1 00:00:00,000 --> 00:00:11,759 [Music] 2 00:00:11,759 --> 00:00:14,080 Chain of events. Cause and effect. We 3 00:00:14,080 --> 00:00:15,519 analyse what went right and what went 4 00:00:15,519 --> 00:00:17,039 wrong, as we discover that many 5 00:00:17,039 --> 00:00:19,039 outcomes can be predicted, planned for, 6 00:00:19,039 --> 00:00:21,679 and even prevented. I'm John Chidgey and 7 00:00:21,679 --> 00:00:24,320 this is Causality. Causality is entirely 8 00:00:24,320 --> 00:00:26,559 supported by you our listeners. If you'd 9 00:00:26,559 --> 00:00:28,480 like to support us and keep the show ad- 10 00:00:28,480 --> 00:00:30,719 free you can by becoming a Premium 11 00:00:30,719 --> 00:00:32,479 Supporter. Premium support is available 12 00:00:32,479 --> 00:00:34,719 via Patreon, through the Apple Podcasts 13 00:00:34,719 --> 00:00:37,039 channel subscription and through Spotify. 14 00:00:37,039 --> 00:00:38,878 Premium supporters have access to high 15 00:00:38,878 --> 00:00:40,960 quality versions of episodes as well as 16 00:00:40,960 --> 00:00:43,200 bonus material from all of our shows not 17 00:00:43,200 --> 00:00:45,280 available anywhere else. Causality is 18 00:00:45,280 --> 00:00:47,840 also a Podcasting 2.0 enhanced show with 19 00:00:47,840 --> 00:00:49,920 value support and a new listener 20 00:00:49,920 --> 00:00:52,238 submitted sound bite option if you'd 21 00:00:52,238 --> 00:00:54,079 like to participate. Just visit 22 00:00:54,079 --> 00:00:56,558 https://engineered.network/causality to learn how 23 00:00:56,558 --> 00:00:58,640 you can help this show to continue to be 24 00:00:58,640 --> 00:01:00,479 made. Thank you! 25 00:01:00,479 --> 00:01:02,079 Hyatt Regency 26 00:01:02,079 --> 00:01:04,319 Kansas City. 27 00:01:04,319 --> 00:01:07,118 Construction of the 40-story high Hyatt 28 00:01:07,118 --> 00:01:09,438 Regency Hotel in Kansas City Missouri 29 00:01:09,438 --> 00:01:12,319 began in May of 1978 and it opened its 30 00:01:12,319 --> 00:01:14,239 doors to the public on the 1st of July 31 00:01:14,239 --> 00:01:16,239 1980 and remained the tallest building 32 00:01:16,239 --> 00:01:19,280 in Missouri until 1986. The hotel was 33 00:01:19,280 --> 00:01:21,280 situated in the Crown Center commercial 34 00:01:21,280 --> 00:01:23,118 complex which was part of an urban 35 00:01:23,118 --> 00:01:25,200 revitalisation of an older part of 36 00:01:25,200 --> 00:01:28,000 Kansas City near Union Station. The hotel 37 00:01:28,000 --> 00:01:29,759 had a revolving restaurant, an exhibit 38 00:01:29,759 --> 00:01:31,920 hall, conference facilities and more than 39 00:01:31,920 --> 00:01:33,920 700 guest rooms. 40 00:01:33,920 --> 00:01:36,078 One of the other widely publicised 41 00:01:36,078 --> 00:01:37,920 features of the new hotel was its 42 00:01:37,920 --> 00:01:40,239 multi-storey atrium which had three 43 00:01:40,239 --> 00:01:42,799 walkways each at a different floor level, 44 00:01:42,799 --> 00:01:45,599 each suspended from the ceiling above. 45 00:01:45,599 --> 00:01:47,759 The atrium has a large open area 46 00:01:47,759 --> 00:01:50,319 approximately 36m (that's 117 feet) 47 00:01:50,319 --> 00:01:54,159 deep 44m (or 145 feet) wide and 48 00:01:54,159 --> 00:01:57,519 15m (or 50 feet) high containing a bar, 49 00:01:57,519 --> 00:01:59,759 a stage, and a large open space that 50 00:01:59,759 --> 00:02:01,920 would sometimes be used as a dance floor. 51 00:02:01,920 --> 00:02:04,959 Each walkway was 47m (that's 121 52 00:02:04,959 --> 00:02:07,280 feet) long and was constructed from steel, 53 00:02:07,280 --> 00:02:09,199 concrete and glass weighing in at 54 00:02:09,199 --> 00:02:12,800 approximately 29 tonnes (or 64,000 pounds) 55 00:02:12,800 --> 00:02:14,080 each. 56 00:02:14,080 --> 00:02:15,759 The hotel itself was split into two 57 00:02:15,759 --> 00:02:18,000 sections: a high-rise section and a 58 00:02:18,000 --> 00:02:20,239 function or events block connected by 59 00:02:20,239 --> 00:02:21,520 the atrium. 60 00:02:21,520 --> 00:02:23,759 Walkways connected those two sections on 61 00:02:23,759 --> 00:02:26,080 levels 2, 3 and 4 with level 62 00:02:26,080 --> 00:02:28,878 3 offset from 2 and 4 such that 63 00:02:28,878 --> 00:02:31,039 levels 2 and 4 walkways were 64 00:02:31,039 --> 00:02:33,120 directly above and below each other, 65 00:02:33,120 --> 00:02:35,598 adjacent to the outside wall, with level 66 00:02:35,598 --> 00:02:38,000 3 independently suspended and adjacent 67 00:02:38,000 --> 00:02:39,840 to the others, closer to the center of 68 00:02:39,840 --> 00:02:41,598 the atrium. 69 00:02:41,598 --> 00:02:43,840 The Hyatt Regency Hotel regularly hosted 70 00:02:43,840 --> 00:02:46,479 a Friday night tea dance with live music 71 00:02:46,479 --> 00:02:48,479 and dance competitions. 72 00:02:48,479 --> 00:02:51,039 a tea dance is a European tradition, 73 00:02:51,039 --> 00:02:53,598 though also called "Thé Dansant" or 74 00:02:53,598 --> 00:02:56,000 "Dancing Tea" in French, was a dance held 75 00:02:56,000 --> 00:02:58,080 in the summer or early autumn, late 76 00:02:58,080 --> 00:03:00,560 afternoons before sunset. With that 77 00:03:00,560 --> 00:03:02,840 background, let's talk about the incident 78 00:03:02,840 --> 00:03:05,680 itself. On Friday evening, the 17th of 79 00:03:05,680 --> 00:03:08,878 July, 1981 the Hyatt Regency Hotel atrium 80 00:03:08,878 --> 00:03:10,719 was once again hosting a regular tea 81 00:03:10,719 --> 00:03:13,039 dance with live music that evening 82 00:03:13,039 --> 00:03:15,680 played by the Steve Miller Orchestra. The 83 00:03:15,680 --> 00:03:17,280 dance was scheduled to go for 3 84 00:03:17,280 --> 00:03:19,439 hours in total starting at 5pm and 85 00:03:19,439 --> 00:03:21,919 finishing by 8pm playing a variety of 86 00:03:21,919 --> 00:03:23,680 Jazz swing music. 87 00:03:23,680 --> 00:03:25,919 At 3pm local time the first people 88 00:03:25,919 --> 00:03:27,919 started arriving and within 90 minutes 89 00:03:27,919 --> 00:03:29,598 the first floor of the atrium was now 90 00:03:29,598 --> 00:03:31,759 fully occupied, forcing people to move to 91 00:03:31,759 --> 00:03:33,919 the atrium terrace and walkways before 92 00:03:33,919 --> 00:03:36,400 the official start at 5pm. 93 00:03:36,400 --> 00:03:38,719 By 7pm the crowd was estimated at 94 00:03:38,719 --> 00:03:42,080 between 1,500 and 2,000 people in the 95 00:03:42,080 --> 00:03:44,158 entire atrium area. 96 00:03:44,158 --> 00:03:46,959 At 7:04pm the band returned and began 97 00:03:46,959 --> 00:03:49,759 playing as part of the dance competition. 98 00:03:49,759 --> 00:03:52,560 At approximately 7:05pm the 4th 99 00:03:52,560 --> 00:03:54,799 floor walkway, with an estimated 20 100 00:03:54,799 --> 00:03:56,959 people standing on it, buckled in the 101 00:03:56,959 --> 00:03:59,680 center and began to fall onto the 2nd 102 00:03:59,680 --> 00:04:02,318 floor walkway directly beneath it. As the 103 00:04:02,318 --> 00:04:04,400 falling walkway impacted the 2nd 104 00:04:04,400 --> 00:04:06,560 floor walkway with approximately 40 105 00:04:06,560 --> 00:04:09,039 people on it, the second walkway then 106 00:04:09,039 --> 00:04:11,120 also collapsed, with both walkways 107 00:04:11,120 --> 00:04:13,680 collapsing onto the ground beneath, near 108 00:04:13,680 --> 00:04:16,639 a very crowded area adjacent to the bar. 109 00:04:16,639 --> 00:04:18,560 The number of people on the walkways has 110 00:04:18,560 --> 00:04:20,000 been impossible to determine with any 111 00:04:20,000 --> 00:04:22,000 certainty, as a television crew covering 112 00:04:22,000 --> 00:04:23,519 the event were changing batteries in 113 00:04:23,519 --> 00:04:25,040 their equipment when the collapse 114 00:04:25,040 --> 00:04:26,240 occurred. 115 00:04:26,240 --> 00:04:28,240 Water pipes were severed in the collapse 116 00:04:28,240 --> 00:04:30,079 and electrical cables were dislodged 117 00:04:30,079 --> 00:04:32,160 leaving the lobby in near darkness with 118 00:04:32,160 --> 00:04:33,918 frequent sparks from arcing electrical 119 00:04:33,918 --> 00:04:35,600 equipment intermittently lighting the 120 00:04:35,600 --> 00:04:37,759 atrium. It took 3 minutes for the 121 00:04:37,759 --> 00:04:39,759 Kansas City Fire Department to be 122 00:04:39,759 --> 00:04:42,079 contacted in the confusion, followed by 123 00:04:42,079 --> 00:04:44,720 the Police Department shortly thereafter. 124 00:04:44,720 --> 00:04:47,759 By 7:18pm a total of 7 ambulances 125 00:04:47,759 --> 00:04:50,000 had arrived at the hotel. 126 00:04:50,000 --> 00:04:53,120 By 7:52pm an estimated 100 firefighters 127 00:04:53,120 --> 00:04:54,959 and emergency workers were now actively 128 00:04:54,959 --> 00:04:56,639 involved in rescuing people from the 129 00:04:56,639 --> 00:04:58,319 wreckage. Due to the weight of the 130 00:04:58,319 --> 00:05:00,720 walkways, heavy cranes were brought in to 131 00:05:00,720 --> 00:05:03,038 remove the debris, with the first walkway 132 00:05:03,038 --> 00:05:06,959 removed by 3:15am the following morning. 133 00:05:06,959 --> 00:05:08,800 The last survivor was pulled from the 134 00:05:08,800 --> 00:05:12,680 debris at approximately 4:30am. 135 00:05:12,680 --> 00:05:15,600 114 people were killed, 136 00:05:15,600 --> 00:05:19,279 and 216 were injured. 137 00:05:19,279 --> 00:05:22,000 It remains the deadliest, unintentional 138 00:05:22,000 --> 00:05:23,918 structural collapse in the history of 139 00:05:23,918 --> 00:05:25,680 the United States. 140 00:05:25,680 --> 00:05:27,918 Let's talk about the investigations. 141 00:05:27,918 --> 00:05:29,600 Yes...there were multiple. 