:06:08,400 and how they evaluate themselves with respect to other transit 1214 01:06:08,400 --> 01:06:10,510 authorities. 1215 01:06:10,510 --> 01:06:11,670 But it's important. 1216 01:06:11,670 --> 01:06:17,820 Because if they don't do performance measurement, 1217 01:06:17,820 --> 01:06:19,050 nobody improves. 1218 01:06:19,050 --> 01:06:20,970 And there's example after example. 1219 01:06:20,970 --> 01:06:24,810 The two best examples, from my perspective, 1220 01:06:24,810 --> 01:06:27,240 are the CTA, about five years ago, 1221 01:06:27,240 --> 01:06:31,950 instituted this rigorous performance measurements regime 1222 01:06:31,950 --> 01:06:39,000 that the general manager sat at a weekly meeting 1223 01:06:39,000 --> 01:06:41,160 on a rotating basis with every department 1224 01:06:41,160 --> 01:06:42,810 and developed performance measures 1225 01:06:42,810 --> 01:06:45,180 and tracked them for each department. 1226 01:06:45,180 --> 01:06:49,810 And the performance, almost universally, went up. 1227 01:06:49,810 --> 01:06:52,470 There's a set of slides that I can't go over. 1228 01:06:52,470 --> 01:06:56,850 But we'll put on Stellar about this. 1229 01:06:56,850 --> 01:07:01,660 It's an amazing story, I think. 1230 01:07:01,660 --> 01:07:03,600 The GM was only there for two years. 1231 01:07:03,600 --> 01:07:07,770 He was one of these management by data types. 1232 01:07:07,770 --> 01:07:10,680 And then he got moved to run the school department, 1233 01:07:10,680 --> 01:07:11,400 for God's sake. 1234 01:07:11,400 --> 01:07:13,470 I don't know why. 1235 01:07:13,470 --> 01:07:15,120 But, anyway. 1236 01:07:15,120 --> 01:07:18,630 And the other example really is the MBTA 1237 01:07:18,630 --> 01:07:22,740 under the fiscal management control board, where they've 1238 01:07:22,740 --> 01:07:25,980 insisted that there be regular updates 1239 01:07:25,980 --> 01:07:29,820 and data-driven analyses. 1240 01:07:29,820 --> 01:07:35,220 And if you look at the progress on things like Overtime, 1241 01:07:35,220 --> 01:07:42,000 and essentially, things that are inefficient, generally, 1242 01:07:42,000 --> 01:07:43,410 they've improved. 1243 01:07:43,410 --> 01:07:46,710 And there's a lot more to improve here at the MBTA. 1244 01:07:46,710 --> 01:07:49,170 And there's a lot of issues that can't 1245 01:07:49,170 --> 01:07:53,310 be solved just by better management and performance 1246 01:07:53,310 --> 01:07:54,560 measurement. 1247 01:07:54,560 --> 01:07:56,100 But there's a lot that can. 1248 01:07:56,100 --> 01:08:02,560 And so, I think this is a really important topic. 1249 01:08:02,560 --> 01:08:07,320 So what we have here is a little bit 1250 01:08:07,320 --> 01:08:09,720 of justification of doing this. 1251 01:08:09,720 --> 01:08:12,480 There is an older report that's really good-- 1252 01:08:12,480 --> 01:08:17,430 TCRP 88 that essentially is a cookbook 1253 01:08:17,430 --> 01:08:19,200 on how to develop a performance management 1254 01:08:19,200 --> 01:08:21,270 system for your transit agency. 1255 01:08:21,270 --> 01:08:25,290 And it's a really good report with respect to the measures 1256 01:08:25,290 --> 01:08:26,220 you use. 1257 01:08:26,220 --> 01:08:29,970 And they divide into small, medium, and large urban areas. 1258 01:08:29,970 --> 01:08:34,380 And they give different measures for four different ones. 1259 01:08:34,380 --> 01:08:36,910 A lot of people want to compare themselves to other people. 