Optimization and Decision Support Methodologies

Before attending this course, students must have passed the Operations Research course unit (2nd year of this degree).

A learning strategy based on the experimentation of the material exposed in the theoretical class is used, either through the resolution of theoretical-practical exercises on the topics covered, or through the computational resolution of the same.

• In the theoretical component the subject is exposed, while the active participation of students is encouraged by frequently asking questions to be answered by them.
• In the theoretical-practical component, students manually solve exercises for the application of the theoretical component.
• In the practical-laboratory component, students solve computationally various types of problems, using Python libraries.

Based on the concepts acquired in Operations Research, in this course it is intended that students broaden their knowledge in the areas of optimization and decision support. In this sense, methodologies will be introduced to be applied to decision problems that are more complex than those previously studied, such as problems involving multiple objectives, integer variables, among others. After attending this course, students should: 1 – Know and understand the fundamental characteristics of the most representative problems of optimization and decision support; 2 – Identify the different approaches that can be used to solve them; 3 – Solve simple practical problems using the appropriate optimization and decision support algorithms and interpret the obtained solution(s); 4 – Be able to computationally implement some of the algorithms / use computational tools, to solve the types of problems addressed.

Theoretical program

0.        Linear Programming Review
1.        Post-Optimization and Sensitivity Analysis
2.        Integer Linear Programming
3.        Multi-Objective Decision Problems
4.        Forecasting Methods

Theoretical-practical/practical program:

• Manual / computational resolution of several types of problems 

Two possible assessment modes:

a) Final exam

• Exam (17 points): covering all the subject taught.
• 2 Practical mini-tests (3 points = 1.5 + 1.5): to be solved in the practical classes.

b) Tests during the semester

• 2 Tests (17 points ​​= 8.5 + 8.5): covering together all the subject taught.
• 2 Practical mini-tests (3 points = 1.5 + 1.5): to be solved in the practical classes.

Notes:

• The assessment tests will be held during the theoretical / theoretical-practical classes (dates to be confirmed).
• The 1st test will cover the 1st and 2nd chapters of the subject, while the 2nd test will cover the last two chapters.
• A minimum classification of 35% is required in each of the tests.
• Failure in taking one of the tests, or failure in obtaining the minimum score in any of them, implies the student's passage to the final exam assessment system.
• The practical mini-tests will be carried out during practical classes, on dates to be confirmed, and will be based on the analysis and adaptation of the computational solution of an optimization problem in Python.
• There is no imposition of a minimum grade for these mini-tests.
• The absence of a student to the mini-test, leads to a grade of 0.0 points assigned to that mini-test.
• Students who obtain a grade greater than or equal to 9.5, in the sum of the test scores with the mini-test scores, provided they have met the minimum required, will be automatically exempted from taking the final exam.
• The final exam will cover the entire subject and cannot be solved in parts.
• The practical mini-test grade will be valid for all exam periods. However, in the special season, the exam can be carried out for 20 points, if the student wants.
• Assessment tests and exams will be without consultation, except the one provided by the professors.
• In the tests and exams, students will be able to use a non-graphic calculator.
  

NAO