The three-month initiative will favour an intelligence-led approach, in a move away from high-volume random breath test sites.

Assistant Commissioner for crime and operations Donna Adams said social media had been part of the change.

``One factor contributing to high awareness of major random breath test locations is the increased use of social media, with drivers easily and quickly able to alert others to avoid them,'' Ms Adams said.

``High-volume RBT operations will still take place, as studies have shown the importance of static RBTs in creating and sustaining deterrence, but our focus will shift during the trial to actively detect offenders.

``It's recognised that a more targeted, intelligence-led approach is more time-consuming but it's more effective.

``Mobile patrols accounted for 54 per cent of all drink-driver detections last year, compared to 19 per cent from static operations.''

During the trial, police will have a bigger focus on back streets, rural roads, private parties and concerts.

There will be an increase in marked and unmarked mobile patrols around licensed premises, including sporting venues and clubs that have temporary or special permits to sell alcohol.

Police will also step up use of the mobile breath-testing van in remote areas, and increase targeting of high-probability areas and individual people.

Ms Adams said evidence supported a shift in focus, with the latest National Survey of Community Satisfaction with Policing showing that Tasmania had the highest percentage of respondents -  10 per cent -  reporting that they had driven while over the legal blood-alcohol alcohol limit.

``In addition, alcohol has been the main causal factor in 20 to 25 per cent of fatal and serious injury crashes,'' she said.