1260 01:08:36,910 --> 01:08:38,640 There's this national database. 1261 01:08:38,640 --> 01:08:42,689 And this is the website. 1262 01:08:42,689 --> 01:08:45,060 And you can see performance measures 1263 01:08:45,060 --> 01:08:49,290 from everyone in the country on a pretty much 1264 01:08:49,290 --> 01:08:52,779 equitable reasonable basis. 1265 01:08:52,779 --> 01:08:56,130 And then there's the concept of performance audits, which 1266 01:08:56,130 --> 01:08:58,260 we'll talk a little bit about. 1267 01:08:58,260 --> 01:09:02,490 So why do it? 1268 01:09:02,490 --> 01:09:04,410 Basically, whatever gets measured 1269 01:09:04,410 --> 01:09:07,700 gets noticed by management. 1270 01:09:07,700 --> 01:09:11,340 And everybody's into these new management dashboards 1271 01:09:11,340 --> 01:09:13,900 with key performance indicators. 1272 01:09:16,620 --> 01:09:20,200 It can be done and not have an impact. 1273 01:09:20,200 --> 01:09:24,180 But I feel like, whatever that indicator 1274 01:09:24,180 --> 01:09:26,609 is, people manage to it. 1275 01:09:26,609 --> 01:09:29,160 Sometimes the indicator is the wrong indicator. 1276 01:09:29,160 --> 01:09:31,319 And it's not so good. 1277 01:09:31,319 --> 01:09:36,840 Because it's the easy one to measure as opposed to what 1278 01:09:36,840 --> 01:09:39,029 they should be focused on. 1279 01:09:39,029 --> 01:09:41,399 Many of them are-- 1280 01:09:41,399 --> 01:09:45,630 the measures that the board and the top management look at 1281 01:09:45,630 --> 01:09:48,750 are oriented to the supply of the transit-- 1282 01:09:48,750 --> 01:09:51,300 what the vehicles did as opposed to what 1283 01:09:51,300 --> 01:09:55,380 the passenger experienced. 1284 01:09:55,380 --> 01:09:59,550 The best example of that is when you 1285 01:09:59,550 --> 01:10:03,750 look at on-time performance. 1286 01:10:03,750 --> 01:10:05,370 They measure it within 2 or 3 minutes. 1287 01:10:07,880 --> 01:10:12,030 And that doesn't mean anything if the bus is supposed 1288 01:10:12,030 --> 01:10:14,310 to come every 7 or 8 minutes. 1289 01:10:14,310 --> 01:10:21,990 Because basically, there's this phenomenon of long gaps 1290 01:10:21,990 --> 01:10:24,330 and then followed by bunches. 1291 01:10:24,330 --> 01:10:27,660 There's very few passengers on the bunch buses, especially 1292 01:10:27,660 --> 01:10:30,830 the second and third one. 1293 01:10:30,830 --> 01:10:34,260 And most of the people are on the crowded bus 1294 01:10:34,260 --> 01:10:37,080 who had to wait, which is the first one. 1295 01:10:37,080 --> 01:10:39,600 And they had to wait much longer for that. 1296 01:10:39,600 --> 01:10:45,360 So the wait time for passengers is not 1297 01:10:45,360 --> 01:10:49,410 linearly related to this on-time performance measure 1298 01:10:49,410 --> 01:10:50,590 that they have. 1299 01:10:50,590 --> 01:10:54,270 So you actually should be trying to measure effective wait 1300 01:10:54,270 --> 01:10:58,210 time for all passengers. 1301 01:10:58,210 --> 01:11:00,510 And that should be your measure for most 1302 01:11:00,510 --> 01:11:03,380 of the transit services. 1303 01:11:03,380 --> 01:11:07,100 They also use some of the performance measures 1304 01:11:07,100 --> 01:11:09,450 on a system basis for funding allocation. 1305 01:11:09,450 --> 01:11:11,870 I'll mention a few things there. 