142 00:05:29,600 --> 00:05:32,240 The mayor of Kansas City, Richard L. 143 00:05:32,240 --> 00:05:34,319 Berkley, formally requested the National 144 00:05:34,319 --> 00:05:36,478 Bureau of Standards to independently 145 00:05:36,478 --> 00:05:38,478 investigate the most probable cause of 146 00:05:38,478 --> 00:05:40,319 the walkway collapse. 147 00:05:40,319 --> 00:05:41,319 Their 148 00:05:41,319 --> 00:05:44,319 378 page report examines in great detail 149 00:05:44,319 --> 00:05:46,399 the physics behind the failure and makes 150 00:05:46,399 --> 00:05:49,519 for interesting reading. In 1983 a grand 151 00:05:49,519 --> 00:05:51,680 jury was convened in Kansas City to 152 00:05:51,680 --> 00:05:53,519 investigate if the collapse was as a 153 00:05:53,519 --> 00:05:55,839 result of any illegal actions of those 154 00:05:55,839 --> 00:05:57,038 involved. 155 00:05:57,038 --> 00:05:59,680 In 1984 the state of Missouri convened 156 00:05:59,680 --> 00:06:01,439 an administrative hearing to determine 157 00:06:01,439 --> 00:06:03,279 whether there had been any violation of 158 00:06:03,279 --> 00:06:05,839 state licensing laws by those involved. 159 00:06:05,839 --> 00:06:09,279 In 1985 the ASCE, that's the American 160 00:06:09,279 --> 00:06:11,759 Society of Civil Engineers also held a 161 00:06:11,759 --> 00:06:13,519 disciplinary hearing. 162 00:06:13,519 --> 00:06:15,360 The relevant findings from all of these 163 00:06:15,360 --> 00:06:17,199 will be discussed in due course, but 164 00:06:17,199 --> 00:06:19,120 before we get to that let's talk a 165 00:06:19,120 --> 00:06:20,879 little bit about the construction of the 166 00:06:20,879 --> 00:06:22,079 hotel. 167 00:06:22,079 --> 00:06:24,800 The hotel was built in the late 1970s 168 00:06:24,800 --> 00:06:26,240 when there was a period of high 169 00:06:26,240 --> 00:06:28,319 inflation, high interest rates, and high 170 00:06:28,319 --> 00:06:30,399 unemployment. The number of available 171 00:06:30,399 --> 00:06:32,399 construction projects at that time was 172 00:06:32,399 --> 00:06:34,000 limited and hence the contracting 173 00:06:34,000 --> 00:06:35,519 companies at that time priced very 174 00:06:35,519 --> 00:06:37,439 aggressively to win work so that they 175 00:06:37,439 --> 00:06:39,519 could stay in business. As projects 176 00:06:39,519 --> 00:06:41,680 increased in availability, those same 177 00:06:41,680 --> 00:06:44,000 companies then pushed to close out their 178 00:06:44,000 --> 00:06:45,759 projects they had on the books so they 179 00:06:45,759 --> 00:06:47,600 could start on the next project as soon 180 00:06:47,600 --> 00:06:48,959 as they could. 181 00:06:48,959 --> 00:06:51,199 This era of construction also saw the 182 00:06:51,199 --> 00:06:53,918 popularisation of the so-called "Fast 183 00:06:53,918 --> 00:06:56,240 Track" method of design and construction. 184 00:06:56,240 --> 00:06:58,720 Projects delivered by Fast Track are 185 00:06:58,720 --> 00:07:00,800 characterized by construction leading 186 00:07:00,800 --> 00:07:03,598 ahead of the final completed design and 187 00:07:03,598 --> 00:07:05,439 in the context of civil engineering 188 00:07:05,439 --> 00:07:08,160 specifically, structural design precedes 189 00:07:08,160 --> 00:07:10,399 architectural's final design. 190 00:07:10,399 --> 00:07:12,720 There's reliance on conceptual sketches, 191 00:07:12,720 --> 00:07:14,800 preliminary drawings and early component 192 00:07:14,800 --> 00:07:16,879 orders with red-lining of shop drawings 193 00:07:16,879 --> 00:07:18,720 and multiple re-issues of design 194 00:07:18,720 --> 00:07:21,519 drawings during the construction phase. 195 00:07:21,519 --> 00:07:22,800 Sounds risky? 196 00:07:22,800 --> 00:07:25,120 Well, if you have good quality assurance 197 00:07:25,120 --> 00:07:27,839 it can still work and on large projects 198 00:07:27,839 --> 00:07:29,839 can reduce the time taken to deliver a 199 00:07:29,839 --> 00:07:31,598 project by 25%, 200 00:07:31,598 --> 00:07:33,360 with the trade-off of increased cost 201 00:07:33,360 --> 00:07:36,478 overall in many cases. Sometimes time is 202 00:07:36,478 --> 00:07:38,560 money too so the sooner it's built, the 203 00:07:38,560 --> 00:07:40,399 sooner it can make income and those 204 00:07:40,399 --> 00:07:42,000 trade-offs are considered to be worth it 205 00:07:42,000 --> 00:07:43,839 in the end. So for this hotel 206 00:07:43,839 --> 00:07:45,839 construction...who are the players 207 00:07:45,839 --> 00:07:48,000 involved? There are a few. Eldridge and 208 00:07:48,000 --> 00:07:50,079 Son Construction Company was the general 209 00:07:50,079 --> 00:07:52,079 contractor on the project. 210 00:07:52,079 --> 00:07:54,160 Eldridge subcontracted the steel work 211 00:07:54,160 --> 00:07:56,720 fabrication to Havens Steel Company as 212 00:07:56,720 --> 00:07:59,839 the fabricator in December, 1978. Gillum 213 00:07:59,839 --> 00:08:01,360 Colico Structural Engineering 214 00:08:01,360 --> 00:08:03,759 Consultants Incorporated (or GCE) was 215 00:08:03,759 --> 00:08:05,598 selected to perform all structural 216 00:08:05,598 --> 00:08:07,519 engineering services for the design and 217 00:08:07,519 --> 00:08:09,360 construction of the hotel in July of 218 00:08:09,360 --> 00:08:11,199 1976. 219 00:08:11,199 --> 00:08:13,199 Jack Gillum was one of the principals 220 00:08:13,199 --> 00:08:16,800 of GCE. GCE subcontracted all structural 221 00:08:16,800 --> 00:08:18,720 engineering services for the project to 222 00:08:18,720 --> 00:08:21,598 Jack D. Gillum & Associates Limited and 223 00:08:21,598 --> 00:08:23,439 designated Jack Gillum as the 224 00:08:23,439 --> 00:08:25,519 professional engineer for the project, 225 00:08:25,519 --> 00:08:28,000 noting that Jack Gillum was also the 226 00:08:28,000 --> 00:08:30,399 president of that firm and hence he was 227 00:08:30,399 --> 00:08:32,639 the Engineer of Record. 228 00:08:32,639 --> 00:08:35,519 Jack Gillum appointed Daniel Duncan from 229 00:08:35,519 --> 00:08:37,918 Jack D. Gillum & Associates as well as the 230 00:08:37,918 --> 00:08:40,399 project manager for daily work execution, 231 00:08:40,399 --> 00:08:42,559 as well as a project engineer and a 232 00:08:42,559 --> 00:08:45,519 senior project designer in late 1976, 233 00:08:45,519 --> 00:08:47,759 along with a full project team. 234 00:08:47,759 --> 00:08:50,639 Patty, Berkebile & Nelson, Herbert Duncan, 235 00:08:50,639 --> 00:08:52,639 and Monroe & Lefebvre 236 00:08:52,639 --> 00:08:54,720 Architects Planners Consortium, 237 00:08:54,720 --> 00:08:58,320 Inc., (or PBNDML for short) served 238 00:08:58,320 --> 00:09:02,000 as associate architects and during 1977 239 00:09:02,000 --> 00:09:04,159 developed the basic design and some of 240 00:09:04,159 --> 00:09:06,320 the detailed designs for different parts 241 00:09:06,320 --> 00:09:10,000 of the hotel. In June of 1978, Jack D. 242 00:09:10,000 --> 00:09:12,559 Gillum & Associates lost both the Hyatt 243 00:09:12,559 --> 00:09:15,039 Regency Hotel project's, project 244 00:09:15,039 --> 00:09:17,600 engineer and its senior project designer 245 00:09:17,600 --> 00:09:19,679 in very quick succession, leaving only 246 00:09:19,679 --> 00:09:21,919 the project manager Daniel Duncan with 247 00:09:21,919 --> 00:09:23,759 any history of the design decisions that 248 00:09:23,759 --> 00:09:25,600 have been made to date. During the 249 00:09:25,600 --> 00:09:28,399 construction phase of the hotel, there 250 00:09:28,399 --> 00:09:31,039 was an incident regarding the atrium. On 251 00:09:31,039 --> 00:09:33,759 Sunday the 14th of October, 1979 during 252 00:09:33,759 --> 00:09:35,919 construction, a section of the East 253 00:09:35,919 --> 00:09:37,679 atrium roof on the North side of the 254 00:09:37,679 --> 00:09:40,000 building above the restaurant area fell 255 00:09:40,000 --> 00:09:42,480 four stories onto the lobby floor. 256 00:09:42,480 --> 00:09:44,639 Fortunately it happened on a Sunday when 257 00:09:44,639 --> 00:09:46,480 no one was working on site and hence 258 00:09:46,480 --> 00:09:48,879 there were no injuries. A spokesman for 259 00:09:48,879 --> 00:09:50,480 the Crown Center Redevelopment 260 00:09:50,480 --> 00:09:52,799 Corporation, the then owner of the hotel 261 00:09:52,799 --> 00:09:54,958 under construction, stated that a: 262 00:09:54,958 --> 00:09:57,200 "...beam fell because of an 263 00:09:57,200 --> 00:09:59,919 installation problem..." 264 00:09:59,919 --> 00:10:01,519 Following the incident it came to light 265 00:10:01,519 --> 00:10:04,078 that the general contractor Eldridge & Son 266 00:10:04,078 --> 00:10:05,559 was fined 267 00:10:05,559 --> 00:10:09,278 $5,920 for 17 violations during the 268 00:10:09,278 --> 00:10:11,759 hotel's construction. Having said that, 269 00:10:11,759 --> 00:10:13,440 none of these issues were related 270 00:10:13,440 --> 00:10:15,679 directly to the collapse, however are 271 00:10:15,679 --> 00:10:18,480 suggestive of a fast-paced construction 272 00:10:18,480 --> 00:10:20,559 process with an inadequate amount of 273 00:10:20,559 --> 00:10:22,000 quality assurance. 274 00:10:22,000 --> 00:10:24,159 There's quite a lengthy analysis of this 275 00:10:24,159 --> 00:10:25,919 specific incident in the show notes by 276 00:10:25,919 --> 00:10:28,639 Gregory Luth: "The Chronology of Hyatt 277 00:10:28,639 --> 00:10:31,039 Regency Collapse" if you're interested. 278 00:10:31,039 --> 00:10:33,440 The key point is that the so-called 279 00:10:33,440 --> 00:10:35,600 installation problem was actually more 280 00:10:35,600 --> 00:10:38,559 of a lack of a cohesive design problem 281 00:10:38,559 --> 00:10:40,240 that during installation was worked 282 00:10:40,240 --> 00:10:42,559 around, rather than questioned and fixed 283 00:10:42,559 --> 00:10:43,679 properly. 