To take “Operations Research” course has been on my to-do list since I was a graduate student. Even though my thesis was done by using dynamic programming and non-linear programming, I solved a scheduling problem by using integer programming in my company.
However, I don’t have a chance to accomplish it due to time constraints. Now, thanks for the Coursera and NTU. I can take the course in my spare time.
The first course covered the fundamentals of Operations Research very well. It provides an overview of the field, workflow, a case study and a specific trick in practice.
Learned Things
Business analytics consists of descriptive analytics, predictive analytics and prescriptive analytics. From my experience, we use statistics and data visualization to handle descriptive analytics. We use data mining and machine learning to do predictive analytics. Operations research focuses on prescriptive analytics. Usually, the output or result of descriptive analytics and predictive analytics are the inputs for prescriptive analytics.
Operations research is a field of mathematical programming
- To support decision-making
- To make resource allocation
- To do prescriptive analytics
There are four types of mathematical programming
The steps to formulate a model:
In the math model part:
It is important to know the first thing to do is to list your decision variables instead of formulating your objective function.
To speed up the solution time, try your best to avoid integer variables and non-linear objective functions and constraints.
If your decision is quantity, the variable should be continuous. If you decide to assign a resource, the variable should be binary. You can use some tricks to make non-linear functions into linear functions.
Summary
The first class is suitable for a beginner to know what operations research is. I also review my knowledge in this field and learn new things. If you want to take the course, you can check the link.
My certification:
