3 ms·
So here's the setup: >Watch for these articles to downplay or ignore our actual 2017 safety data and to instead focus on a small number of complaints and anecd
by 6stringmerc 9y ago
So here's the setup:
>Watch for these articles to downplay or ignore our actual 2017 safety data and to instead focus on a small number of complaints and anecdotes that are not representative of what is actually occurring in our factory of over 10,000 workers.
...and a citation to drive the point of Safety home:
>In addition, through the end of Q1 2017, the factory’s total recordable incident rate (TRIR), the leading metric for workplace safety, is 4.6, which is 32% better than the industry average of 6.7.
...and then we get to the whole "Can Elon Musk really scale Tesla to the heights he likes to claim are possible" notion of context:
>Ford had 50,703 U.S. hourly workers as of Feb. 1, according to a report from the UAW's Ford department. A GM spokesman said the company has about 50,300 U.S. hourly workers.[1]
So a 32% better safety record with 1/5 the workforce. Hm. Got it.
[1] http://www.autonews.com/article/20150215/OEM/302169970/ford-tops-gm-in-u.s.-factory-jobs http://www.autonews.com/article/20150215/OEM/302169970/ford-...
- scott00 9y agoTRIR = (Number of OSHA recordable incidents)*200,000 / (total number of hours worked) It's already normalized for workforce size. Source: https://www.csmgroup.com/trir-a-key-safety-performance-indicator-you-should-know/ https://www.csmgroup.com/trir-a-key-safety-performance-indic...
- mhermher 9y agoMaybe he's implying that it doesn't scale linearly.