284 00:10:43,679 --> 00:10:45,600 Had the designers taken that opportunity 285 00:10:45,600 --> 00:10:47,759 to stop the job and review the design 286 00:10:47,759 --> 00:10:49,519 from end to end they would have more 287 00:10:49,519 --> 00:10:51,519 than likely uncovered the walkway design 288 00:10:51,519 --> 00:10:53,919 issues we're about to go through...but 289 00:10:53,919 --> 00:10:55,440 they didn't. 290 00:10:55,440 --> 00:10:57,919 Let's talk about the walkway design. 291 00:10:57,919 --> 00:11:00,159 The walkway is comprised of four spans 292 00:11:00,159 --> 00:11:02,399 in total with each span denoted by its 293 00:11:02,399 --> 00:11:04,399 interconnecting joints and hanging rods 294 00:11:04,399 --> 00:11:07,278 starting at 7 through to 11. Hence the 295 00:11:07,278 --> 00:11:10,639 first span is designated Span 7-8 then 296 00:11:10,639 --> 00:11:14,958 8-9, 9-10 and 10-11. Before you wonder why 297 00:11:14,958 --> 00:11:17,039 the numbering starts at 7...the numbers 298 00:11:17,039 --> 00:11:18,559 line up with the building's column 299 00:11:18,559 --> 00:11:20,720 numbers, hence the walkways hung between 300 00:11:20,720 --> 00:11:23,600 column 7 and 11 of the overall building 301 00:11:23,600 --> 00:11:24,720 space. 302 00:11:24,720 --> 00:11:26,639 Because the walkways on levels 2 and 4 303 00:11:26,639 --> 00:11:28,639 were joined together by hanging rods, the 304 00:11:28,639 --> 00:11:30,639 4th floor walkway was referred to as 305 00:11:30,639 --> 00:11:33,278 upper and the 2nd floor as lower. 306 00:11:33,278 --> 00:11:35,679 Finally the hanging rods were positioned 307 00:11:35,679 --> 00:11:37,919 on either the East or West side of the 308 00:11:37,919 --> 00:11:39,919 walkway. For example the eastern 309 00:11:39,919 --> 00:11:42,000 centre-most rod connection point on the 310 00:11:42,000 --> 00:11:45,360 4th floor was denoted 9UE. 311 00:11:45,360 --> 00:11:47,360 The box girders connecting the walkway 312 00:11:47,360 --> 00:11:49,200 to the hanger rods were made of 200mm 313 00:11:49,200 --> 00:11:52,078 wide (that's 8") of 314 00:11:52,078 --> 00:11:53,839 c-channel, welded together along their 315 00:11:53,839 --> 00:11:57,440 length hence creating the so-called box. 316 00:11:57,440 --> 00:11:59,200 How heavy was it though? 317 00:11:59,200 --> 00:12:00,879 Because during the investigation they 318 00:12:00,879 --> 00:12:02,799 found significant variability in the 319 00:12:02,799 --> 00:12:04,480 thickness of the concrete decking and 320 00:12:04,480 --> 00:12:06,480 topping materials, so they confirmed the 321 00:12:06,480 --> 00:12:08,559 final weights by weighing several 322 00:12:08,559 --> 00:12:11,120 walkway spans as the design drawings did 323 00:12:11,120 --> 00:12:12,720 not align to what was constructed in 324 00:12:12,720 --> 00:12:14,078 that regard. 325 00:12:14,078 --> 00:12:16,000 Analyzing the fallen walkway sections 326 00:12:16,000 --> 00:12:17,839 led to an estimate of the mass of the 327 00:12:17,839 --> 00:12:20,399 walkway to within a worst case error of 328 00:12:20,399 --> 00:12:23,600 +/- 136kg (or 300 329 00:12:23,600 --> 00:12:24,399 lbs). 330 00:12:24,399 --> 00:12:26,799 The addition of gypsum board to meet 331 00:12:26,799 --> 00:12:28,559 fire endurance requirements added 332 00:12:28,559 --> 00:12:32,000 approximately 1,050kg (that's 333 00:12:32,000 --> 00:12:35,120 2,310lbs) to each walkway span. We'll 334 00:12:35,120 --> 00:12:37,120 get to that shortly. The investigators 335 00:12:37,120 --> 00:12:39,440 therefore concluded that the average 336 00:12:39,440 --> 00:12:42,480 final weight of a single walkway span as 337 00:12:42,480 --> 00:12:44,559 built was 8,050kg 338 00:12:44,559 --> 00:12:46,240 (that's 339 00:12:46,240 --> 00:12:48,159 17,750lbs) which 340 00:12:48,159 --> 00:12:51,600 exceeded the as-designed span weight 341 00:12:51,600 --> 00:12:54,639 by about 8%. 342 00:12:54,639 --> 00:12:56,078 Not a good start. 343 00:12:56,078 --> 00:12:58,078 So what went wrong? We know that the 344 00:12:58,078 --> 00:12:59,919 walkways were 8% heavier than 345 00:12:59,919 --> 00:13:03,440 their design but that's only one issue. 346 00:13:03,440 --> 00:13:05,600 It was determined that the most likely 347 00:13:05,600 --> 00:13:08,078 connection point that failed first was 348 00:13:08,078 --> 00:13:11,839 9UE with transferred load to spans U8-9 349 00:13:11,839 --> 00:13:15,360 and U9-10 on the East connections, then 350 00:13:15,360 --> 00:13:18,480 causing 8UE and 10UE to rapidly fail 351 00:13:18,480 --> 00:13:20,159 with the other side of the walkways 352 00:13:20,159 --> 00:13:24,078 joint 9UW failing as well. The method of 353 00:13:24,078 --> 00:13:26,159 failure was that the nut and washer 354 00:13:26,159 --> 00:13:27,759 pulled through the center of the box 355 00:13:27,759 --> 00:13:29,839 beam under high load. 356 00:13:29,839 --> 00:13:31,679 Given the construction method of welding 357 00:13:31,679 --> 00:13:33,039 x2 c-channels together, the 358 00:13:33,039 --> 00:13:35,120 investigators spent significant time 359 00:13:35,120 --> 00:13:36,720 confirming that the quality of the 360 00:13:36,720 --> 00:13:39,198 welding job on those channels was not a 361 00:13:39,198 --> 00:13:41,600 factor in the incident. 362 00:13:41,600 --> 00:13:43,519 In addition, whilst there were people 363 00:13:43,519 --> 00:13:45,919 present on the walkways it was found in 364 00:13:45,919 --> 00:13:47,600 the investigation that the mass of the 365 00:13:47,600 --> 00:13:49,839 people on the walkways was not a 366 00:13:49,839 --> 00:13:52,078 significant contributing factor to the 367 00:13:52,078 --> 00:13:54,720 failure that they saw in the incident. 368 00:13:54,720 --> 00:13:57,519 The two centre-most spans then rotated 369 00:13:57,519 --> 00:14:00,078 downwards pulling span U10-11 off its 370 00:14:00,078 --> 00:14:01,839 bearing seats at the building column 371 00:14:01,839 --> 00:14:03,360 number 11. 372 00:14:03,360 --> 00:14:05,600 As the upper walkway then fell onto the 373 00:14:05,600 --> 00:14:07,759 lower walkway, it followed a similar 374 00:14:07,759 --> 00:14:10,078 failure progression. The investigation 375 00:14:10,078 --> 00:14:11,759 determined that there were two design 376 00:14:11,759 --> 00:14:13,278 changes that contributed to this 377 00:14:13,278 --> 00:14:17,039 incident, as well as a design gap. 378 00:14:17,039 --> 00:14:19,360 We'll talk about each in turn. The first: 379 00:14:19,360 --> 00:14:23,120 Fire-proofing in March, 1978. A routine 380 00:14:23,120 --> 00:14:24,480 design review by the Codes 381 00:14:24,480 --> 00:14:26,720 Administration Office, Kansas City Public 382 00:14:26,720 --> 00:14:28,958 Works Department resulted in a 6 page 383 00:14:28,958 --> 00:14:30,480 handwritten assessment of the fire 384 00:14:30,480 --> 00:14:32,639 endurance of the walkways structural 385 00:14:32,639 --> 00:14:36,078 steel, dated the 10th of March, 1978. A 386 00:14:36,078 --> 00:14:38,078 meeting was held on the 16th of March, 387 00:14:38,078 --> 00:14:40,320 1978 between representatives of the 388 00:14:40,320 --> 00:14:42,078 Codes Administration Office and the 389 00:14:42,078 --> 00:14:45,039 architects PBNDML with 2 pages of 390 00:14:45,039 --> 00:14:46,720 minutes which included an agreement to 391 00:14:46,720 --> 00:14:48,958 clad the walkway structural steel with 392 00:14:48,958 --> 00:14:51,519 gypsum board to address fire resilience 393 00:14:51,519 --> 00:14:54,799 concerns. Specifically x2 layers of 394 00:14:54,799 --> 00:14:56,639 16mm (that's 5/8") thick 395 00:14:56,639 --> 00:14:59,120 gypsum board supported by 0.8mm 396 00:14:59,120 --> 00:15:01,360 metal studs and nailing strips, then 397 00:15:01,360 --> 00:15:03,600 attached to the structural steel by 398 00:15:03,600 --> 00:15:05,839 power driven fasteners. The structural 399 00:15:05,839 --> 00:15:08,159 drawings were reissued on the 30th of 400 00:15:08,159 --> 00:15:11,198 March, 1978 and it is unclear in the 401 00:15:11,198 --> 00:15:13,600 documentation at the time if any changes 402 00:15:13,600 --> 00:15:16,078 were made to the structural details as a 403 00:15:16,078 --> 00:15:18,480 result of gypsum board being added to 404 00:15:18,480 --> 00:15:21,198 each walkway span. The investigation 405 00:15:21,198 --> 00:15:23,440 concluded that the walkway design did 406 00:15:23,440 --> 00:15:25,120 not get adjusted to account for the 407 00:15:25,120 --> 00:15:27,120 additional load from retrofitting gypsum 408 00:15:27,120 --> 00:15:28,879 board as a fire retardant material 409 00:15:28,879 --> 00:15:31,759 during design. That said, they also 410 00:15:31,759 --> 00:15:34,639 concluded that even if the load from the 411 00:15:34,639 --> 00:15:36,879 gypsum board wasn't present, there would 412 00:15:36,879 --> 00:15:39,440 still have been insufficient redundancy 413 00:15:39,440 --> 00:15:42,078 (also called reserve load capacity) in the 414 00:15:42,078 --> 00:15:44,399 hangar rod design to resist the failure 415 00:15:44,399 --> 00:15:46,000 of a single rod. 416 00:15:46,000 --> 00:15:49,440 Hence, if one breaks, they'll all break. 417 00:15:49,440 --> 00:15:52,078 The second design change, also in March 418 00:15:52,078 --> 00:15:54,879 1978, related to the hangar rod 419 00:15:54,879 --> 00:15:56,240 connection. 