International Railroad Discussion > Rio Tinto arrives at automated rail completion

https://www.afr.com/business/mining/iron-ore/rio-tinto-arrives-at-automated-rail-completion-20181227-h19i4b

AutoHaul technology has now been installed on all of Rio Tinto's 200 Pilbara locomotives. Trains now controlled from Perth, 970 miles to the south

Article is behind a pay wall.

Sorry, apparently not in Australia. Below is the guts of the article without photos

Rio Tinto has declared mission nearly accomplished on the ambitious $US940 million ($1.3 billion) plan to automate its Pilbara long-haul iron ore rail network with confirmation that its "pioneering" technology has now been successfully and fully commissioned.The leading-edge AutoHaul technology has been installed on all of its 200 locomotives and its 50-strong fleet of trains and they have been successfully linked to remotely operated facilities at all 16 of Rio Tinto's Pilbara mines and at its four export port terminals.The scale and unanticipated complexity of the train automation project is revealed by the fact that it has arrived at this point of near-completion nearly four years behind schedule and 81 per cent over budget.The original budget for the automation project was $517 million and the full retro-fitting of Rio Tinto's rail system with its in-house AutoHaul technology was supposed to be completed early in 2015.The delayed passing of this latest milestone leaves Rio Tinto within reach of fully justifying a marketing pitch that describes the miner's Pilbara rail system as "the world's largest robot and first automated heavy-haul distance rail network".In a statement Rio Tinto said that the ramp-up of the automated network would continue through 2019, a process that meant that all of its 600 train drivers would retain their positions for at least another year.Of those 600 drivers, some 240 are employed to steer the 2.5km long trains up and down the four mainlines of what is said to be the world's busiest rail network. At some point in the future, those drivers will be redeployed to other tasks or lose their jobs.At this stage some 90 per cent of the available train fleet is deemed capable of running without a driver. Of the 34 trains that Rio Tinto deployed across the system on Boxing Day, 28 ran without drivers, two ran on autopilot but with drivers on-board and the other four required human direction.

Peak performance to rise
The longer-term target is to incrementally increase the AutoHaul journeys to about 95 per cent of all network haulage and to have the robot's running solo.The biggest initial benefit of automation is that the 32,300-tonne trains will not have to stop and start to allow driver shift changes. The other first-generation incremental benefit is that the range of human driver performance is broad while the robots' offer the equivalent of peak-human performance more consistently.Over time though, as the trains learn more about themselves and how to optimise their performance though collection of data in the management of forces, power requirements, fuel loads and speeds, that peak performance will rise across the system.The final safe and successful commissioning of AutoHaul almost a decade after its inception stands a tribute to "the pioneering spirit inside Rio Tinto", according to the man driving the process, Rio Tinto Iron Ore managing director Rail, Port and Core Services, Ivan Vella."It's been a challenging journey to automate a rail network of this size and scale in a remote location like the Pilbara, but early results indicate significant potential to improve productivity, providing increased system flexibility and reducing bottlenecks," Mr Vella said in a statement.Rio Tinto's investment in automation of both its rail and mine haul fleets has left it a leader in what some have dubbed the mining industry's new arms race.Both ventures, which will have the truck fleet robotised within two years and has the rail system already fully autonomous, sit critical to the $US2.6 billion investment that Rio Tinto confirmed in its first fully digital mine, Koodairderi, just a month ago.

>some 240 are employed to steer the 2.5km long trains up and down the four mainlines of what is said to be the world's busiest rail network. 

Hmmm . . . Is this "factoid" coming from a CNN reporter?  

So future Rio Tinto locomotives will be cabless?

That's a lot of money to just get rid of employees.  Now if eveyone would do the same then nobody would have any money to buy their products.!

1 or 2 or 3 Billion.  Fraction of  a worthless border wall.  It will be interesting how this might be adopted to other rail lines.
RB

tomstp Wrote:
-------------------------------------------------------
> That's a lot of money to just get rid of
> employees.  Now if eveyone would do the same then
> nobody would have any money to buy their
> products.!

Not long-term it isn't. Machines/systems are better at some things than people, people do some things better than automated systems, so that's what people should do.