1306 01:11:11,870 --> 01:11:14,525 I talked about this TCRP Report. 1307 01:11:18,260 --> 01:11:20,180 It's nice. 1308 01:11:20,180 --> 01:11:24,276 And I think it used to have a website associated with it. 1309 01:11:24,276 --> 01:11:25,650 So you could actually go in there 1310 01:11:25,650 --> 01:11:27,290 and put your own program together. 1311 01:11:27,290 --> 01:11:28,790 I'm not sure if it still does. 1312 01:11:28,790 --> 01:11:30,110 Because it's fairly old. 1313 01:11:30,110 --> 01:11:36,470 But it's nice in that it isn't one prescribed thing. 1314 01:11:36,470 --> 01:11:38,450 It gives you a lot of different options. 1315 01:11:38,450 --> 01:11:41,300 And you can tailor it to your urban area. 1316 01:11:45,120 --> 01:11:50,490 This is the process that they recommend in the report. 1317 01:11:50,490 --> 01:11:53,460 Again, this is an internal performance measurement 1318 01:11:53,460 --> 01:11:53,970 system-- 1319 01:11:53,970 --> 01:11:56,190 performance management system. 1320 01:11:56,190 --> 01:12:02,190 And it's a top-down to find your goals and objectives get 1321 01:12:02,190 --> 01:12:03,300 management support. 1322 01:12:03,300 --> 01:12:08,590 And I guess, the second one is extremely important. 1323 01:12:08,590 --> 01:12:12,600 I used to, as my olden days as a consultant, 1324 01:12:12,600 --> 01:12:16,800 I used to do transit performance audits for transit agencies. 1325 01:12:16,800 --> 01:12:21,570 And the experiences varied further 1326 01:12:21,570 --> 01:12:24,300 than anyone could imagine. 1327 01:12:24,300 --> 01:12:26,490 I'd been in some transit authorities. 1328 01:12:26,490 --> 01:12:28,870 The first meeting was with the general manager. 1329 01:12:28,870 --> 01:12:33,170 And he says, go get 'em. 1330 01:12:33,170 --> 01:12:35,490 And he says, and come back with all the information, 1331 01:12:35,490 --> 01:12:37,400 because I want to know it. 1332 01:12:37,400 --> 01:12:39,920 And you would do it. 1333 01:12:39,920 --> 01:12:42,590 And we'd come back with the indicators. 1334 01:12:42,590 --> 01:12:45,480 And he'd say, OK, what do I do about this? 1335 01:12:45,480 --> 01:12:46,810 What I do about this? 1336 01:12:46,810 --> 01:12:48,590 He'd bring the departments in and says, 1337 01:12:48,590 --> 01:12:50,600 I'm not blaming anybody. 1338 01:12:50,600 --> 01:12:52,370 We just gotta develop action plans 1339 01:12:52,370 --> 01:12:54,140 to get these things better. 1340 01:12:54,140 --> 01:12:56,150 And I've been in other properties 1341 01:12:56,150 --> 01:12:59,270 where the general manager said, literally, 1342 01:12:59,270 --> 01:13:02,420 on the first meeting, you're here 1343 01:13:02,420 --> 01:13:06,090 because the state makes me do this. 1344 01:13:06,090 --> 01:13:11,090 So I want you to get in and out of here as quick as you can. 1345 01:13:11,090 --> 01:13:15,470 And basically, I know how to manage my place. 1346 01:13:15,470 --> 01:13:17,260 You don't have to tell me anything. 1347 01:13:17,260 --> 01:13:23,350 And now, all of these were relatively small properties. 1348 01:13:23,350 --> 01:13:25,520 We did some larger properties. 1349 01:13:25,520 --> 01:13:30,430 But basically, there's a tremendous variation 1350 01:13:30,430 --> 01:13:32,920 among management styles on this. 1351 01:13:32,920 --> 01:13:36,940 And so, if you're going to do this, 1352 01:13:36,940 --> 01:13:40,060 you want top management to be supportive right from the start 1353 01:13:40,060 --> 01:13:42,940 and to participate in developing it. 1354 01:13:45,772 --> 01:13:49,250 The rest of it is intuitive. 