420 00:15:56,240 --> 00:15:59,039 In March, 1978 drawings for the walkway 421 00:15:59,039 --> 00:16:01,839 hangar showed a single rod continuously 422 00:16:01,839 --> 00:16:03,759 threaded from top to bottom, that was 423 00:16:03,759 --> 00:16:05,839 suspended from the beams on the atrium 424 00:16:05,839 --> 00:16:07,759 ceiling, passing through the fourth floor 425 00:16:07,759 --> 00:16:10,320 walkway and second floor box beams with 426 00:16:10,320 --> 00:16:12,000 securing nuts and washers at each 427 00:16:12,000 --> 00:16:15,679 walkway level. In early January, 1979 the 428 00:16:15,679 --> 00:16:17,519 drawings were now at the fabricators, 429 00:16:17,519 --> 00:16:18,958 Havens Steel Company, and their 430 00:16:18,958 --> 00:16:20,159 engineering manager called the 431 00:16:20,159 --> 00:16:22,480 structural engineers project manager. 432 00:16:22,480 --> 00:16:24,000 Havens engineering manager raised 433 00:16:24,000 --> 00:16:25,600 concerns about the durability of a 434 00:16:25,600 --> 00:16:27,759 continuously threaded rod over that 435 00:16:27,759 --> 00:16:30,078 length during construction. The original 436 00:16:30,078 --> 00:16:32,879 design called for a nut 6.1m up 437 00:16:32,879 --> 00:16:34,639 the hanger rod and didn't use sleeve 438 00:16:34,639 --> 00:16:37,679 nuts. That, coupled with the higher cost 439 00:16:37,679 --> 00:16:39,679 of manufacture, they proposed to change 440 00:16:39,679 --> 00:16:42,399 the design from a single continuous rod 441 00:16:42,399 --> 00:16:45,519 to two rods instead, with threads on each 442 00:16:45,519 --> 00:16:48,399 end. According to Havens testimony the 443 00:16:48,399 --> 00:16:50,159 structural engineer checked the turning 444 00:16:50,159 --> 00:16:52,078 moment and the sheer force presented at 445 00:16:52,078 --> 00:16:54,879 the box beam for the offset condition 446 00:16:54,879 --> 00:16:57,519 while on the phone and Daniel Duncan 447 00:16:57,519 --> 00:17:00,159 accepted that change over the phone. The 448 00:17:00,159 --> 00:17:01,919 structural engineer then asked the 449 00:17:01,919 --> 00:17:03,919 fabricator to submit the change request 450 00:17:03,919 --> 00:17:06,798 through the "normal channels for approval." 451 00:17:06,798 --> 00:17:08,640 The shop drawings were then marked up to 452 00:17:08,640 --> 00:17:11,038 now have two hanger rods with a 453 00:17:11,038 --> 00:17:13,679 102mm (that's a 4") inset at 454 00:17:13,679 --> 00:17:15,599 the fourth floor walkway connection 455 00:17:15,599 --> 00:17:18,160 point, in place of the original single 456 00:17:18,160 --> 00:17:20,400 continuous hanger rod arrangement. To put 457 00:17:20,400 --> 00:17:21,679 it another way, 458 00:17:21,679 --> 00:17:23,679 the ceiling to upper walkway rods were 459 00:17:23,679 --> 00:17:25,759 on the outside and the upper to lower 460 00:17:25,759 --> 00:17:28,720 walkway rods were offset on the inside 461 00:17:28,720 --> 00:17:29,759 of those. 462 00:17:29,759 --> 00:17:32,880 On the 12th of January, 1979 Havens Steel 463 00:17:32,880 --> 00:17:35,440 Company pulled the Hyatt Regency Project 464 00:17:35,440 --> 00:17:37,440 out of their engineering department and 465 00:17:37,440 --> 00:17:40,000 then subcontracted the for-construction 466 00:17:40,000 --> 00:17:42,160 drafting of the partially completed shop 467 00:17:42,160 --> 00:17:44,558 drawings to an external engineering firm. 468 00:17:44,558 --> 00:17:46,798 This was due to Havens winning a large 469 00:17:46,798 --> 00:17:48,319 project and they required their 470 00:17:48,319 --> 00:17:50,079 engineering resources to be focused on 471 00:17:50,079 --> 00:17:51,599 the new job instead. 472 00:17:51,599 --> 00:17:53,679 The outsourced drawings included the box 473 00:17:53,679 --> 00:17:56,000 beam detail but no rod connection was 474 00:17:56,000 --> 00:17:58,798 shown in its detail, which unfortunately 475 00:17:58,798 --> 00:18:00,798 is common practice in the industry. 476 00:18:00,798 --> 00:18:02,319 Different drawings show different 477 00:18:02,319 --> 00:18:04,640 details based on their final intended 478 00:18:04,640 --> 00:18:06,160 use cases. 479 00:18:06,160 --> 00:18:07,839 To understand the overall design of the 480 00:18:07,839 --> 00:18:09,599 connection configuration to the box 481 00:18:09,599 --> 00:18:11,359 beams, a person would need to consider 482 00:18:11,359 --> 00:18:13,279 all of the drawings and specifications 483 00:18:13,279 --> 00:18:14,480 collectively... 484 00:18:14,480 --> 00:18:17,359 not individually. The offset rod design 485 00:18:17,359 --> 00:18:19,200 change was therefore never submitted for 486 00:18:19,200 --> 00:18:21,200 a formal review in its entirety and the 487 00:18:21,200 --> 00:18:23,038 final connection to the box beams was 488 00:18:23,038 --> 00:18:25,519 only implied on the shop drawings. 489 00:18:25,519 --> 00:18:27,279 They assumed that the connection design 490 00:18:27,279 --> 00:18:30,000 was complete and specified elsewhere, so 491 00:18:30,000 --> 00:18:31,679 they simply added the "weld" symbol at 492 00:18:31,679 --> 00:18:33,519 the connection point, before sending the 493 00:18:33,519 --> 00:18:35,919 drawings out for final approval. 494 00:18:35,919 --> 00:18:38,000 Ordinarily if there are any design 495 00:18:38,000 --> 00:18:40,000 changes required to the shop drawings 496 00:18:40,000 --> 00:18:41,679 they would be submitted under a separate 497 00:18:41,679 --> 00:18:43,519 cover sheet along with a request for 498 00:18:43,519 --> 00:18:45,359 engineering validation of the shop 499 00:18:45,359 --> 00:18:46,640 drawings. 500 00:18:46,640 --> 00:18:48,400 This did not occur. 501 00:18:48,400 --> 00:18:50,720 On the 7th of February the outside 502 00:18:50,720 --> 00:18:52,798 detailer made their final consistency 503 00:18:52,798 --> 00:18:54,480 and completeness checks and the drawings 504 00:18:54,480 --> 00:18:56,480 were sent to the structural engineer for 505 00:18:56,480 --> 00:18:58,160 engineering approval. 506 00:18:58,160 --> 00:19:00,480 With various communication delays the 507 00:19:00,480 --> 00:19:02,000 drawings didn't reach the engineer's 508 00:19:02,000 --> 00:19:04,720 desk until Thursday the 16th of February 509 00:19:04,720 --> 00:19:06,240 and the contractor had requested 510 00:19:06,240 --> 00:19:07,919 expedited approval due to the 511 00:19:07,919 --> 00:19:09,679 construction timetable pressures under 512 00:19:09,679 --> 00:19:12,240 the contract. At that time both the 513 00:19:12,240 --> 00:19:14,319 project engineer and designer had long 514 00:19:14,319 --> 00:19:16,400 since departed and the project manager 515 00:19:16,400 --> 00:19:18,480 Daniel Duncan was too short of time to 516 00:19:18,480 --> 00:19:20,480 review the drawings himself so he 517 00:19:20,480 --> 00:19:22,160 delegated their checking to a senior 518 00:19:22,160 --> 00:19:23,279 technician. 519 00:19:23,279 --> 00:19:24,798 The senior technician was trained in 520 00:19:24,798 --> 00:19:26,079 structural engineering and had two 521 00:19:26,079 --> 00:19:27,679 decades of experience but wasn't 522 00:19:27,679 --> 00:19:29,759 licensed as an engineer. 523 00:19:29,759 --> 00:19:31,519 Whilst he queried about the rod 524 00:19:31,519 --> 00:19:33,759 specification (more on that in a moment) 525 00:19:33,759 --> 00:19:35,679 there were no other significant findings 526 00:19:35,679 --> 00:19:37,200 and the drawings were returned approved 527 00:19:37,200 --> 00:19:39,919 for construction on Sunday the 27th of 528 00:19:39,919 --> 00:19:41,759 February. That's right...they were working 529 00:19:41,759 --> 00:19:43,839 the Sunday to get it done. 530 00:19:43,839 --> 00:19:46,160 This modification essentially doubled 531 00:19:46,160 --> 00:19:48,319 the load transferred from the fourth 532 00:19:48,319 --> 00:19:50,798 floor walkway box beam to the holding 533 00:19:50,798 --> 00:19:53,759 nut for the upper hanging rod from 91kN 534 00:19:53,759 --> 00:19:57,279 to 181kN. 535 00:19:57,279 --> 00:19:59,759 Another way to describe this is the nut 536 00:19:59,759 --> 00:20:01,679 holding the weight of the fourth floor 537 00:20:01,679 --> 00:20:04,000 walkway in the original design only had 538 00:20:04,000 --> 00:20:06,319 the weight of its own floor, in that 539 00:20:06,319 --> 00:20:08,720 segment, applied to it with a continuous 540 00:20:08,720 --> 00:20:11,599 rod. Once a second rod was added, instead 541 00:20:11,599 --> 00:20:14,079 of the load from the lower walkway being 542 00:20:14,079 --> 00:20:16,079 transferred up through the rod to the 543 00:20:16,079 --> 00:20:17,200 ceiling, 544 00:20:17,200 --> 00:20:18,798 since it was hanging from the upper 545 00:20:18,798 --> 00:20:22,400 walkway now (not a common rod) the entire 546 00:20:22,400 --> 00:20:24,640 mass of the lower walkway segment was 547 00:20:24,640 --> 00:20:27,519 applied to the upper walkway as well. 548 00:20:27,519 --> 00:20:29,599 Therefore the mass on the holding nut on 549 00:20:29,599 --> 00:20:31,599 the fourth floor walkway was effectively 550 00:20:31,599 --> 00:20:32,798 doubled. 