BAN437 Stochastic optimization using algebraic modelling languages

• Topics

  The seminar will introduce uncertainty, in the form of random variables, into optimization models. The focus will be on why we need this, and what can go wrong with deterministic modeling. We shall use professional software - AMPL - to formulate and solve simple numerical models, so as to see how solutions change to adjust to the uncertain future. In other words, see how tactical and strategic decisions change to facilitate operational handling of uncertainty. The technical level of the course is limited, we focus on concepts and simple examples. 

• Learning outcome

  By the end of this seminar the students

  Knowledge

  • are able to explain and discuss key concepts in decision making under uncertainty, and understand why deterministic modeling does not deliver what it promises
  • understand why taking uncertainty into account leads to better use of resources
  • are able to verbally formulate relevant research questions where uncertainty in optimization is involved

  Skills

  • are able to algebraically formulate simple optimization models involving uncertainty
  • are able to solve simple problems using high-level optimization software, like GAMS or AMPL

  General competence

  • are able to use computational tools for implementing and solving simple decision models involving uncertainty 

• Teaching

  The course is taught intensively over one week, combining teaching sessions with work on the implementation of simple models using AMPL.

• Required prerequisites

  Students need to have taken a basic course in optimization, operations research, management science, industrial engineering or something similar so as to be able to read and write simple optimization models, including the use of high-level software at a basic level (like AMPL or GAMS - not just Excel). For NHH students, BAN402 is a perfect background.

• Credit reduction due to overlap

  None.

• Compulsory Activity

  Mandatory attendance (100%).

  Course approvals from spring 2025 remains valid. Course approvals acquired in earlier years are no longer valid. 

• Assessment

  A group presentation (2-5 students in a group) on a given topic.

• Grading Scale

  Pass-Fail.

• Computer tools

  Standard laptop, AMPL modelling language with solvers CPLEX, Gurobi and MINOS (licenses will be provided during the course).

• Literature

  To be given in Canvas