1355 01:13:49,250 --> 01:13:52,160 The most important thing is, how do 1356 01:13:52,160 --> 01:13:54,920 the results, when you update all these measures, 1357 01:13:54,920 --> 01:13:58,190 get integrated into how you actually 1358 01:13:58,190 --> 01:13:59,990 carry out the business? 1359 01:13:59,990 --> 01:14:03,300 And that is a hard part sometimes. 1360 01:14:03,300 --> 01:14:07,190 And it's a part that you need to think about. 1361 01:14:07,190 --> 01:14:11,210 And you gotta think about, what can I do about this indicator? 1362 01:14:11,210 --> 01:14:13,970 And I think part of when you're deciding 1363 01:14:13,970 --> 01:14:17,870 what the indicators are, that's a big part is thinking 1364 01:14:17,870 --> 01:14:21,190 about that step 7 and 8. 1365 01:14:21,190 --> 01:14:24,050 And essentially, if something's working 1366 01:14:24,050 --> 01:14:25,880 and you're making progress, you keep it in. 1367 01:14:25,880 --> 01:14:31,130 If you don't, an 8, you throw it out and try something else. 1368 01:14:31,130 --> 01:14:37,850 This is the way they've got lists and lists of measures. 1369 01:14:37,850 --> 01:14:39,790 They've got bunches of tables. 1370 01:14:39,790 --> 01:14:43,910 And again, they have them for large properties. 1371 01:14:43,910 --> 01:14:46,790 It makes sense for large properties, medium, small, 1372 01:14:46,790 --> 01:14:49,850 actually under 50,000 people in the metropolitan area. 1373 01:14:49,850 --> 01:14:52,580 So they basically say, this is kind 1374 01:14:52,580 --> 01:14:54,800 of what we think your capabilities are 1375 01:14:54,800 --> 01:15:00,180 based on staffing and these kind of properties. 1376 01:15:00,180 --> 01:15:01,210 There's some more. 1377 01:15:01,210 --> 01:15:02,750 I'm not going to go through these. 1378 01:15:02,750 --> 01:15:06,190 The NDT data. 1379 01:15:06,190 --> 01:15:08,320 How many people have actually gone in 1380 01:15:08,320 --> 01:15:09,830 and looked at these tables? 1381 01:15:09,830 --> 01:15:11,230 Anybody? 1382 01:15:11,230 --> 01:15:12,330 Two. 1383 01:15:12,330 --> 01:15:13,470 OK. 1384 01:15:13,470 --> 01:15:14,820 You should all do this. 1385 01:15:14,820 --> 01:15:18,180 It's pretty interesting. 1386 01:15:18,180 --> 01:15:20,910 They've got a whole range of transit properties 1387 01:15:20,910 --> 01:15:22,660 across the country. 1388 01:15:22,660 --> 01:15:25,530 And you can sort them in various ways. 1389 01:15:25,530 --> 01:15:26,850 You can download reports. 1390 01:15:26,850 --> 01:15:29,760 You can download it in CSV files. 1391 01:15:29,760 --> 01:15:33,890 You can see how your city stacks up against other cities. 1392 01:15:33,890 --> 01:15:36,460 Consultants use these all the time. 1393 01:15:36,460 --> 01:15:38,760 And the good thing about it is at the beginning when 1394 01:15:38,760 --> 01:15:41,880 they first developed this, definitions 1395 01:15:41,880 --> 01:15:44,760 were variable across cities. 1396 01:15:44,760 --> 01:15:49,200 And each year, the government has done 1397 01:15:49,200 --> 01:15:51,420 what big bureaucracies do well. 1398 01:15:51,420 --> 01:15:55,020 They started to standardize this. 1399 01:15:55,020 --> 01:15:57,910 And they call you up if they don't like your number. 