551 00:20:32,798 --> 00:20:34,558 Having said all of that, 552 00:20:34,558 --> 00:20:36,480 with the correct redundancy factors 553 00:20:36,480 --> 00:20:39,919 applied and under the original design, 554 00:20:39,919 --> 00:20:42,240 accounting for the correct weight of the 555 00:20:42,240 --> 00:20:44,480 walkways as they were built, the minimum 556 00:20:44,480 --> 00:20:48,079 load value needed to be 151kN 557 00:20:48,079 --> 00:20:51,038 even with a single rod design, and this 558 00:20:51,038 --> 00:20:53,839 still didn't meet the AISC that's the 559 00:20:53,839 --> 00:20:56,640 American Institute of Steel Construction 560 00:20:56,640 --> 00:20:59,679 specifications requirements. Specifically 561 00:20:59,679 --> 00:21:02,240 the AISC Specification for "Design, 562 00:21:02,240 --> 00:21:04,079 Fabrication and Erection of Structural 563 00:21:04,079 --> 00:21:06,079 Steel for Buildings" as well as the "Code 564 00:21:06,079 --> 00:21:08,159 of Standard Practice for Steel Buildings 565 00:21:08,159 --> 00:21:10,240 and Bridges" as required by the Kansas 566 00:21:10,240 --> 00:21:12,720 City Building Code. Another key problem 567 00:21:12,720 --> 00:21:14,400 was the design of the walkway support 568 00:21:14,400 --> 00:21:16,640 beam's connection to the hanger rod. The 569 00:21:16,640 --> 00:21:18,159 original design sketch for the rod 570 00:21:18,159 --> 00:21:20,319 connection showed a bracket extending 571 00:21:20,319 --> 00:21:23,440 off of the web of the W8 (that's a Wide 572 00:21:23,440 --> 00:21:26,000 flange I-Beam). W8 (that's the vertical 573 00:21:26,000 --> 00:21:28,798 section of the beam where 8 in W8 is the 574 00:21:28,798 --> 00:21:31,759 height of that beam in inches) 575 00:21:31,759 --> 00:21:33,519 and that had an indicative spacing for 576 00:21:33,519 --> 00:21:35,679 the rod connection with the rod sizing 577 00:21:35,679 --> 00:21:37,359 and forces all drawn on the engineer's 578 00:21:37,359 --> 00:21:40,480 sketch. In March, 1978 the project manager 579 00:21:40,480 --> 00:21:43,440 replaced the W8 beam at the hangar 580 00:21:43,440 --> 00:21:45,440 location with a pair of 8" 581 00:21:45,440 --> 00:21:47,119 channels we previously spoke about, 582 00:21:47,119 --> 00:21:49,599 turned toe to toe, and extended these 583 00:21:49,599 --> 00:21:52,720 past the interconnecting W16 beam. This 584 00:21:52,720 --> 00:21:54,798 was done to eliminate concerns about the 585 00:21:54,798 --> 00:21:57,359 eccentricity of the original suggested 586 00:21:57,359 --> 00:21:59,440 bracket connection with the I-Beam. The 587 00:21:59,440 --> 00:22:02,480 revised design was in effect symmetrical 588 00:22:02,480 --> 00:22:04,880 where the original design wasn't. 589 00:22:04,880 --> 00:22:07,200 The revised engineers sketch also showed 590 00:22:07,200 --> 00:22:09,919 the rod size force and rod grade. When 591 00:22:09,919 --> 00:22:12,159 the draftsman transcribed the engineer's 592 00:22:12,159 --> 00:22:15,119 sketch to a for-construction drawing, 593 00:22:15,119 --> 00:22:17,759 they omitted the note specifying the 594 00:22:17,759 --> 00:22:21,679 strength of a 413MPa (or 60kips) 595 00:22:21,679 --> 00:22:23,919 for the hanger rods. In civil engineering 596 00:22:23,919 --> 00:22:26,640 in that era, writing the load detail on 597 00:22:26,640 --> 00:22:28,079 the sketch 598 00:22:28,079 --> 00:22:29,919 indicated to the fabricator that the 599 00:22:29,919 --> 00:22:31,919 connection design still needed to be 600 00:22:31,919 --> 00:22:33,919 completed, and this was normally part of 601 00:22:33,919 --> 00:22:36,079 the fabricator's scope of work. The 602 00:22:36,079 --> 00:22:38,319 engineer had assumed that the fabricator 603 00:22:38,319 --> 00:22:40,319 would complete the design detail for the 604 00:22:40,319 --> 00:22:42,880 connection to the box beam hanger. 605 00:22:42,880 --> 00:22:45,200 It is not clear why the additional 606 00:22:45,200 --> 00:22:47,279 details were not transcribed by the 607 00:22:47,279 --> 00:22:49,599 draftsman, however the fact that they 608 00:22:49,599 --> 00:22:52,400 weren't, led to a mis-assumption by the 609 00:22:52,400 --> 00:22:54,400 fabricator about whether the connection 610 00:22:54,400 --> 00:22:56,640 design was actually complete. They 611 00:22:56,640 --> 00:22:58,000 believed it was, 612 00:22:58,000 --> 00:23:00,079 when it wasn't, and equally the engineer 613 00:23:00,079 --> 00:23:01,279 believed it was being handled by the 614 00:23:01,279 --> 00:23:04,159 fabricator when it wasn't. Placing the 615 00:23:04,159 --> 00:23:06,798 rod through the weak point of the welded 616 00:23:06,798 --> 00:23:09,279 c-channels may have looked the most 617 00:23:09,279 --> 00:23:12,000 pleasant to the eye however the nature 618 00:23:12,000 --> 00:23:14,640 of c-channels is that they taper from 619 00:23:14,640 --> 00:23:16,960 each corner to each tip furthest from 620 00:23:16,960 --> 00:23:19,279 the back of the "C" such that the tips 621 00:23:19,279 --> 00:23:22,240 are the narrowest parts of the channel. 622 00:23:22,240 --> 00:23:24,079 Placing those tip to tip and then 623 00:23:24,079 --> 00:23:26,000 running a weld seam along them to 624 00:23:26,000 --> 00:23:28,159 join them together and then grinding 625 00:23:28,159 --> 00:23:29,839 that weld flat 626 00:23:29,839 --> 00:23:33,279 for assembly, resulted in a very weak 627 00:23:33,279 --> 00:23:35,200 connection. The welders on site 628 00:23:35,200 --> 00:23:37,679 recognised this and attempted to weld 629 00:23:37,679 --> 00:23:40,000 the inner join line inside the box 630 00:23:40,000 --> 00:23:42,319 section, but they could only reach as far 631 00:23:42,319 --> 00:23:44,000 in as the end of their welding rod 632 00:23:44,000 --> 00:23:45,519 allowed them to. 633 00:23:45,519 --> 00:23:47,679 Hardly precision work and pretty much 634 00:23:47,679 --> 00:23:49,839 unconstructable in that way...laws of 635 00:23:49,839 --> 00:23:51,759 physics and all that. 636 00:23:51,759 --> 00:23:54,880 A final note though about the rod design 637 00:23:54,880 --> 00:23:58,640 change that also occurred in March 1978. 638 00:23:58,640 --> 00:24:01,119 The architect requested the rods be 639 00:24:01,119 --> 00:24:04,159 changed from 44.5mm (that's 640 00:24:04,159 --> 00:24:06,558 1-3/4") to 32mm 641 00:24:06,558 --> 00:24:09,119 (or 1-1/4") to "lighten up" 642 00:24:09,119 --> 00:24:11,278 the appearance of the bridges. The 643 00:24:11,278 --> 00:24:13,359 original drawing showing the rod detail 644 00:24:13,359 --> 00:24:15,519 did not specify the material on the 645 00:24:15,519 --> 00:24:18,079 drawing though the other specifications 646 00:24:18,079 --> 00:24:21,119 suggested standard A36 grade rods (that's 647 00:24:21,119 --> 00:24:24,880 36ksi tensile strength) should be used 648 00:24:24,880 --> 00:24:27,440 based on the original design. To achieve 649 00:24:27,440 --> 00:24:29,038 the load-bearing requirements for the 650 00:24:29,038 --> 00:24:32,240 narrower rod, a grade 60 rod (therefore 651 00:24:32,240 --> 00:24:34,880 with a 60ksi tensile strength) would be 652 00:24:34,880 --> 00:24:36,798 required and this was marked up on the 653 00:24:36,798 --> 00:24:39,278 revised engineer's sketch. Just a quick 654 00:24:39,278 --> 00:24:42,079 note...I talk about minimum yield strength... 655 00:24:42,079 --> 00:24:44,079 that's the amount of stress a material 656 00:24:44,079 --> 00:24:46,159 can withstand before it succumbs to a 657 00:24:46,159 --> 00:24:48,000 permanent deformation and doesn't return 658 00:24:48,000 --> 00:24:50,079 back to where it started. In imperial 659 00:24:50,079 --> 00:24:52,319 units it's usually expressed in ksi 660 00:24:52,319 --> 00:24:54,880 which is kilo pounds per square inch of 661 00:24:54,880 --> 00:24:57,038 pressure. Metric units is normally in 662 00:24:57,038 --> 00:25:00,480 MPa or Mega Pascals of pressure. If you 663 00:25:00,480 --> 00:25:02,880 recall, the final drafting for the for- 664 00:25:02,880 --> 00:25:04,798 construction drawing set was outsourced 665 00:25:04,798 --> 00:25:06,480 by the fabricator to another engineering 666 00:25:06,480 --> 00:25:08,240 firm that had a long-standing business 667 00:25:08,240 --> 00:25:10,159 relationship with the fabricator. 668 00:25:10,159 --> 00:25:11,278 During the course of checking the 669 00:25:11,278 --> 00:25:12,720 drawings the technician from the 670 00:25:12,720 --> 00:25:14,558 outsourced back-drafting company 671 00:25:14,558 --> 00:25:16,640 involved asked the project manager about 672 00:25:16,640 --> 00:25:18,480 the strength of the hanger rods. The 673 00:25:18,480 --> 00:25:20,880 technician's own calculations found that 674 00:25:20,880 --> 00:25:23,919 36ksi steel rod would not fully support 675 00:25:23,919 --> 00:25:25,759 the load indicated on the other 676 00:25:25,759 --> 00:25:28,159 available drawings. The project manager 677 00:25:28,159 --> 00:25:30,400 responded from memory that it was a high 678 00:25:30,400 --> 00:25:32,798 strength rod however no mention of the 679 00:25:32,798 --> 00:25:35,440 specific rod grade was provided. 680 00:25:35,440 --> 00:25:37,440 Further than that, no attempt was made to 681 00:25:37,440 --> 00:25:39,599 verify this by looking at a complete set 682 00:25:39,599 --> 00:25:41,359 of drawings and specifications and the 683 00:25:41,359 --> 00:25:42,798 drawings were issued without this 684 00:25:42,798 --> 00:25:44,558 information detailed on the for- 685 00:25:44,558 --> 00:25:47,119 construction drawing set. Some analyses 686 00:25:47,119 --> 00:25:48,960 of this incident call this out as a 687 00:25:48,960 --> 00:25:51,278 cause of the incident but that's not 688 00:25:51,278 --> 00:25:53,599 really true as the rods themselves 689 00:25:53,599 --> 00:25:55,599 didn't actually fail. 690 00:25:55,599 --> 00:25:58,558 Having said that, I have no doubt at all 691 00:25:58,558 --> 00:26:00,960 that had the box beam joints not failed 692 00:26:00,960 --> 00:26:03,440 first with the passage of enough time 693 00:26:03,440 --> 00:26:05,919 and cumulative deformation of the rods 694 00:26:05,919 --> 00:26:08,079 eventually would have led to a collapse 695 00:26:08,079 --> 00:26:09,278 in the future. 