Study Casts Doubt On Extended Clearway Benefits

A seven-month travel time study has found no noticeable improvement in tram travel times from extended clearway hours on Sydney Road in Brunswick.

For some time, PTUA Secretary Tony Morton has been travelling to and from work with a stopwatch. His aim is to measure—carefully and scientifically—what it is that’s making Melbourne trams so slow. His stopwatch counts up the ‘dead time’ on tram journeys: the lost time when the tram isn’t actually picking up or dropping off passengers, yet is not moving.

His results have proved surprising. While trams do spend time in traffic queues in places where trams and cars share a lane, that’s not the biggest source of delay. More often, a stationary tram is just waiting for a red traffic light to turn green.

“A couple of years ago I did a study on the Lygon Street tram,” Dr Morton said. “What I found is that even if you don’t count boarding time, trams are delayed twice as long in the central city as they are in the suburbs. This is surprising at first, because in the CBD, trams and cars occupy separate lanes. But it starts to make sense once you realise the biggest problem is traffic lights, not car congestion.”

Dr Morton’s conclusion—that trams spend up to one-third of their travel time just waiting for red lights—was presented at a Melbourne transport conference in 2007.

Now, Dr Morton has turned his attention to a newly controversial question: whether clearways improve tram travel speeds in ‘peak shoulder’ times.

“In July this year, the clearway finish time in Sydney Road was extended from 6pm to 7pm,” he said. “As it happens, this is exactly the time when I’m travelling home on the number 19 tram. So I’ve been able to time the trams before and after the clearway time was extended, to see what sort of difference it would really make.”

His results? “In June before the clearway was extended, northbound trams took an average of 9 minutes and 12 seconds to travel the length of Sydney Road Brunswick. In the 4 months or so since July, they have taken an average of 8 minutes and 58 seconds. The 14 second difference is not statistically significant, and may well be down to pure chance,” Dr Morton explained.

Dr Morton also collected data for morning peak travel to the city, and for travel in Royal Parade, immediately south of Sydney Road. This evidence confirmed his earlier finding: that trams are delayed more by red lights close to the city than by traffic queues further out. “In Sydney Road the trams averaged 16.5kph in the evening, but in Royal Parade they only averaged 15.1kph. Yet as we all know, Royal Parade has a barrier to keep the cars off the tram tracks—there’s no traffic for the trams to be caught in. That’s why when people in officialdom talk about delays to trams, just talking about ‘traffic congestion’ is far too simplistic.”

There was also the opportunity to time the tram in the middle of the day on weekends, a time when trams are known to be significantly affected by traffic queues. “We certainly recorded a significant amount of dead time,” Dr Morton said. “But to put this in perspective, it’s about as much delay as occurs to Swanston Street trams at the same time, without any traffic queues to delay them.”

“The longest delays in Sydney Road, without question, occur in the morning peak on sporadic occasions when traffic queues back up Sydney Road from Brunswick Road. The longest of these I observed was a 20 minute delay. Yet even including these in the average, the average speed is faster than just to the south, on the reserved tramway in Royal Parade.”

Dr Morton urged attention to traffic light priority for trams. “Clearways seem to be mainly about squeezing in more cars. We have a great opportunity with a new tram operator coming to Melbourne, to refocus on what world’s best practice can do to speed up our trams.”

Summary of Results

Location, time and direction Av.travel time
(min:sec)
Av. speed
(kph)
Av.dead time
(min:sec)
Sydney Road, PM northbound, May–June 2009 9:12 16.3 1:19
Sydney Road, PM northbound, July–November 2009 8:58 16.7 1:22
Royal Parade, PM northbound 10:44 15.1 2:26
Sydney Road, AM southbound 10:20 14.5 2:01
Royal Parade, AM southbound 11:46 13.8 3:02

Observations were collected between 25 May 2009 and 18 November 2009. AM observations were collected between 8am and 8:30am. PM observations were collected between 6:00pm and 7:00pm.

The difference of 14 seconds between mean travel time in May–June and in July–November has a t-statistic of 0.77. The null hypothesis—that the true mean travel time is the same before and after the clearway extension—is not rejected at the 80% level.

The median travel time, at 9 minutes, is identical before and after the clearway extension. This is consistent with the running time according to the Yarra Trams timetable, which was not changed with the September 2009 timetable revisions.