1400 01:15:57,910 --> 01:16:00,300 And they also call you up if your numbers change 1401 01:16:00,300 --> 01:16:02,400 from year to year too much. 1402 01:16:02,400 --> 01:16:03,930 And they ask you why. 1403 01:16:03,930 --> 01:16:06,300 And so there's an auditing process 1404 01:16:06,300 --> 01:16:09,040 of the data that's in here. 1405 01:16:09,040 --> 01:16:12,880 Now the only downside of it is it 1406 01:16:12,880 --> 01:16:17,810 got big quick when they first developed it in the 80s. 1407 01:16:17,810 --> 01:16:20,890 And then there was a backlash from the transit agencies 1408 01:16:20,890 --> 01:16:23,070 in the city saying, you're asking us 1409 01:16:23,070 --> 01:16:24,250 for too much information. 1410 01:16:24,250 --> 01:16:26,900 This costs too much. 1411 01:16:26,900 --> 01:16:33,610 And so, I don't think they've added the proper updating of it 1412 01:16:33,610 --> 01:16:39,310 with respect to developments that have been in the transit 1413 01:16:39,310 --> 01:16:42,270 over the last 10 or 15 years or so. 1414 01:16:45,500 --> 01:16:48,670 And there still is some inconsistency. 1415 01:16:48,670 --> 01:16:51,260 There's still a little bit of latitude 1416 01:16:51,260 --> 01:16:54,019 on how you allocate your costs. 1417 01:16:54,019 --> 01:16:55,060 That's the hardest thing. 1418 01:16:55,060 --> 01:16:59,670 And have you guys had cost allocation models? 1419 01:16:59,670 --> 01:17:00,770 Yeah. 1420 01:17:00,770 --> 01:17:05,810 So you saw the uncertainty of exactly what you put-- 1421 01:17:05,810 --> 01:17:09,530 all these costs-- what category you put them in. 1422 01:17:09,530 --> 01:17:14,240 And so, that part, while they define a lot of it, 1423 01:17:14,240 --> 01:17:17,440 they don't define it down to the bottom. 1424 01:17:17,440 --> 01:17:22,550 And so, I think that there is, especially in the costs. 1425 01:17:22,550 --> 01:17:25,490 So the cost effectiveness numbers that come out of 1426 01:17:25,490 --> 01:17:30,180 here aren't necessarily directly comparable 1427 01:17:30,180 --> 01:17:34,310 depending on the property and how seriously they take it. 1428 01:17:34,310 --> 01:17:36,890 The other issue with this is this 1429 01:17:36,890 --> 01:17:39,770 is another thing where the management style-- 1430 01:17:39,770 --> 01:17:42,500 some management actually ask their people 1431 01:17:42,500 --> 01:17:46,010 to look at other properties here and report back. 1432 01:17:46,010 --> 01:17:47,490 And they care about it. 1433 01:17:47,490 --> 01:17:50,000 And other management says, just get that report done. 1434 01:17:50,000 --> 01:17:52,020 Get it out of here. 1435 01:17:52,020 --> 01:17:55,340 So this is a similar thing. 1436 01:17:55,340 --> 01:17:56,660 Leadership is important. 1437 01:18:00,220 --> 01:18:06,360 So this issue of performance audits. 1438 01:18:06,360 --> 01:18:11,730 So outside of doing it with key performance indicators going 1439 01:18:11,730 --> 01:18:14,580 to the board, there's also these efforts 1440 01:18:14,580 --> 01:18:19,950 to do it comprehensively, maybe a deeper dive, than you 1441 01:18:19,950 --> 01:18:22,470 would do on a regular basis. 