696 00:26:09,278 --> 00:26:10,880 When exactly that could have happened 697 00:26:10,880 --> 00:26:13,440 who can say? Technically...not a cause of 698 00:26:13,440 --> 00:26:16,000 this incident but clearly an error that 699 00:26:16,000 --> 00:26:18,000 could have led to a different incident 700 00:26:18,000 --> 00:26:19,679 had the other issues not beaten them to 701 00:26:19,679 --> 00:26:21,278 it first. 702 00:26:21,278 --> 00:26:22,960 Let's talk a little bit about the legal 703 00:26:22,960 --> 00:26:26,159 fallout. In 1984, Missouri's Board of 704 00:26:26,159 --> 00:26:27,839 Architects, Professional Engineers and 705 00:26:27,839 --> 00:26:29,599 Land Surveyors commenced disciplinary 706 00:26:29,599 --> 00:26:31,759 proceedings against Duncan, Gillum and 707 00:26:31,759 --> 00:26:33,200 GCE. 708 00:26:33,200 --> 00:26:35,679 After a 27-day hearing and weeks of 709 00:26:35,679 --> 00:26:37,679 compiling their report the commission 710 00:26:37,679 --> 00:26:39,359 eventually issued its findings which 711 00:26:39,359 --> 00:26:42,640 were 442 pages long and found all three 712 00:26:42,640 --> 00:26:45,038 parties grossly negligent and revoked 713 00:26:45,038 --> 00:26:46,640 their licences. 714 00:26:46,640 --> 00:26:49,519 Judge James B. Deutsch, an administrative law 715 00:26:49,519 --> 00:26:51,038 judge for Missouri's Administrative 716 00:26:51,038 --> 00:26:52,640 Hearing Commission found the structural 717 00:26:52,640 --> 00:26:54,480 engineers guilty of gross negligence, 718 00:26:54,480 --> 00:26:57,359 misconduct and unprofessional conduct. 719 00:26:57,359 --> 00:26:59,038 They claimed that the design flaws had 720 00:26:59,038 --> 00:27:01,119 resulted from a mis-communication between 721 00:27:01,119 --> 00:27:03,200 Jack D. Gillum & Associates and the 722 00:27:03,200 --> 00:27:06,079 Havens Steel Company. Mr Duncan, Mr Gillum 723 00:27:06,079 --> 00:27:08,240 and GCE unsuccessfully appealed their 724 00:27:08,240 --> 00:27:10,000 decision up to the Missouri Court of 725 00:27:10,000 --> 00:27:11,038 Appeals. 726 00:27:11,038 --> 00:27:13,119 The Court of Appeals also rejected the 727 00:27:13,119 --> 00:27:15,038 more substantive attacks on the 728 00:27:15,038 --> 00:27:16,798 sufficiency of the evidence in the 729 00:27:16,798 --> 00:27:18,240 initial proceedings, stating the 730 00:27:18,240 --> 00:27:21,839 following: 1) "...Mr Duncan was 731 00:27:21,839 --> 00:27:23,759 responsible for designing and approving 732 00:27:23,759 --> 00:27:25,839 the building structure..." of which the 733 00:27:25,839 --> 00:27:28,398 walkways fell within that scope; "...[T]he 734 00:27:28,398 --> 00:27:30,720 walkways offered a potential of great 735 00:27:30,720 --> 00:27:33,038 danger to human life if defectively 736 00:27:33,038 --> 00:27:34,558 designed..." 737 00:27:34,558 --> 00:27:35,919 2) 738 00:27:35,919 --> 00:27:38,000 "...Mr Duncan approved the fabricator's 739 00:27:38,000 --> 00:27:40,079 change, recommending it to the architect, 740 00:27:40,079 --> 00:27:41,919 and approved shop drawings reflecting it 741 00:27:41,919 --> 00:27:44,159 without confirming its acceptability; 742 00:27:44,159 --> 00:27:46,398 (noting that) the change effectively 743 00:27:46,398 --> 00:27:49,119 doubled the box connection load..." and 744 00:27:49,119 --> 00:27:51,759 3) "...Mr Duncan never reviewed the shop 745 00:27:51,759 --> 00:27:54,319 drawings even though such a review is an 746 00:27:54,319 --> 00:27:57,119 'Engineering Function' that even GCE's 747 00:27:57,119 --> 00:28:00,159 in-house policies required that he 748 00:28:00,159 --> 00:28:01,440 do..." 749 00:28:01,440 --> 00:28:03,839 The Court of Appeals upheld the original 750 00:28:03,839 --> 00:28:06,720 conclusion that: "...[t]he conduct 751 00:28:06,720 --> 00:28:08,640 of Duncan from initial design through 752 00:28:08,640 --> 00:28:10,558 shop drawing review and through the 753 00:28:10,558 --> 00:28:12,880 subsequent requested connection review... 754 00:28:12,880 --> 00:28:14,720 supports the Commission's finding of 755 00:28:14,720 --> 00:28:16,200 conscious indifference to 756 00:28:16,200 --> 00:28:19,038 [non-delegatable] professional duty." 757 00:28:19,038 --> 00:28:21,679 The Court of Appeal also upheld 758 00:28:21,679 --> 00:28:23,599 the original conclusion that Mr Gillum 759 00:28:23,599 --> 00:28:25,519 had displayed gross negligence because 760 00:28:25,519 --> 00:28:28,398 he failed as the engineer of record: 761 00:28:28,398 --> 00:28:29,440 762 00:28:29,440 --> 00:28:31,519 "...to assure that the Hyatt engineering 763 00:28:31,519 --> 00:28:33,440 designs and drawings were structurally 764 00:28:33,440 --> 00:28:35,919 sound...prior to impressing their upon his 765 00:28:35,919 --> 00:28:38,319 seal..." and failed "...to assure adequate 766 00:28:38,319 --> 00:28:40,240 shop drawing review." 767 00:28:40,240 --> 00:28:41,440 768 00:28:41,440 --> 00:28:44,000 During the trial the detailer, architect, 769 00:28:44,000 --> 00:28:46,319 fabricator, and technician, all testified 770 00:28:46,319 --> 00:28:47,679 that during construction they had 771 00:28:47,679 --> 00:28:49,679 contacted the project engineer regarding 772 00:28:49,679 --> 00:28:51,599 the structural integrity of the walkway 773 00:28:51,599 --> 00:28:54,000 connection detail. They claimed that each 774 00:28:54,000 --> 00:28:56,480 time it was raised the project engineer 775 00:28:56,480 --> 00:28:58,558 assured them that the connection design 776 00:28:58,558 --> 00:29:00,558 was sound and whilst he claimed to have 777 00:29:00,558 --> 00:29:02,720 checked the detail, in reality there was 778 00:29:02,720 --> 00:29:04,558 no evidence found that he had ever 779 00:29:04,558 --> 00:29:07,440 performed any calculations at all. The 780 00:29:07,440 --> 00:29:09,839 legal repercussions for the Hyatt 781 00:29:09,839 --> 00:29:11,919 engineers firmly established the 782 00:29:11,919 --> 00:29:14,000 Engineer of Records responsibilities for 783 00:29:14,000 --> 00:29:16,000 the structural integrity of an entire 784 00:29:16,000 --> 00:29:17,919 building during construction...including 785 00:29:17,919 --> 00:29:20,159 the shop drawings. Hence in the civil 786 00:29:20,159 --> 00:29:22,079 engineering context, the Engineer of 787 00:29:22,079 --> 00:29:24,240 Record should design and detail all 788 00:29:24,240 --> 00:29:27,119 non-standard structural connections. More 789 00:29:27,119 --> 00:29:29,519 broadly all new designs need to be 790 00:29:29,519 --> 00:29:31,519 thoroughly checked and all modifications 791 00:29:31,519 --> 00:29:33,679 to design details require a formal 792 00:29:33,679 --> 00:29:35,519 written approval from the Engineer of 793 00:29:35,519 --> 00:29:37,119 Record. 794 00:29:37,119 --> 00:29:39,359 The day before that decision was handed 795 00:29:39,359 --> 00:29:41,119 down by Judge Deutsch, 796 00:29:41,119 --> 00:29:43,359 the American Society of Civil Engineers 797 00:29:43,359 --> 00:29:45,759 had announced a new policy of holding 798 00:29:45,759 --> 00:29:48,079 structural engineers responsible for all 799 00:29:48,079 --> 00:29:50,319 elements of structural safety in the 800 00:29:50,319 --> 00:29:52,398 buildings they design. Let's talk about 801 00:29:52,398 --> 00:29:55,359 the aftermath. The hotel reopened 802 00:29:55,359 --> 00:29:57,919 3 months after the incident, after 803 00:29:57,919 --> 00:30:00,319 $5M USD of reconstruction. 804 00:30:00,319 --> 00:30:02,558 A single walkway was reinstalled, only 805 00:30:02,558 --> 00:30:04,319 this time supported from beneath by 806 00:30:04,319 --> 00:30:05,759 concrete pillars. 807 00:30:05,759 --> 00:30:07,440 Other than the 3rd floor now being 808 00:30:07,440 --> 00:30:09,200 left without a connecting walkway the 809 00:30:09,200 --> 00:30:11,200 lobby generally retained its original 810 00:30:11,200 --> 00:30:12,240 design. 811 00:30:12,240 --> 00:30:14,640 The hotel was renamed the Hyatt Regency 812 00:30:14,640 --> 00:30:17,359 Crown Center in 1987, but when Starwood 813 00:30:17,359 --> 00:30:20,000 took over the hotel in November of 2011, 814 00:30:20,000 --> 00:30:22,319 it was renamed the Sheraton Kansas City 815 00:30:22,319 --> 00:30:24,480 Hotel at Crown Center. 816 00:30:24,480 --> 00:30:26,558 Mr Gillum and Mr Duncan had their 817 00:30:26,558 --> 00:30:28,960 engineering licences revoked, as did the 818 00:30:28,960 --> 00:30:31,200 company GCE, and the two had their 819 00:30:31,200 --> 00:30:34,720 memberships to the ASCE revoked as well. 820 00:30:34,720 --> 00:30:37,200 GCE survived only by merging with a 821 00:30:37,200 --> 00:30:39,839 Denver engineering firm: Ketcham, Konkel, 822 00:30:39,839 --> 00:30:41,440 Nichol, Barrett. 823 00:30:41,440 --> 00:30:43,679 No criminal charges were filed and no 824 00:30:43,679 --> 00:30:45,599 one was prosecuted. 825 00:30:45,599 --> 00:30:47,278 Compensation claims from the courts 826 00:30:47,278 --> 00:30:50,079 awarded victims around $140M USD 827 00:30:50,079 --> 00:30:52,000 paid by the Crown Center 828 00:30:52,000 --> 00:30:54,240 Redevelopment Corporation, the owner of 829 00:30:54,240 --> 00:30:56,398 the hotel, with multiple significant 830 00:30:56,398 --> 00:30:59,759 insurance payouts on top of that. 831 00:30:59,759 --> 00:31:02,319 In 2015, over three decades later, a 832 00:31:02,319 --> 00:31:04,720 memorial was finally built adjacent to 833 00:31:04,720 --> 00:31:06,558 the hotel and was paid for by the 834 00:31:06,558 --> 00:31:09,440 Skywalk Memorial Foundation. It had taken 835 00:31:09,440 --> 00:31:12,960 the foundation a decade to raise the $550,000 836 00:31:12,960 --> 00:31:15,440 USD to pay for the memorial 837 00:31:15,440 --> 00:31:18,398 and associated maintenance endowment. The 838 00:31:18,398 --> 00:31:20,240 names of each of the victims are etched 839 00:31:20,240 --> 00:31:22,558 in the memorial of which the centerpiece 840 00:31:22,558 --> 00:31:25,119 is a six meter or twenty feet high 841 00:31:25,119 --> 00:31:28,159 abstract sculpture of a couple embraced 842 00:31:28,159 --> 00:31:29,440 as they dance. 