ANPR as a low-cost alternative to the proposed East Reading MRT

(Observation on East Reading MRT Planning Application: Wokingham – application number 172048; Reading – application number 171108).

Dr John Walker MA, MSc, PhD, CPhys, FIET
Honorary Secretary, ITS(UK) Road User Charging Interest Group
Visiting Senior Research Fellow, Transportation Research Group, University of Southampton, UK

Executive Summary

A limited road charging scheme on the A4 in East Reading at peak hours to reduce congestion, using Automatic Number Plate Recognition (ANPR), is much cheaper than the proposed £24 Million East Reading Mass Rapid transit Scheme (MRT). The capital cost would be around £31k, with annual cost £3k and no environmental disruption or visual intrusion.

The resulting reduction in congestion would reduce delays to buses on this stretch of their route, and improve journey time reliability, as well as reducing delays for other traffic in East Reading.

Integrated ANPR modules cost £5k, plus 4G mobile communications of £30/month, installation £1k-£3k (depending on whether existing poles or gantries can be used), and annual maintenance cost of £300. The existing Reading bus lane ANPR Back Office could possibly be used, or Transport for London may offer to provide this service, as they have offered to for similar schemes in the past.

An ANPR module on each side of the A4 between Cumberland & Amity Roads, and another pair of modules between Liverpool Rd and the A3290 roundabout, monitor east-bound and west-bound traffic; this makes an initial capital cost of 4*£5k =£20k, installation cost of around £8k, and annual running costs (4G communications plus maintenance) of £3k.

Such a scheme could generate income, depending on the charge made, though ideally, especially for public acceptability, it should just be used as a traffic management scheme and charges should just cover the costs of the scheme. A similar scheme in Saddler St in Durham (see figures 1-4 below) which has been operational since 2003 charges £2/day.

An alternative is a Dutch scheme (“spitsmijden” – “avoiding peak traffic”) which involves paying drivers a few pounds per day to avoid the area during peak hours – monitored by ANPR.

The proposed charging scheme, and the “spitsmijden” alternative, both need traffic modelling before implementation, to optimise the scheme, and to decide whether additional cameras are needed on other roads such as the Wokingham Rd and the Kings Rd, plus traffic monitoring during implementation to measure the benefits. Additional cameras would of course increase the cost of the scheme, but even if it works out an order of magnitude more expensive than the £30k quoted above, at £300k it would still be two orders of magnitude cheaper than the proposed MRT at £24 million, and would not incur significant environmental damage or visual intrusion alongside the river Thames.

Introduction

New roads & improved Public Transport do not solve road traffic congestion problems. According to the classic paper by Duranton & Turner of Toronto University (“The Fundamental Law of Road Congestion: Evidence from US Cities”, published in 2009), the total Vehicle Miles Travelled (VMT) is proportional to road lane-miles – in other words, the amount of road space. And building more roads does NOT result in diversion from other roads. Furthermore, the provision of extra public transport has NO EFFECT on VMT.

The evidence from cities such as London, Stockholm and Milan is that the optimal way to control road traffic congestion is to introduce road pricing/congestion charging. Furthermore, the evidence from these and other cities world-wide is that road pricing is acceptable to public opinion, and to voters, provided that:

  • it is equitable;
  • it is revenue-neutral (i.e. there are compensating reductions in other taxes and charges), or that any surplus revenues (over and above the cost of running a scheme) are reinvested in transport;
  • it has a low cost overhead;
  • above all that the people affected have experience that road pricing works.

As the authorities in Minnesota and Milan have stated, the best way to get public acceptability is “Show, don’t just tell” and “introduce first, get acceptability later” – the evidence from an existing scheme is far more persuasive than anything else. For much more detail of this see Walker (2018).

Automatic Number Plate Recognition (ANPR) for traffic and congestion management

The price of ANPR systems has dropped considerably over the last 25 years, as is typical with any technology that sees wider use. There has also been a general movement from cameras connected to a controller to integrated units complete with illumination, processing and communications. A typical integrated camera, which can cover up to three lanes, will cost around £4,500.

Typical monthly communications costs per module, using 4G mobile communications, would typically be £30 per month. If there is a hardwired or WiFi communications network already, the communications cost may be zero. Installation cost of each camera module, if it can be mounted on existing street furniture, is relatively cheap and easy (e.g. £600); a new pole and foundation is more expensive (e.g. £3,000). Location and traffic management requirements are also relevant.

Back office costs depend on what method is used – physical server location or cloud based, just data capture or managed service with revenue collection. It could be as little as £10k, especially if it can use the existing Reading bus lane enforcement back office, or as much as £200k if implemented from scratch. An alternative would be to subcontract it to an existing road user charging back office such as Transport for London’s; TfL have offered this service in the past.

