A big data approach to schedule optimization

Workforce scheduling algorithms are used by businesses around the world, including hospitals, factories, and retail stores, to determine when and where employees should come to work. Usually, the needs of organizations change over time, but the scheduling algorithms used in these systems usually do not change. Kronos is an industry leader in scheduling software, and has access to a lot of data reflecting schedule changes in many organizations. This research project leverages big data methods in order to mine this data for interesting patterns, and to provide scheduling algorithms that adapt automatically in response to new requirements. In particular, we aim to learn a model of scheduling user intentions from the data, and use it to infer changes to the schedules provided by Kronos software that the user would be likely to desire.

Faculty Supervisor:

Doina Precup

Student:

Emil Janulewicz

Partner:

Kronos Canadian Systems Inc

Discipline:

Computer science

Sector:

Information and communications technologies

University:

McGill University