News & Discussion: Roads & Traffic

[Nort]

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

[abc]

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

'apparently extensive modelling'

[Nort]

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

'apparently extensive modelling'

[abc]

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

'apparently extensive modelling'

Screenshot_20240520-212612.png

I call BS

[Nort]

Fine.

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

On the other other hand there are those calling BS because I dunno conspiracy or something.

[abc]

Fine.

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

On the other other hand there are those calling BS because I dunno conspiracy or something.

I'd like to see how they can model random human driver behaviour. It would be about as accurate as their 'computer models' for climate.
In this case its just buzzwords. Your last sentence makes you sound like a lot of online drones.

[SBD]

Fine.

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

On the other other hand there are those calling BS because I dunno conspiracy or something.

I'd like to see how they can model random human driver behaviour. It would be about as accurate as their 'computer models' for climate.
In this case its just buzzwords. Your last sentence makes you sound like a lot of online drones.

There are two quite different ways this modelling could be done. I have no idea if either or both (or something else) has been done.

• Make an agent that "acts like a human" (including occasional random misbehaviour) and run the road with lots of replicants of that agent.
• Treat traffic as a fluid and model it that way

In either case, you run the model lots of times until the averages converge. Presumably in this case, you model the current road and test the model against history to get the random number generators tuned. Then run the same agents/model in the proposed new road configuration and see if there's a statistically-significant difference in the outcome.

If this has been done, one or two junior staff in DIT will have written a report on it, and that could probably be accessed through an FOI request if anybody cared enough.

[abc]

Fine.

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

On the other other hand there are those calling BS because I dunno conspiracy or something.

I'd like to see how they can model random human driver behaviour. It would be about as accurate as their 'computer models' for climate.
In this case its just buzzwords. Your last sentence makes you sound like a lot of online drones.

There are two quite different ways this modelling could be done. I have no idea if either or both (or something else) has been done.

• Make an agent that "acts like a human" (including occasional random misbehaviour) and run the road with lots of replicants of that agent.
• Treat traffic as a fluid and model it that way

In either case, you run the model lots of times until the averages converge. Presumably in this case, you model the current road and test the model against history to get the random number generators tuned. Then run the same agents/model in the proposed new road configuration and see if there's a statistically-significant difference in the outcome.

If this has been done, one or two junior staff in DIT will have written a report on it, and that could probably be accessed through an FOI request if anybody cared enough.

lol okay

[mattblack]

Fine.

On the one hand there is apparently extensive modelling and evidence from crashes to support the change.

On the other hand there are those complaining that a drive through this section will take an extra 30 seconds under ideal conditions with this change.

On the other other hand there are those calling BS because I dunno conspiracy or something.

I'd like to see how they can model random human driver behaviour. It would be about as accurate as their 'computer models' for climate.
In this case its just buzzwords. Your last sentence makes you sound like a lot of online drones.

There are two quite different ways this modelling could be done. I have no idea if either or both (or something else) has been done.

• Make an agent that "acts like a human" (including occasional random misbehaviour) and run the road with lots of replicants of that agent.
• Treat traffic as a fluid and model it that way

In either case, you run the model lots of times until the averages converge. Presumably in this case, you model the current road and test the model against history to get the random number generators tuned. Then run the same agents/model in the proposed new road configuration and see if there's a statistically-significant difference in the outcome.

If this has been done, one or two junior staff in DIT will have written a report on it, and that could probably be accessed through an FOI request if anybody cared enough.

I couldn't agree with you more abc. Can't trust those computers. They always spit out answers you don't want to hear or believe. I use a typewriter to do all my modelling.

BTW your still on ignore list so don't bother.

[Nort]

Somewhat relevant to this discussion, it's much more primitive modelling than what is used professionally, but would be very interesting to take a game like this, model real Adelaide intersections and roads and see how well the game models match reality: https://80.lv/articles/use-real-world-m ... ngineering

[Saltwater]

Somewhat relevant to this discussion, it's much more primitive modelling than what is used professionally, but would be very interesting to take a game like this, model real Adelaide intersections and roads and see how well the game models match reality: https://80.lv/articles/use-real-world-m ... ngineering

Thanks for the tip, this game looks like fun. It's been added to the wishlist.

[rev]

Independent MLC Frank Pangallo calls for trial of hook turns in Adelaide CBD and suburbs
Adelaide would adopt a controversial road rule from across the border if a renewed push from an SA MP gets up. Have your say.

Natalie Vikhrov
less than 2 min read
September 2, 2024 - 3:15PM
The Messenger

Could a Melbourne-style hook turn be the answer to Adelaide CBD’s ‘no right turn’ woes? South Australian independent upper house MP Frank Pangallo is advocating for a trial of the manoeuvre in the CBD and some busy suburban intersections.

The road rule is largely used in Melbourne and requires drivers to turn right from the left hand lane.

But Mr Pangallo believes it could help alleviate congestion as well as improve safety and traffic flow at peak times in Adelaide too.

“There are more than 300 no-right turns in the city and greater Adelaide, causing delays and frustrations to drivers,” he said.

“Since the Marshall government banned right turns at some intersections along King William Street to make things easier for the trams, it has created a time-wasting nightmare just navigating your way to get to either the west or the east.

“And if you miss one of the few opportunities to make a right turn, it will add to the journey in costs and productivity if you can’t make your scheduled appointments.

“Hook turns can look terrifying the first time, but they have proven to be very effective in keeping traffic flowing smoothly in Melbourne, which has around 49 of them.”



Mr Pangallo said hook turns were safer than making right turns at busy intersections with no dedicated right-turn lanes, where sometimes only one or two vehicles can pass through traffic light signal cycles.

Adelaide currently has one hook turn, restricted to buses, turning from King William St eastbound on to North Tce.

Mr Pangallo wants to see the rule expanded to motorists, flagging the intersections of King William and Grenfell streets (for westbound traffic to Grenfell) and King William and Currie streets (for eastbound traffic to Currie) as good options to trial them.

Transport Minister Tom Koutsantonis said he was happy to consider the proposal.

“I’m always happy to speak with Frank Pangallo about his proposals and give them due consideration,” he said.

But not everyone is on board with the idea.

RAA’s Charles Mountain said hook turns were “not a particularly desirable treatment”.

“I think it would take a lot of driver education to be comfortable with it,” he told 7NEWS.

https://www.adelaidenow.com.au/subscrib ... nt-1-SCORE

[ChillyPhilly]

I don't agree, I think hook turns will actually increase congestion by making driving in the CBD even more convenient than it already is.

[Norman]

So many errors with the story, does this paper even do basic fact checking?

Firstly, the right turns along King William Street were banned in 2007 under the Rann government, not the Marshall government.

Secondly, the is a second hook turn on Currie Street turning onto King William Street.

Personally, I think it's not a necessity for Adelaide. Melbourne's streets are narrower and trams run in more streets, making them essential for getting around the city. In Adelaide we have plenty of alternatives that can be used to turn right.

[abc]

So many errors with the story, does this paper even do basic fact checking?

Firstly, the right turns along King William Street were banned in 2007 under the Rann government, not the Marshall government.

Secondly, the is a second hook turn on Currie Street turning onto King William Street.

Personally, I think it's not a necessity for Adelaide. Melbourne's streets are narrower and trams run in more streets, making them essential for getting around the city. In Adelaide we have plenty of alternatives that can be used to turn right.

no you're wrong

I actually live in the CBD and right turns were indeed banned during the time I've been here as when I first arrived in 2014 I could turn right from Sturt into KW, now I cannot.