Other costs would include annual maintenance & support (typically £300 per year per camera), Scheme Design, Project Management, Integration with existing systems, Legal advice, Signage regime etc. These are difficult to quantify in advance but are unlikely to cost anything like as much as the MRT at £24 million.

European number-plates – of which there are probably not very many in Reading – tend not to be a problem, as long as they aren’t caked in dirt and are in the correct format.

Calculations for East Reading MRT

So Capital Cost would be say £5k/module, with average installation cost of say £2k, making £7k per location, and annual running costs of £300 pa per module or site – say £3k in total.

Back office costs are difficult to assess at this stage, depending very much on whether the existing bus lane monitoring scheme could be adapted and used.

Eastbound and westbound traffic along London Rd

It would seem that a camera system on the A4 between Cumberland Road & Amity Road, and another between Liverpool Rd and the A3290 roundabout, would be sufficient to monitor out-bound/eastbound traffic, with a similar pair at the same locations to monitor inbound/westbound – a capital cost of 4*£5k + installation at £8k plus £3k annual running costs =£31k would seem to be adequate, in addition to whatever the Back Office might cost.

There would also need to be targeted road traffic monitoring during the implementation phase to measure the benefit and gauge whether changes to traffic require further action.

Such a scheme could generate revenue, though ideally, especially for public acceptability, it should be used only as a traffic management scheme, with charges just covering costs of the scheme.

A similar scheme -Saddler St in Durham

A somewhat similar scheme has been operating in Saddler St in Durham since October 2002 – it was in fact the first congestion charging zone in the UK, predating the London scheme but covering only one street in the centre of Durham. It originally used rising bollards to control access but these were replaced by ANPR in 2011. See figures 1-4. It currently charges £2/day.

Saddler Street in Durham which is part of a congestion charging scheme.
Figure 1: Saddler St in Durham
Congestion charging zone showing access to Market Place and Saddler St
Figure 2: Access to Market Place and Saddler St
Entry sign to congestion charging zone.
Figure 3: Sign indicating entry to charging zone
Entry and exit signs at the Durham congestion charging zone
Figure 4: Signs indicating entry to and exit from the charging zone

Potential problems with the Reading scheme

With locations like Saddler St in Durham, motorists have the option not to drive up it. But in the case of the London Rd in Reading, inbound and outbound traffic, apart from changing the time of travel, has only the Wokingham Rd as the alternative route. So it may be necessary to have cameras on the Wokingham Rd just east of Cemetery Junction – and similarly on the Kings Rd just west of Cemetery Junction. This would increase the cost of the scheme, but even if it works out ten times as expensive as the £30k quoted above, at £300k it would still be only around one hundredth of the cost of the proposed MRT at £24 million. And in a larger scheme there would be potential economies of scale; a small implementation has a disproportionately large cost to set it up and manage it.

“Spitsmijden” (‘avoiding peak traffic’)

An interesting alternative technique to address traffic congestion is “spitsmijden” (‘avoiding peak traffic’): motorists, rather than paying to drive, are paid NOT TO DRIVE, at certain times (especially peak hours) and locations (such as roadworks). Used only in the Netherlands so far, “Spitsmijden” projects have been shown to be an effective and cost-effective congestion reduction measure, using small payments of between 1 euro and 6 euros per day. Especially during large road construction works, Spitsmijden projects can significantly mitigate the negative effects of capacity reduction and increased travel time, and the effects may persist long afterwards. They can also be used to alleviate recurrent congestion on roads, either as an alternative to, or in advance of, road charging schemes. Spitsmijden seems to work well for such localised areas; whether it would work equally well on a larger scale is yet to be determined.

For more details of “spitsmijden” see chapter 20 of Walker (2018)

Recommendations

The proposed charging scheme, and the “spitsmidjen” alternative, should be investigated thoroughly before any MRT scheme is implemented, modelled using suitable traffic models, and monitored during implementation, to ensure that the scheme is optimised; badly designed schemes may make congestion worse.

Such schemes would not only be much cheaper but would not incur significant environmental damage or visual intrusion alongside the river Thames, unlike the MRT. The reduction in congestion would reduce delays to buses on this stretch of their route, and improve their journey time reliability, as well as reducing delays for other traffic in East Reading. And they would leave open the possibility of proceeding with the MRT subsequently in the unlikely event that they are found not to work.

References

Duranton G & Turner MA (2009) “The fundamental law of road congestion: evidence from US cities”, Working Paper 15376, http://www.nber.org/papers/w15376, National Bureau Of Economic Research, Cambridge, MA 02138, September 2009

Walker, J (2018) “Road Pricing: Technologies, economics and acceptability”. Published by the Institution of Engineering and Technology, February 2018. The contents of this book can be viewed at http://www.theiet.org/resources/books/transport/rdpri.cfm
Dr John Walker
25 April 2018

Leave a Comment