1442 01:18:22,470 --> 01:18:25,620 And they usually bring in outside people consultants 1443 01:18:25,620 --> 01:18:29,520 or maybe a state group that comes in. 1444 01:18:29,520 --> 01:18:32,130 And they go around, especially, like, say, 1445 01:18:32,130 --> 01:18:35,700 for the RTAs here in Massachusetts that are all much 1446 01:18:35,700 --> 01:18:37,500 smaller than the MBTA-- 1447 01:18:37,500 --> 01:18:39,480 they have a standardized performance audit 1448 01:18:39,480 --> 01:18:41,880 group that goes around. 1449 01:18:41,880 --> 01:18:44,700 And they give them information. 1450 01:18:44,700 --> 01:18:47,310 I think the big issue, especially in the smaller 1451 01:18:47,310 --> 01:18:51,540 properties, is they don't have the resources to find out 1452 01:18:51,540 --> 01:18:54,840 what other people are doing very easily as compared 1453 01:18:54,840 --> 01:18:57,482 to the larger cities who just call up people and say, 1454 01:18:57,482 --> 01:18:58,440 I'm going to come down. 1455 01:18:58,440 --> 01:18:59,820 I want to hear about this. 1456 01:18:59,820 --> 01:19:01,920 Or they set up a special panel and they 1457 01:19:01,920 --> 01:19:04,840 bring in some experts on a particular issue. 1458 01:19:04,840 --> 01:19:09,510 What this process allows is identifying critical issues 1459 01:19:09,510 --> 01:19:14,670 that are performance issues and then allowing the consultant 1460 01:19:14,670 --> 01:19:16,530 to focus just on those two or three 1461 01:19:16,530 --> 01:19:20,700 issues or the outside group and say, give us information 1462 01:19:20,700 --> 01:19:24,600 about how we can improve just in these areas. 1463 01:19:24,600 --> 01:19:28,920 And then it brings information from other properties. 1464 01:19:28,920 --> 01:19:32,249 Because these people have been doing it in other places. 1465 01:19:37,090 --> 01:19:42,020 And the big debate when this first all started 1466 01:19:42,020 --> 01:19:47,860 and among the people who did it was, should you focus on 1467 01:19:47,860 --> 01:19:52,840 internally in the trends in your internal operation or should 1468 01:19:52,840 --> 01:19:55,450 you focus on how you compare to other people? 1469 01:19:55,450 --> 01:19:59,440 And there's no the answer here. 1470 01:19:59,440 --> 01:20:03,520 I always said, it doesn't hurt to look at outside. 1471 01:20:03,520 --> 01:20:08,500 But clearly, if you're looking internally first, 1472 01:20:08,500 --> 01:20:10,900 you can see whether you're getting better or worse. 1473 01:20:10,900 --> 01:20:13,270 And you know the data is consistent. 1474 01:20:13,270 --> 01:20:17,500 And you know the issues that you're having trouble with. 1475 01:20:17,500 --> 01:20:22,730 Because intuitively, you know it as well as from the data. 1476 01:20:22,730 --> 01:20:30,880 So I would say, generally, the peer comparisons were 1477 01:20:30,880 --> 01:20:34,540 often used by the outside groups to say, where should we 1478 01:20:34,540 --> 01:20:36,730 focus going in? 1479 01:20:36,730 --> 01:20:39,470 And let's dig deeper in these areas, 1480 01:20:39,470 --> 01:20:43,615 because these numbers look a little bit out of context. 1481 01:20:46,670 --> 01:20:47,820 Almost done here. 1482 01:20:47,820 --> 01:20:51,530 So I'm going to press on for-- 1483 01:20:51,530 --> 01:20:53,480 to-- 1484 01:20:53,480 --> 01:20:57,440 So the bulk of the effort was internally. 1485 01:20:57,440 --> 01:21:00,020 One thing that this allows is because you 1486 01:21:00,020 --> 01:21:02,930 have this facilitator and the outside group, 1487 01:21:02,930 --> 01:21:08,710 it allows you to tell them, like a counselor, 1488 01:21:08,710 --> 01:21:11,380 this is why I really can't get my job done. 1489 01:21:11,380 --> 01:21:17,880 And we really developed a mode with these folks. 