Every year the NZTA conduct post implementation reviews (PIR) on a small handful of projects. They perform three functions, to see how the outcomes compare what was expected before the project was built, explain any variation in those measures and identify lessons that can be learned to improve other projects. As you’d expect, the PIRs often throw up some interesting insights into projects and we’ve covered a number of these before. Here are some of the interesting points I’ve picked up from the latest batch (for 2015/16).
This actually covers four projects, the Victoria Park Tunnel (VPT), the Newmarket Viaduct replacement, Greenlane auxiliary lane and some ramp metering. The PIR notes they “were reviewed together because of their proximity and interrelated influence on SH1 motorway traffic“. The VPT was the first of the governments Roads of National Significance to be completed, in part because already considered a high priority project and so much of the work needed had already been completed.
The summary confirms that while more vehicles are being moved by the increased capacity the projects provided, the initial time savings have already disappeared, a comment noted numerous times in the document. This is also significant as they note that the original economic evaluation forecast that travel time savings would account for 72.6% of all benefits. Positively, Safety has improved though with crashes down noticeably and these weren’t counted in the assessment as “were thought to be low“.
Here is one of the travel time summaries showing that trips are now slower than before the projects.
Overall the costs of the project were less than expected which was mostly due to VPT which was completed months early.
Finally, there are a number of lessons learned listed. I found these three the most interesting, particularly the comments about not making VPT 4 lanes.
This is the HOP system. On the project outcomes, the NZTA essentially claim that it is too hard to determine any benefits. This is in part because HOP has been just one of many improvements we’ve seen in recent years which is contributing to ridership growth.
One interesting comment about the forecasts notes that the original evaluation assumed that integrated ticketing would boost ridership by about 1% and that fares would add another 1.5%. Given the excellent growth that has continued to occur, it wouldn’t surprise me if that was much higher.
In their response to the findings of the review, Auckland Transport disagree with some of these comments, make some good points and highlight some interesting facts about HOP. They note that:
- Over 1 million HOP cards have been sold against an initial forecast of 350k.
- The system was designed to handle 500,000 transactions per day but is regularly doing 70-90k more than that.
- They note that they benchmark operating costs against Transport for London’s Oyster card. They say using the measure of operating costs versus system revenue, TfL manage 8.8% while AT sits around 9%. I recall the Northern Gateway toll road is something like 30% so AT aren’t doing too badly.
Perhaps their best response is to the comment about bus travel times, effectively saying that without HOP, times would be much worse than they are now.
The Northern busway runs to two-minute interval timetable at peak times how would that be possible with a paper ticket system where each sale takes at least fifteen seconds and the bus loads both at the front and back of the bus.
I quite like this side of AT defending themselves, they should do it more often.
At $99.2 million, the costs of the project were quite a bit higher than expected (38%) and they say that was mainly due to scope changes made by the NZTA. The project was also two years late which is attributed to both scope changes and dealing with the Snapper debacle.
This project realigned 4.5km of SH2 between Napier and Gisborne, shortening it by 1.4km. This includes a substantial bridge across the gorge and a new crossing of the since mothballed rail line. They say the proposals for the project have been around for over 50 years and that it’s been assessed multiple times in the 20 years prior to being built. The project was one the government accelerated in 2009
The PIR makes a number of fairly critical statements about the assessment of the project, especially the forecasting, which it calls “basic”. This modelling included using a base level higher than the actual counts and using way overoptimistic growth assumptions.
On top of this, travel time savings, while good, haven’t been anywhere near as strong as was forecast with an average of 2 mins 37 seconds vs 4 mins 18 seconds.
The most positive aspect to the PIR though is the cost. Initially expected to cost over $31 million, it ended up being built for just $19.7 million
As the name suggests, this bypasses Kamo which is in the north of Whangarei and like a few other projects in this post, includes a crossing of a rail line (although this one is not grade separated). It follows Stage 1 was completed in 2005.
Like other projects, it appears that the travel time savings forecast, which were a significant proportion of the assumed benefits, haven’t eventuated.
In addition, the project cost more than expected and is having a number of issues which have required remedial work, including already having to be resurfaced.
This was another realignment project that also involves crossing a rail line (grade separated). It was primarily built to improve safety in an area with a high crash rate but improving travel times was a factor too. The new route is shown in red below.
The NZTA say the project has been successful in reducing crashes with all crashes reducing from 42 in the 5-6 years prior to construction to just 7 in the ~4.5 years since. Injury crashes reduced from 10 to 4.
Travel time savings were also seen but times are now back to what they were before the project, which the review says can’t be explained as there has been no changes in the area or significant traffic growth that could have influenced speeds.
As for costs, the project was completed early and $1.3 million under budget for a total of $13.6 million.
One of the most notable comments in all of these PIRs are those about travel time savings, or the lack of them. That’s important as so many projects are justified based on travel time savings and yet at best they’re gone within just a few years. How would it change what projects we built if we didn’t include travel time savings?
What is good are the improvements in safety many of these projects have delivered. I wonder what could be achieved if instead of spending billions on low value motorways, we instead focused say $1 billion of that on safety improvements. The NZTA could probably deliver 50-100 of the types of rural state highway projects shown above which could contribute significantly to improving our dreadful road toll.