Scheduling boats, crews and missions

CSIRO scientists have worked with the Royal Australian Navy and the Defence Science and Technology Organisation (DSTO) to solve some important fleet scheduling challenges.

In planning a replacement for their current Fremantle-class patrol boats, the Navy has needed to determine a fleet size that will be viable in terms of maintenance, crewing and deployment commitments. Another challenge has been to find ways to deploy the activities of the fleet more efficiently. The two issues are related because in general, scheduling the fleet's activities more efficiently involves a more intensive use of boats, and can make smaller fleet sizes possible.

One way to improve efficiency is through "multi-crewing" arrangements. When crews can be alternated on each boat during a given period, fewer boats are needed than crews because the boats require less time in port than their crews. Multi-crewing thus permits more intensive utilisation of the boats. However, because each crew is no longer tied to a particular boat, special attention must be given to the provision of training, leave and other working conditions.

CSIRO's task was to determine week-by-week timings for mission-deployments and maintenance activities for the Navy's patrol boat fleet over the course of a year, and to assign those activities amongst crews, boats and ports. The scheduling conditions included the following.

• Missions should be spread evenly over time, subject to specified time-windows
• There must be sufficient boats available at all times to handle unplanned “surge” missions.
• Maintenances should be planned within pre-given time-windows, and to conform to the capacity of port facilities.
• Mission-deployments undertaken by each crew should be spread evenly over time and should include adequate allowances for leave and training.
• Each crew must have adequate leave, coinciding with school holidays where possible.
• Limits were placed on the number of crews assigned to each boat, and of boats assigned to each crew; in addition, the number of "handovers" between crews was to be minimised.
• Additional travel costs due to remote handovers between crews with different home ports should be minimised.

These conditions made for an irregular optimisation problem which is unsuitable for treatment with any conventional "one size fits all" package.  CSIRO's expertise in the mathematical solution of such problems is demonstrated in a software package developed to address the conditions outlined above.

The package, called CBM (“Crews, Boats and Missions”), provides a week-by-week schedule of activities for each boat and each crew, so as to meet the conditions outlined above. CBM has already proven useful in the planning of the replacement fleet, and may also be valuable as the basis for development of a “hands-on” tool.