DBA3701 / DSC3214 Introduction To Optimisation

Module: DBA3701 / DSC3214 Introduction To Optimisation

Semester taken: AY 2020/21 Semester 2

Lecturer: No lecturer, module is sectional-based

Tutor: Dr Chaitanya Bandi

Textbook: Multiple, but not strictly necessary

What it is about

This module provides students with a glimpse on optimisation techniques that are available in today's world, along with examples in logistics and pricing. Students will be exposed to the mathematical theories involved in optimisation, as well as examples on how these optimisation techniques can be used in real life. Topics include linear optimisation, network optimisation, integer programming and non-linear optimisation.

Assessment components

The most accurate assessment breakdown was not provided, due to various changes throughout the semester. The components are:

• 3 graded homework assignments (originally supposed to have 5)

• Final project report and presentation

• Final exam

Comments

This was the first semester that this module is offered by this professor, and I specifically took it as he previously studied at MIT and was a lecturer at the Kellogg School of Management. It was a different teaching style, and definitely something that I was not prepared for, as the module was rather messy and the professor did not have all the materials ready.

This module also requires some linear algebra knowledge, which luckily I took in the previous semester. It is not entirely necessary, but knowing linear algebra helps in understanding some of the concepts taught in the module, especially with eigenvectors involved. The professor provided us with some materials to self-study, but they were barely sufficient and you are better off finding good YouTube videos that explain the concept better.

Overall, this module can be a little heavy in theory and mathematics, which may not be suitable for some business students who are more comfortable with presentations and fluff. Most students taking this module will be specialising in Business Analytics anyway, and they should be comfortable with the content taught in this module. Students coming from Operations and Supply Chain Management will need to play a little catch-up.

Sectionals

There is a 3-hour sectional teaching session per week, and it is conducted similar to a lecture style where the professor will go through the content in the lecture slides and also conduct some demonstration using Jupyter notebook and Gurobi. There are also opportunities to ask the professor questions during the session, similar to how other sectionals are conducted in NUS Business School.

Homework and Project

In this semester, there were 3 graded homework in total. Some of the questions in the homework are done individually, whereas the others were done in groups. Originally, there were 3 people in a group, and the individual and group answers are compiled together and submitted to the professor. However, the groups were merged halfway through the semester, as there were too many groups. The combined groups were mainly for working on the final project.

For the final project, we are supposed to come up with a topic on our own and show how optimisation techniques can be used to solve the problem. However, we were also given the option to ask the professor for a topic that he had in mind. We decided to ask for one, as it was easier in our opinion, and were given a topic related to food production and distribution. We were also promised to be given data to work on, but the professor kept delaying, and we eventually decided to come up with some mock data on our own based on publicly available sources.

Unfortunately, the combined group was difficult to work with, as the other half of the group were very reluctant to work together. They were all staying together at Cinnamon College of the University Scholars Programme (USP), and had secretly met up to finish the project. This was the second time that I have had bad experiences working with USP students, the first being in IS1103. Luckily, we could still "save" ourselves by working on the group presentations, and the group project was still a satisfactory submission at the end.

Final Exam

The final exam was held physically in one of the seminar rooms in BIZ1. The format of the exam was pen and paper, and there were a total of 4 questions (with multiple parts in each question). The exam was closed book with only one double-sided A4 size cheatsheet. The professor provided some sample questions for us to practice, which are based on the tests given in past semesters. The final exam tested all topics except for non-linear programming (which itself is an extremely difficult topic).

Being a math-based module, it is important for you to practice as much as possible, so that you can train yourself to see patterns in the questions and become more confident in answering the questions during the exam. The paper itself is decent, and is a similar standard to those given in the homework assignments. However, the bell curve may be a little steep as the students taking this module are often more mathematically inclined from their training in Business Analytics, so always check your answers to avoid making careless mistakes.

Other information

Assignment workload: Moderate. There were 3 graded homework assignments given.

Project workload: Moderate, although most of the project is done in the second half of the semester.

Readings: None

Recommended if: You are interested in optimisation (which can be quite useful in a variety of fields), but do not wish to take them under the Mathematics department.

Rating: 3.5/5. The content of the module is decent, but had a bad experience with how the module is organised, and of course bad group mates.

Expected grade: A-

Actual grade: A- (thankfully)