843 00:31:29,440 --> 00:31:32,000 So what do we learn from all this? 844 00:31:32,000 --> 00:31:33,599 There are three points I'd like to go 845 00:31:33,599 --> 00:31:35,359 over here. Responsibility of the 846 00:31:35,359 --> 00:31:37,759 engineers during design; Complete design 847 00:31:37,759 --> 00:31:39,599 reviews rather than piecemeal; and 848 00:31:39,599 --> 00:31:41,359 personnel churn. 849 00:31:41,359 --> 00:31:42,960 The Engineer of Record is a commonly 850 00:31:42,960 --> 00:31:45,200 used term in North America which has in 851 00:31:45,200 --> 00:31:46,960 the past and particularly prior to this 852 00:31:46,960 --> 00:31:49,038 incident, had varying interpretations as 853 00:31:49,038 --> 00:31:51,038 the ultimate responsibility of that job 854 00:31:51,038 --> 00:31:53,119 role in a construction project. 855 00:31:53,119 --> 00:31:54,960 The Hyatt Regency walkway collapse has 856 00:31:54,960 --> 00:31:56,880 become the legal precedent against which 857 00:31:56,880 --> 00:31:59,119 the role of EoR has ever since been 858 00:31:59,119 --> 00:32:01,919 defined in the USA. The buck stops with 859 00:32:01,919 --> 00:32:04,159 the Engineer of Record. They are 860 00:32:04,159 --> 00:32:05,440 responsible. 861 00:32:05,440 --> 00:32:07,679 In Australia we have legal requirements 862 00:32:07,679 --> 00:32:09,599 under the Professional Engineers Act and 863 00:32:09,599 --> 00:32:11,759 for many engineering disciplines today, 864 00:32:11,759 --> 00:32:13,839 if you aren't a Registered Professional 865 00:32:13,839 --> 00:32:15,839 Engineer in the state you're operating 866 00:32:15,839 --> 00:32:17,759 in and/or a Chartered Professional 867 00:32:17,759 --> 00:32:19,440 Engineer in Australia then you aren't 868 00:32:19,440 --> 00:32:22,000 allowed to sign off on a design drawing. 869 00:32:22,000 --> 00:32:24,398 I've been an RPEQ in Electrical 870 00:32:24,398 --> 00:32:27,679 Engineering for 17 years and also an RPEQ 871 00:32:27,679 --> 00:32:29,200 in ITEE and a Chartered 872 00:32:29,200 --> 00:32:30,880 Professional Engineer in both as well 873 00:32:30,880 --> 00:32:32,558 and I take those responsibilities 874 00:32:32,558 --> 00:32:35,440 extremely seriously. When I stamp and 875 00:32:35,440 --> 00:32:37,599 sign a drawing it means I've checked 876 00:32:37,599 --> 00:32:39,839 every detail of that drawing to the best 877 00:32:39,839 --> 00:32:41,759 of my ability which is in effect what 878 00:32:41,759 --> 00:32:43,759 the Engineer of Record is required to do 879 00:32:43,759 --> 00:32:46,798 as well. When I'm asked to RPEQ a 880 00:32:46,798 --> 00:32:48,960 drawing, there's also additional 881 00:32:48,960 --> 00:32:50,880 requirements. I need to have been 882 00:32:50,880 --> 00:32:52,880 involved with the design from start to 883 00:32:52,880 --> 00:32:55,519 finish...an aspect that was lacking in the 884 00:32:55,519 --> 00:32:58,000 case of this specific incident. I've been 885 00:32:58,000 --> 00:32:59,679 involved in projects where we've spent 886 00:32:59,679 --> 00:33:01,200 significant amounts of money 887 00:33:01,200 --> 00:33:03,919 subcontracting to find the right RPEQ 888 00:33:03,919 --> 00:33:06,159 from other organizations in order to get 889 00:33:06,159 --> 00:33:08,480 the design correctly validated after 890 00:33:08,480 --> 00:33:10,640 that specific engineer had left the 891 00:33:10,640 --> 00:33:13,038 design organization mid-project. We 892 00:33:13,038 --> 00:33:15,278 tracked them down to their new job 893 00:33:15,278 --> 00:33:17,759 and we got them back. Design reviews for 894 00:33:17,759 --> 00:33:19,759 complex systems aren't as simple as 895 00:33:19,759 --> 00:33:21,839 you'd like sometimes, especially when we 896 00:33:21,839 --> 00:33:24,240 fractionalise our designs. I like to call 897 00:33:24,240 --> 00:33:26,319 it "Fractional Engineering." 898 00:33:26,319 --> 00:33:28,319 The idea that reviewing the design or 899 00:33:28,319 --> 00:33:31,038 change as a whole just takes too long or 900 00:33:31,038 --> 00:33:32,880 has too many hurdles when it's 901 00:33:32,880 --> 00:33:35,519 considered in its entirety, so we break 902 00:33:35,519 --> 00:33:37,119 it into fractions of the whole and 903 00:33:37,119 --> 00:33:38,398 review those, 904 00:33:38,398 --> 00:33:40,960 one fraction or one slice at a time...in 905 00:33:40,960 --> 00:33:43,679 isolation. There's often too much detail 906 00:33:43,679 --> 00:33:45,599 to put on a single drawing or in a 907 00:33:45,599 --> 00:33:47,599 single specification which means the 908 00:33:47,599 --> 00:33:50,880 engineer needs to often read, ingest and 909 00:33:50,880 --> 00:33:53,278 comprehend a significant number of 910 00:33:53,278 --> 00:33:55,440 drawings and information in order to 911 00:33:55,440 --> 00:33:58,159 determine if the design in its entirety 912 00:33:58,159 --> 00:34:00,398 is safe and correct. 913 00:34:00,398 --> 00:34:03,440 That takes more time and as they say, "It 914 00:34:03,440 --> 00:34:05,519 takes as long as it takes, if you have to 915 00:34:05,519 --> 00:34:06,960 do it right." 916 00:34:06,960 --> 00:34:08,719 The time pressures and distractions from 917 00:34:08,719 --> 00:34:09,918 other elements of the building's 918 00:34:09,918 --> 00:34:11,679 construction clearly affected the 919 00:34:11,679 --> 00:34:14,079 outcome in this incident. "It's just a 920 00:34:14,079 --> 00:34:16,960 walkway...I have bigger concerns..." I have no 921 00:34:16,960 --> 00:34:20,239 doubt that kind of mindset was a factor. 922 00:34:20,239 --> 00:34:21,519 Finally though... 923 00:34:21,519 --> 00:34:23,440 personnel churn. 924 00:34:23,440 --> 00:34:26,239 It's insidious and it comes from a 925 00:34:26,239 --> 00:34:29,039 belief that people can be plug and play. 926 00:34:29,039 --> 00:34:30,960 "We're all just numbers." I've worked for 927 00:34:30,960 --> 00:34:33,280 companies that push hard on documenting 928 00:34:33,280 --> 00:34:35,280 everything...turning everything into a 929 00:34:35,280 --> 00:34:37,358 process, with evidence provided for every 930 00:34:37,358 --> 00:34:39,280 single step and then when someone leaves 931 00:34:39,280 --> 00:34:41,519 the organization the management layer 932 00:34:41,519 --> 00:34:44,079 hires someone new and think everything 933 00:34:44,079 --> 00:34:46,480 can just pick up from where it was...no 934 00:34:46,480 --> 00:34:49,119 risk, no problem. Thinking is...any new 935 00:34:49,119 --> 00:34:51,280 person can just read all the documents, 936 00:34:51,280 --> 00:34:54,398 comprehend the design and finish the job. 937 00:34:54,398 --> 00:34:56,878 If only it worked that way. 938 00:34:56,878 --> 00:34:59,358 It is simply not possible to write down 939 00:34:59,358 --> 00:35:02,239 every last detail, every conversation, 940 00:35:02,239 --> 00:35:03,760 every side note, every corridor 941 00:35:03,760 --> 00:35:05,280 conversation, site meeting, inspection 942 00:35:05,280 --> 00:35:07,679 note, and even if it was...could the next 943 00:35:07,679 --> 00:35:09,760 person that wasn't present early in the 944 00:35:09,760 --> 00:35:11,679 project then absorb all of that 945 00:35:11,679 --> 00:35:13,679 information and comprehend its 946 00:35:13,679 --> 00:35:16,639 importance and context in any reasonable 947 00:35:16,639 --> 00:35:18,079 amount of time? 948 00:35:18,079 --> 00:35:20,079 Enough to finish the job without an 949 00:35:20,079 --> 00:35:21,039 incident? 950 00:35:21,039 --> 00:35:23,679 It's complete fiction. 951 00:35:23,679 --> 00:35:26,559 We are not plug and play. We are the sum 952 00:35:26,559 --> 00:35:28,880 of our experiences and we are unique. So 953 00:35:28,880 --> 00:35:30,960 few things in engineering are truly copy 954 00:35:30,960 --> 00:35:33,119 and paste widgets and even if they are 955 00:35:33,119 --> 00:35:35,119 the larger the scale, the differences 956 00:35:35,119 --> 00:35:37,679 location and materials makes can vastly 957 00:35:37,679 --> 00:35:40,079 impact the outcome. It's so important to 958 00:35:40,079 --> 00:35:41,679 keep the core team consistent in 959 00:35:41,679 --> 00:35:44,000 projects and if that means paying more, 960 00:35:44,000 --> 00:35:45,599 giving people what they want to get them 961 00:35:45,599 --> 00:35:48,400 to stay, that's not pandering, that's a 962 00:35:48,400 --> 00:35:50,880 long-term vision about what's the best 963 00:35:50,880 --> 00:35:53,119 for the project and the company overall 964 00:35:53,119 --> 00:35:54,559 and that's what it means to be a good 965 00:35:54,559 --> 00:35:57,119 manager. Not trying the sell of plug and 966 00:35:57,119 --> 00:35:59,519 play, "We'll roll on, it'll be fine..." that's 967 00:35:59,519 --> 00:36:00,880 not management. 968 00:36:00,880 --> 00:36:02,880 Building and maintaining a good, stable, 969 00:36:02,880 --> 00:36:04,639 competent team that communicates well 970 00:36:04,639 --> 00:36:06,880 together makes an enormous difference. 971 00:36:06,880 --> 00:36:08,639 Creating time and space to execute 972 00:36:08,639 --> 00:36:10,880 thorough design reviews does too. 973 00:36:10,880 --> 00:36:13,358 Somewhat of an unusual footnote however. 