$53.8m in science funding announced

University of Auckland statistician Ilze Ziedins' research on traffic and routing patterns was one of several receiving funding from the Royal Society of New Zealand's Marsden Fund. Photo / Thinkstock

Clogged roads and slow internet could be cured not with more highways or cables... but with less selfishness, says the recipient of a major grant.

A new statistical study will attempt to fix the problem of "selfish routers" - people or systems which make decisions that slow down transport and other parts of daily life.

University of Auckland statistician Ilze Ziedins' research is one of 88 projects which have attracted $53.8 million in funding from the Royal Society of New Zealand's Marsden Fund.

Dr Ziedins said past research has found that when travelling, either by car or through cyberspace, people chose their route without thinking of other people. Statisticians called this behaviour selfish routing, and said it was the cause of long delays and queues.

Road users could already choose to use GPS or webcams to plan their journeys. Backed by the $465,000 grant, Dr Ziedins will investigate whether this more detailed, accurate information could change people's behaviour and reduce congestion on the road or on the internet.

"The main thing we're looking at is: 'Will giving people more information lead to a better distribution of traffic, just in itself? If you more information about the state of the network, will your choices end up making the network work better?," she said.

The answers to solving these problems were not simply in adding lanes to highways or building more roads.

"In fact, increasing capacity may lead to worse behaviour overall," said Dr Ziedins.

"This is the paradox they found in Stuttgart, for example. They ended up closing roads because traffic flow had become so bad after adding capacity. After closing roads, traffic improved."

A small-scale study found that if selfish users of roads were given more information before they joined a network it improved the functioning of that roading system.

Dr Ziedens and colleague Mark Holmes would now attempt to see if that model works on larger, more complex networks.

The issue of selfish routing was also a problem for the internet, because routers chose the least congested path through cyberspace until that route became clogged, then switched to a less neglected route. Overall the transfer of data online was slowed down under this method.

Dr Ziedens has recently worked with Auckland City Hospital's cardiovascular ward to work out how to cut down on the long waiting times for treatment.

The hospital considered her probability model - which was based on the number of beds, staff, and treatment times- to plan the ward's expansion.

Applications to the Marsden Fund this year were extremely competitive. The 88 winning projects were chosen from 1078 preliminary proposals, which were whittled down to 250 full proposals.

Twenty-eight of the three-year grants went to Auckland universities.

Marsden Grants:
- Dr Stephen Banniser, GNS Science, $765,000
Seismologists will investigate the low frequency "chatter" made by slow-moving earthquakes which occur around 5km to 15km below Poverty Bay. They will investigate how fluids in the ground play a role in causing these "slow-slip" events to start, continue, and stop.

- Professor Mick Roberts, Massey University (Auckland), $390,000
Prof Roberts will try to find better ways to understand how epidemics such as swine flu and and SARS spread. He will develop models which show how epidemics spread through human contact and the projected rate of infection over time.

- Dr Andrew Taberner, University of Auckland, $830,000Dr Taberner will contruct an tiny device which simultaneously analyses five factors behind every single heart beat - force, contraction, heat production, oxygen consumption and calcium concentration. This cardiac myometer will increase understanding of heart muscle behaviour.