Statistical Analysis of Bus OnTime Data:Route 370 case study

[BroadGauge]

Much of the data seems a bit dubious. I find it hard to believe that so many routes are apparently as tardy as that site claims them to be. For example, looking at some outer suburban services that personally I wouldn't have thought would cop many delays:

Bus route 677
Richmond to Penrith via Londonderry

Sydney Buses Network
Last two weeks:

2018-01-15 - 23.1% of trips were on-time
2018-01-16 - 15.4% of trips were on-time
2018-01-17 - 30.8% of trips were on-time

Bus route 32

Warragamba to Camden via Werombi & Theresa Park

Sydney Buses Network
Last two weeks:

2018-01-15 - 14.3% of trips were on-time
2018-01-16 - 28.6% of trips were on-time
2018-01-17 - 35.7% of trips were on-time

Bus route 596

Hornsby to Hornsby Heights (Loop Service)

Sydney Buses Network
Last two weeks:

2018-01-14 - 63.6% of trips were on-time
2018-01-15 - 30.6% of trips were on-time
2018-01-16 - 30.6% of trips were on-time
2018-01-17 - 26.5% of trips were on-time

I was actually in Hornsby Heights earlier this month, and all of the buses that I saw in the PM peak were pretty much on time to the minute!

[neilrex]

Yes, I thought that was unconvincing, too.

Those timetables would already have so much padding in them.

[neilrex]

And frequency is the key part.
It doesn't matter if every 55 or 370 is running 20min late - to the pax it doesn't matter.

It's when one in 20min late, the next 10min late and the final one on time so clumping in 3s, that the passenger is upset

Sure, if the bus is 15 minutes late, then you might be catching the one which was supposed to come 15 minutes ago.

It's when there is no bus for 45 minutes, that you have a problem. This seems to be a common occurence on the 400.

Part of the problem is, a lot of the routes are so long. I am in favour of the long cross-town routes, but you need to realise that this is one of the issues that makes it worse.

The alternative would be simpler routes, needing more changes. Opal more or less has fixed the fare penalty for that. But the bus arrangements in downtown Sydney are militantly unfavourable for bus interchanges, compared to most other cities I have visited.

[rpy]

This seems to be a measurement of the quality of arrival forecasting in PTIPS more than a reflection on bus on time running - the number of loop services measured running almost exactly an hour late is a huge red flag.

[andy_centralcoast]

Are you saying the real time data isn't very accurate for buses?

It seems pretty good for trains.

[simonl]

It's the old 'why do buses always come in 3?' problem. The reasons are fairly well documented and short of slamming doors in peoples faces, skipping stops or building in lots of dwell time, there is little that can be done to avoid without significantly increasing costs.

Sure there are. All door boarding, only stopping every 800m rather than every 300m, all pre paid, and priority measures will all reduce travel time variability which will make it more reliable. Then there's the aspect largely within the STA's power which is actually starting the trip on time - not practised by STA as their own data attests.

[Frosty]

I noticed a 370 this afternoon left Marketplace 2 mins late by the time it got Coogee it was 30 mins late. Looking at it it lost quite a few mins along Glebe Pt Rd. Then 9 mins from King St to Sydney Park Rd. 6 mins from Sydney Park Rd onto Euston Rd probably due to roundabout works.

[GazzaOak]

Well, what do u expect from the 370 route where 90% of the route has traffic