974 00:36:13,358 --> 00:36:14,800 When something traumatic like this 975 00:36:14,800 --> 00:36:17,199 occurs to a person, they can handle it in 976 00:36:17,199 --> 00:36:19,119 a multitude of ways. 977 00:36:19,119 --> 00:36:21,599 We saw the tragedy of the Challenger 978 00:36:21,599 --> 00:36:23,599 Space Shuttle engineers with Bob Ebling 979 00:36:23,599 --> 00:36:25,440 retiring from engineering, unable to 980 00:36:25,440 --> 00:36:27,679 practice, consumed by guilt, although 981 00:36:27,679 --> 00:36:28,800 admittedly he didn't cause the 982 00:36:28,800 --> 00:36:30,559 Challenger explosion but he did fail to 983 00:36:30,559 --> 00:36:32,639 stop the launch, but in this case 984 00:36:32,639 --> 00:36:35,440 something altogether different happened. 985 00:36:35,440 --> 00:36:37,679 Jack Gillium was born on the 21st of 986 00:36:37,679 --> 00:36:40,239 November, 1928 and graduated from the 987 00:36:40,239 --> 00:36:41,920 University of Kansas with a degree in 988 00:36:41,920 --> 00:36:43,838 Architectural Engineering before being 989 00:36:43,838 --> 00:36:45,760 drafted into the army to serve in the 990 00:36:45,760 --> 00:36:49,519 Korean war between 1950 and 1952. 991 00:36:49,519 --> 00:36:51,599 In the early 1960s he found his own 992 00:36:51,599 --> 00:36:53,599 structural engineering company, Jack D. 993 00:36:53,599 --> 00:36:55,519 Gillum & Associates and was known for 994 00:36:55,519 --> 00:36:57,280 many prominent engineering designs 995 00:36:57,280 --> 00:37:00,639 around the world until Hyatt Regency. 996 00:37:00,639 --> 00:37:01,920 Whilst he initially fought the 997 00:37:01,920 --> 00:37:04,159 allegations against him he ultimately 998 00:37:04,159 --> 00:37:06,079 came to a different conclusion about his 999 00:37:06,079 --> 00:37:08,159 involvement, and accepted that he was in 1000 00:37:08,159 --> 00:37:10,719 fact, partly to blame for the incident. As 1001 00:37:10,719 --> 00:37:12,639 he lost his engineering license he 1002 00:37:12,639 --> 00:37:14,639 instead decided to spend the rest of his 1003 00:37:14,639 --> 00:37:16,800 working career advising, teaching and 1004 00:37:16,800 --> 00:37:19,440 publicly speaking about how he and his 1005 00:37:19,440 --> 00:37:21,519 company had got it wrong and how we can 1006 00:37:21,519 --> 00:37:24,159 all learn from it. He wrote multiple 1007 00:37:24,159 --> 00:37:26,079 papers including the excellent: "The 1008 00:37:26,079 --> 00:37:28,000 Engineer of Record and Design 1009 00:37:28,000 --> 00:37:29,920 Responsibility" paper that's linked in 1010 00:37:29,920 --> 00:37:31,199 the show notes. 1011 00:37:31,199 --> 00:37:34,239 He passed away on the 4th of July, 2012 1012 00:37:34,239 --> 00:37:37,519 aged 83 years old. I've gone a bit back 1013 00:37:37,519 --> 00:37:39,679 and forth on how I feel about him but in 1014 00:37:39,679 --> 00:37:41,838 the end Jack Gillum could have gone into 1015 00:37:41,838 --> 00:37:44,559 seclusion, wracked by guilt, but he chose 1016 00:37:44,559 --> 00:37:47,119 to stay out in front and essentially do 1017 00:37:47,119 --> 00:37:48,960 exactly what I'm doing here with 1018 00:37:48,960 --> 00:37:51,679 Causality. He tried to educate people 1019 00:37:51,679 --> 00:37:54,320 about how it can go wrong as it had for 1020 00:37:54,320 --> 00:37:56,880 him personally in his case, to try and 1021 00:37:56,880 --> 00:37:59,920 prevent future incidents from occurring. 1022 00:37:59,920 --> 00:38:02,559 And that, on balance, 1023 00:38:02,559 --> 00:38:04,880 I think that was the right call to make, 1024 00:38:04,880 --> 00:38:07,280 and I'm glad he chose that in the end 1025 00:38:07,280 --> 00:38:09,599 over alternatives. 1026 00:38:09,599 --> 00:38:11,599 At the risk of being a touch bit 1027 00:38:11,599 --> 00:38:14,639 flippant, the old and mostly untraceable 1028 00:38:14,639 --> 00:38:17,358 saying is: "When you assume, you make an 1029 00:38:17,358 --> 00:38:20,000 ASS out of U and ME." 1030 00:38:20,000 --> 00:38:22,239 The "Fast Track" method of construction 1031 00:38:22,239 --> 00:38:24,400 employed here led engineers and 1032 00:38:24,400 --> 00:38:26,719 fabricators leaning on assumptions out 1033 00:38:26,719 --> 00:38:28,800 of a lack of available time with 1034 00:38:28,800 --> 00:38:31,199 horrible results. For those engineers 1035 00:38:31,199 --> 00:38:33,119 listening to this now, whether you're 1036 00:38:33,119 --> 00:38:35,440 reviewing a design, even if you're not 1037 00:38:35,440 --> 00:38:38,960 the EoR, or your CPEng or RPEng aren't on 1038 00:38:38,960 --> 00:38:41,358 the line, take that review with the same 1039 00:38:41,358 --> 00:38:43,760 level of vigor as if they were. Gather 1040 00:38:43,760 --> 00:38:45,760 the information you need to be sure if 1041 00:38:45,760 --> 00:38:48,559 the design isn't clear and push back... 1042 00:38:48,559 --> 00:38:52,559 if you need more time...you NEED MORE TIME! 1043 00:38:52,559 --> 00:38:54,079 If you need to speak with the original 1044 00:38:54,079 --> 00:38:56,960 designer...find them! Ask them. 1045 00:38:56,960 --> 00:38:58,800 Get clarity. 1046 00:38:58,800 --> 00:39:00,480 Get it right. 1047 00:39:00,480 --> 00:39:02,880 Being an engineer can be a stressful job 1048 00:39:02,880 --> 00:39:05,679 but can also be a rewarding one too. I 1049 00:39:05,679 --> 00:39:07,838 regularly think back over my 25 year 1050 00:39:07,838 --> 00:39:10,239 career to date and I wonder if someday 1051 00:39:10,239 --> 00:39:12,480 an error I made in the past, might turn 1052 00:39:12,480 --> 00:39:14,400 out with negative consequences - ones that 1053 00:39:14,400 --> 00:39:17,119 I didn't foresee at the time in the past. 1054 00:39:17,119 --> 00:39:19,599 I try not to dwell on it but that's not 1055 00:39:19,599 --> 00:39:21,199 really the point. 1056 00:39:21,199 --> 00:39:24,159 I use those thoughts, those fears, to keep 1057 00:39:24,159 --> 00:39:27,280 me sharp. A bit of fear isn't a bad thing. 1058 00:39:27,280 --> 00:39:30,159 It's a reminder to take what I do, what 1059 00:39:30,159 --> 00:39:32,559 many of us do for a living, very, very 1060 00:39:32,559 --> 00:39:33,760 seriously. 1061 00:39:33,760 --> 00:39:35,280 Had the structural engineers and 1062 00:39:35,280 --> 00:39:36,880 fabricators done the same on this 1063 00:39:36,880 --> 00:39:37,920 project, 1064 00:39:37,920 --> 00:39:42,320 those 114 people might not have died. 1065 00:39:42,320 --> 00:39:44,320 Make the time. 1066 00:39:44,320 --> 00:39:46,320 Do better. 1067 00:39:46,320 --> 00:39:48,000 If you're enjoying Causality and you'd 1068 00:39:48,000 --> 00:39:49,838 like to support us and keep the show ad- 1069 00:39:49,838 --> 00:39:52,079 free you can by becoming a Premium 1070 00:39:52,079 --> 00:39:54,000 Supporter. Premium support is available 1071 00:39:54,000 --> 00:39:56,079 via Patreon, through the Apple Podcasts 1072 00:39:56,079 --> 00:39:58,400 channel subscription, and through Spotify. 1073 00:39:58,400 --> 00:40:00,960 Just visit https://engineered.network/causality to 1074 00:40:00,960 --> 00:40:02,719 learn how you can help this show to 1075 00:40:02,719 --> 00:40:04,880 continue to be made. Thank you. 1076 00:40:04,880 --> 00:40:06,880 A big thank you to all of our supporters, 1077 00:40:06,880 --> 00:40:08,400 a special thank you to our Silver 1078 00:40:08,400 --> 00:40:10,639 Producers: Mitch Biegler, Kevin Koch, 1079 00:40:10,639 --> 00:40:14,079 Lesley, Shane O'Neill, Hafthor, Jared, Bill, 1080 00:40:14,079 --> 00:40:16,960 Joel Maher, Katharina Will and Dave Jones, 1081 00:40:16,960 --> 00:40:18,960 and an extra special thank you to both 1082 00:40:18,960 --> 00:40:21,440 of our Gold Producers: Steven Bridle and 1083 00:40:21,440 --> 00:40:24,000 our Gold Producer known only as 'R'. 1084 00:40:24,000 --> 00:40:26,079 Causality is heavily researched and 1085 00:40:26,079 --> 00:40:27,679 links to all materials used for the 1086 00:40:27,679 --> 00:40:29,440 creation of this episode are contained 1087 00:40:29,440 --> 00:40:30,960 in the show notes. You can find them in 1088 00:40:30,960 --> 00:40:32,400 the text of the episode description of 1089 00:40:32,400 --> 00:40:35,119 your podcast player or, on our website. 1090 00:40:35,119 --> 00:40:38,000 Causality is a Podcasting 2.0 enhanced 1091 00:40:38,000 --> 00:40:39,760 show and with the right podcast player 1092 00:40:39,760 --> 00:40:41,838 you'll have episode locations, enhanced 1093 00:40:41,838 --> 00:40:43,760 chapters, and real-time subtitles on 1094 00:40:43,760 --> 00:40:46,000 selected episodes and you can also 1095 00:40:46,000 --> 00:40:47,440 stream satoshi's and Boost with a 1096 00:40:47,440 --> 00:40:49,358 message if you like...there's details on 1097 00:40:49,358 --> 00:40:50,719 how along with the Boostagram 1098 00:40:50,719 --> 00:40:53,039 leaderboard on our website. Causality now 1099 00:40:53,039 --> 00:40:55,119 has support for listener submitted sound 1100 00:40:55,119 --> 00:40:57,760 bites from any episode! You can create 1101 00:40:57,760 --> 00:40:59,920 and Email in for inclusion on the 1102 00:40:59,920 --> 00:41:02,639 website and in Podcasting 2.0 compliant 1103 00:41:02,639 --> 00:41:04,800 apps. If you'd like to make a credited 1104 00:41:04,800 --> 00:41:07,119 submission visit https://engineered.network 1105 00:41:07,119 --> 00:41:10,000 /createsoundbite/ 1106 00:41:10,000 --> 00:41:12,000 to learn more...there's also a link 1107 00:41:12,000 --> 00:41:13,760 in the notes. You can follow me on the 1108 00:41:13,760 --> 00:41:16,480 Fediverse @chidgey@engineered.space on 1109 00:41:16,480 --> 00:41:18,719 Twitter @johnchidgey or the 1110 00:41:18,719 --> 00:41:21,358 network @Engineered_Net. 1111 00:41:21,358 --> 00:41:24,480 This was Causality. I'm John Chidgey. 1112 00:41:24,480 --> 00:41:28,280 Thanks so much for listening.