The Highways Agency has a glut of data on the condition of its roads, but it is costly to collect and just about useless. That's changing.
With modern survey methods a tidal wave of road condition data is pouring into the Highways Agency today. Properly managed, this data can help those responsible for road maintenance meet their performance targets, it can help the quest for greater efficiency, and can, ultimately, ensure safer, more reliable road journeys.
Capturing this data is relatively easy (see box). But by itself it is costly and just about worthless.
To be used properly it must be interrogated, analysed and understood. This is another matter and is only now becoming possible.
'In the last three years there has been a very significant improvement in the effective management and use of data, ' says independent consultant Nick Lamb, who started advising the Highways Agency Pavement Management System (HAPMS) team six years ago.
'When I started with the HAPMS team, the immediate task was software specification and procurement. The business case for finding a much more sophisticated and useful way of managing the data had already been made. There was a whole raft of problems with the old DOS based software the Agency was using to interrogate its data.
'Most of these problems could be traced to a single factor, ' he says. 'The Agency was using five different systems that couldn't talk to each other.'
Non-communicating software systems - Babel Syndrome - is a recurring theme in the public sector, but the effects were particularly acute in the case of the Highways Agency.
Properly informed road works decisions are impossible without access to the whole picture.
But data was sometimes lost - for example when managing agents (MAs) or their contractors changed. Proper evaluation of the inventory was impossible.
'The legacy systems were not all bad, ' explains Lamb. 'They allowed the Agency to identify the roads in the worst condition, which is of course essential. But dealing with 'worst first' is not necessarily the most economic or efficient way to schedule works. The Agency needed the data to tell them where they could make the most timely and cost efficient interventions.'
HAPMS project director Les Hawker, who was responsible for putting the current team together in 1998, admits that the Agency needed much smarter software: 'What I saw at the time was that the way we wanted to interrogate the data had implications far beyond better data management.
'If we could get the software to do what we wanted it to do, it would effectively allow us to completely re-engineer our business process, ' says Hawker.
'What we were really looking for was 'a single source of truth' available to us and all our suppliers.
Hawker's single source of truth turned out to be Confirm, an infrastructure management package developed by Southbank Systems and already in use by a number of local authorities including Surrey County Council and the London Borough of Merton.
'The starting point for any effective highway data management system is a definition of the network, ' says Lamb.
Having a definitive, detailed model of the motorway and trunk road network is essential for a number of reasons. Not least because the permitted duration, density and road length of works that interrupt or close a carriageway are governed by performance measures agreed with the Department for Transport.
MAs, or managing agent contractors (MACs) must make works scheduling decisions based on a view of the whole network.
Confirm allows MAs and MACs to access the whole network picture independently and refers proposed road works to closures already booked into the system.
This helps avoid the sort of pillar to post road works that once blighted so many journeys.
The second must-have for Hawker was a full inventory, that includes more than 5,000 emergency telephones and 50,000km of white line.
The business case for making it happen is compelling. Where contractors are working to fixed price contracts anything that generates greater efficiency - say by ensuring the correct replacement telephone is picked and delivered - is immediately felt as improved service for no increase in cost.
Capturing the data
Historically, most road condition data came from visual surveys by managing agents (MAs) and their contractors on the ground.
This was expensive and the data, gathered by individuals across a wide network was, inevitably, inconsistent.
More recently the Highways Agency has been collecting data using road assessment vehicles. There are three sorts of survey data collected.
SCRIM surveys measure wet skidding resistance while deflectograph surveys measure the resilience of the road structure.
But it is the Traffic Speed Condition Survey (TRACS) that is responsible for the giant step change in the quality and volume of the data now available.
TRACS vehicles record eight different types of data including cross-fall, transverse profile and cracks.
Bristling with lasers, cameras and strobe lights, and laden with laptops, generators and refrigeration (the strobe lights have to be kept cool), TRACS vehicles assess the condition of the road while travelling at the prevailing traffic speed. This offers the big advantage that data collection and road surface evaluation can take place without closing the highways.
The sheer volume of data collected is staggering. Road surface texture, for example, is measured by the TRACS lasers every 0.1 mm then averaged out by the onboard computers to produce a reading for every 1mm travelled - 10M measurements per kilometre. Further analysis of that data reduces a typical survey's 500 megabytes of information into just one megabyte for delivery to the Confirm infrastructure management system.