1490 01:21:17,880 --> 01:21:18,780 You can tell us. 1491 01:21:18,780 --> 01:21:20,880 We're not going to attribute it to you. 1492 01:21:20,880 --> 01:21:23,610 We're not going to necessarily go to your direct boss 1493 01:21:23,610 --> 01:21:25,560 about your complaint to the direct boss. 1494 01:21:25,560 --> 01:21:27,720 But we might mention this to somebody 1495 01:21:27,720 --> 01:21:31,150 higher up in a more general way. 1496 01:21:31,150 --> 01:21:33,660 And we had ways of assuring them. 1497 01:21:33,660 --> 01:21:37,260 And so, employees liked this, especially 1498 01:21:37,260 --> 01:21:41,970 if they were hitting roadblocks in doing a better job. 1499 01:21:41,970 --> 01:21:44,970 And in small properties, sometimes personalities 1500 01:21:44,970 --> 01:21:46,810 got in the way. 1501 01:21:46,810 --> 01:21:49,930 And so we would provide that feedback to the top management 1502 01:21:49,930 --> 01:21:51,370 as long as they wanted it. 1503 01:21:54,100 --> 01:21:58,960 And just I'm aware of a very significant state 1504 01:21:58,960 --> 01:22:00,640 program in California. 1505 01:22:00,640 --> 01:22:04,600 They have to do this every three years. 1506 01:22:04,600 --> 01:22:05,620 It started at two. 1507 01:22:05,620 --> 01:22:07,210 Now it's down to three. 1508 01:22:07,210 --> 01:22:08,830 And they have a set format. 1509 01:22:08,830 --> 01:22:13,390 And they send the same kind of people around to each one. 1510 01:22:13,390 --> 01:22:18,790 And there's some compliance with state requirements 1511 01:22:18,790 --> 01:22:21,280 and funding and things like that. 1512 01:22:21,280 --> 01:22:24,460 But it really was a comprehensive. 1513 01:22:24,460 --> 01:22:27,770 And they looked in these five focus areas. 1514 01:22:27,770 --> 01:22:31,960 And they essentially-- when you got the feedback 1515 01:22:31,960 --> 01:22:34,450 from most of the smaller properties, at least, 1516 01:22:34,450 --> 01:22:36,340 they felt like this was a great program. 1517 01:22:36,340 --> 01:22:41,770 Because it gave them the ability to focus harder 1518 01:22:41,770 --> 01:22:45,230 on things on a periodic basis. 1519 01:22:45,230 --> 01:22:48,790 And then when you came back, you measured those areas and said, 1520 01:22:48,790 --> 01:22:54,810 OK, what else was going on at that point? 1521 01:22:54,810 --> 01:22:56,500 I think that's where I'm going to stop. 1522 01:22:56,500 --> 01:22:58,330 Yeah. 1523 01:22:58,330 --> 01:23:00,730 Anybody have any last questions? 1524 01:23:00,730 --> 01:23:04,581 We can pick this up a little bit in the service contracting 1525 01:23:04,581 --> 01:23:05,080 stuff. 1526 01:23:05,080 --> 01:23:07,240 Because it comes up there, too. 1527 01:23:07,240 --> 01:23:10,420 The difference between performance 1528 01:23:10,420 --> 01:23:15,160 in your internal operations and performance by your contractors 1529 01:23:15,160 --> 01:23:17,080 is nothing. 1530 01:23:17,080 --> 01:23:21,280 In fact, it's harder to manage the service contract. 1531 01:23:21,280 --> 01:23:25,510 And we can talk about that at the next one. 1532 01:23:25,510 --> 01:23:27,156 OK. 1533 01:23:27,156 --> 01:23:28,840 Anybody? 1534 01:23:28,840 --> 01:23:29,440 Great. 1535 01:23:29,440 --> 01:23:30,990 Thanks.