Around 200 train drivers, up to 600 haul truck drivers

These guys are FIFO (https://en.wikipedia.org/wiki/Fly-in_fly-out), so 2 incumbants for each position, Rio pays for airfares and messing (employees flown in from Perth)

With oncosts each head costing at least $US300k per annum

So Pilbera cost saving of at least $US480 mil per annum, without costing in the fewer camp support staff required

Perth control room staff possibly costing around $US80k per annum, massive savings

Payback period well under 4 years, good investment

 

 Rio has just advised that there will be no job losses. There are many additional costs and delays to their operation due to Auto Haul so don't believe the media hype or BS.

Please tell the family of the law enforcement officer killed this week by an illegal alien that a boarder wall would be "worthless." Just one life saved would justify the cost. 

andersonb109 Wrote:
-------------------------------------------------------
> Please tell the family of the law enforcement
> officer killed this week by an illegal alien that
> a boarder wall would be "worthless." Just one life
> saved would justify the cost. 

Why don't you dream of a lap dance with Ann Coulter,  you'd feel much better.

CA_Sou_MA_Agent Wrote:
-------------------------------------------------------
> >some 240 are employed to steer the 2.5km long trains up and down the four mainlines of what is said to be the world's busiest rail network. 
>
> Hmmm . . . Is this "factoid" coming from a CNN reporter?  

I laughed pretty hard at the "steer" part...
Wow.  With statements like that, how can ANY of the rest of it be taken seriously?

HH

andersonb109 Wrote:
-------------------------------------------------------
> Please tell the family of the law enforcement  officer killed this week by an illegal alien that a boarder wall would be "worthless." Just one life saved would justify the cost. 

I agree with you here Mr. Anderson.  But I fear we're vastly outnumbered.  Note that Trainorders is headquartered in liberal California and I think the vast majority of its subscribers are libs.  California, the state where the governor just granted clemency to 95 convicted murderers.  I'm embarrassed to say that I live in CA.  

We are not all in California, far from it.  

Bruce:
Please leave your political comments off this site. There are plenty of other sites where such comments would be appreciated, lets keep TO out of such comments. If I want to hear pro Trump or pro Dem propiganda I too can find other venues to satisffy my personal interests. 

Megafuggley Wrote:
-------------------------------------------------------
>  Rio has just advised that there will be no job
> losses. There are many additional costs and delays
> to their operation due to Auto Haul so don't
> believe the media hype or BS.

I agree with that. Phrases like "nearly complete" has the integrity of financial media (a.k.a. none).

It's a guess, but the idea of "artificial intelligence" has massively cost time and money for management who bought into the hype. If Rio Tinto has learned anything, more sensors, less intelligence is the way to go. Machines are still inferior to humans in decision making, especially when it comes to instinct and anticipation.

All the arguments about long term cost savings are just hot air to appease "shareholders" to cover up the delays and shortfalls.  While I'm not dismissing the progress being made on automation, those that deal with the physical world (transportation) are still being tested, nowhere near "nearly complete."

up833 Wrote:
-------------------------------------------------------
> 1 or 2 or 3 Billion.  Fraction of  a worthless
> border wall.  It will be interesting how this
> might be adopted to other rail lines.
> RB

Realize that the other political posts here were replies to your off topic remark, to place proper "blame." Maybe you committed the sin of speed reading and mistakenly replied to PHall's comment about "pay wall" which has nothing to do with politics?

There's a handful of TO members who realize the importance of automation as related to workforce reduction. Behind the manipulative narrative by either "side" is a serious economic issue. Since it's off topic, I won't elaborate further but it makes "cost savings" peanuts compared to the larger problems looming underneath. The information is out there to research, but mainstream media by either "side" is NOT a good start.

In this thread and others, Rio Tinto's decision to automate this operation has been discussed. Remote operation with partial automation instead of complete automation is probably the solution they're looking for, based on jmt's comments above about the need to shuttle crews from one location to another. I think Rio Tinto management got tricked into the effectivess of "self-driving" and "AI" and they've been regrouping for a more practical solution, IMHO.

well, how much profit was rio tinto making with engineer's driving trains?  how much more profit will rio tinto make with "driverless" (automated) trains?

just curious.......

66

goneon66 Wrote:
-------------------------------------------------------
> well, how much profit was rio tinto making with
> engineer's driving trains?  how much more profit
> will rio tinto make with "driverless" (automated)
> trains?
>
> just curious.......
>
> 66

Not always just about more profits... Increasing efficiency and cost reductions can allow a company to maintain profit levels while avoiding/minimising price increases. That can make a company more competitive.