This unit will be a core elective in the Master of Artificial Intelligence to be introduced in 2020.

Combinatorial Optimisation technology is key for providing good solutions to decision making problems that appear in every area of our lives. It is a research strength of the Faculty and has led to successful collaboration with many industries, including Melbourne Water and Woodside. Yet, FIT does not have any units that teach modern optimisation modelling and solving methods at either the UG or PG level. This misalignment contributes to the lack of access our research group has to HDR students with the appropriate background.

17/04/2019 - Admin - updating 100% in-semester justification to include validation on behalf of the CE.

18/04/2019 - Guido Tack - update assessment to include electronic exam (to address authentication of assessment)

29/10/2019: Updating the prerequisites to include new foundation units. Effective semester 1, 2020.

11/11/2019: Adding prerequisite clause for C6007 Master of AI students to enable enrolment. Effective 2020.

12/12/2019: Changing activity type from "lecture" to "workshop" in Workload/ContactHours, to reflect the fact that the unit is taught in flipped classroom style. Effective semester 1, 2020, (Activities in Allocate+ for S1 2020 were already created as "workshop").

18/09/2020 - Admin: Update to include new assessment and teaching approach fields as per Handbook requirements.

25/05/2021 - Changing assessment structure to 3 assignments instead of 4, and changing weight of final exam from 35% to 40%, to address high in-semester workload. Also fixed small typos in unit description and required technology.

On successful completion of this unit, students should be able to:

1. model a discrete optimisation problem using a mix of basic and more advanced modelling techniques in a high level modelling language;
2. interpret and explain models written by others;
3. explain how models are mapped to solver-level input;
4. identify and fix errors in models;
5. evaluate the limitations, appropriateness and benefits of different modelling patterns for common problem classes;
6. evaluate and improve the efficiency of models by applying different model transformations.

020307 Decision Support Systems

This unit introduces the fundamentals of modelling for discrete optimisation, focussing on how to rigorously express a discrete optimisation problem in a manner that it can be solved. Topics covered will include decision variables, basic constraints, modelling with sets, modelling with functions, multiple modelling viewpoints, modelling time, common modelling patterns, model translation, and debugging discrete optimisation models. We will examine complex real world problems and see how they can be translated so that they can be solved by modern discrete optimisation technology.

Technological requirements

All code examples, lab tasks and assignments use the MiniZinc constraint modelling language. MiniZinc is available free from https://www.minizinc.org for Windows, Linux and macOS. We recommend that you install MiniZinc on your own laptop, however you can also access it on MoVE at https://move.monash.edu.

On-campus

Active, Problem-based and Peer assisted learning

Examination (2 hours): 40%; In-semester assessment: 60%

1. In-class participation - 10% - ULO: 1, 2, 3, 4, 5, 6
2. Assignment 1 - 5% - ULO: 1, 4
3. Assignment 2 - 15% - ULO: 1, 4
4. Mid-semester test - 5% - ULO: 1, 2, 4, 5
5. Assignment 3 - 25% - ULO: 1, 3, 4, 5, 6
6. Final exam - 40% - ULO: 1, 2, 3, 4, 5, 6

Minimum total expected workload equals 12 hours per week comprising:

• Two hours/week workshops
• Two hours/week laboratories

A minimum of 8 hours per week of personal study for completing lab/tutorial activities, assignments, private study and revision.

FIT9131 or FIT9133 or FIT9136.

Students enrolled in C6007: None.