Information About This Unit
Table of contents
Context in the Track and MSc Program
This website is for the Probabilistic Design unit that is part of a B module in the Hydraulic and Offshore Structures (HOS) track in the Civil Engineering MSc program at Delft University of Technology in the Netherlands. In this case, the term “unit” can be considered synonymous with the term “course.”
The Hydraulic and Offshore Structures (HOS) track offers a choice between three B modules during the fourth quarter of Year 1, each of which focuses on a different set of structures:
- CIEM4210 Marine Renewables
- CIEM4220 Dams, Dikes and Breakwaters
- CIEM4230 Floating and Submerged Structures
Each B module has three units:
- Unit B-1 Probabilistic Design (this unit),
- Unit B-2 Computational Modelling, and
- Unit B-3, which covers design aspects of the specific structures related to each module listed above.
The units Probabilistic Design and Computational Modelling introduce fundamental methods that are independent of the structure considered and essential for all three B modules. This is why they have separate websites (otherwise there would be 3 separate Brightspace pages! Links to these pages are provided at the top right of this website).
What We Will Learn
Although each of these B modules look at a wide range of structures, there are a couple important things in common: hydraulic loads govern the design of each system and structure; and failures can lead to significant consequences related to economic damages and loss of life. There is a high level of uncertainty in the hydraulic loads as well as the geotechnical conditions where these systems are operating. This makes probability, risk analysis and reliability-based design key tools for the design, assessment and maintenance of these structures. The purpose of the unit CIEM42X0 Probabilistic Design is to support you in this effort for your system and structure of interest. As such, the unit is built on several key elements:
- Three primary technical topics: 1) extreme value analysis, 2) component reliability and 3) system reliability
- A series of (online) reading assignments and exercises (homework)
- A weekly lecture to cover theoretical aspects of each topic
- A weekly workshop that will apply concepts to design cases
Weekly Topics
Topics covered each week can be found via the Calendar page.
What We Will Do
In-Class Sessions
The Probabilistic Design Unit meets regularly on Tuesdays and Fridays each week (no meetings during week 3). All sessions on Tuesday start at 15:45 and end at 16:45. Sessions on Friday start at 9:45 and end at 12:30. Tuesday is a lecture format; Friday starts with a short introduction followed by workshop assignments completed in groups (laptop required) and finishes with a short discussion (participation required). Exercises and reading will be provided each week to complete at home (2 hours max) to prepare for the Friday sessions (homework is not turned in or graded). Workshops 2, 5 and 8 are turned in as part of the exercise portfolio and will receive feedback and a grade from the instructors. Contribution of the Probabilistic Design Unit workshops to the final module grade depends on which unit you are in (i.e., 4210, 4220, 4230).
Week X: Outline of a Typical Week
- Lecture
- Short description of the lecture
Tuesday, 15:45-16:45, Hall 1.98 - Homework
- Short description of homework (reading, exercises, software setup, etc.)
what you should do at home, before Friday! - Workshop
- Short description of the workshop
Friday, 9:45-12:30, Hall 1.95
The main goal for the week are provided in boxes like this:
1. along with a sub-goal
… and more sub-goals
n. until the n-th sub-goal
Links to reading and study material are provided in boxes like this.
Links to exercises are provided in boxes like this. The workshops that are to be turned in for a grade and feedback as part of the exercise portfolio use a button like this: Portfolio
Collegerama Recordings
During the 2023-24 academic year Collegerama is recorded via CIEM4220 Dams Dikes and Breakwaters, so the Probabilistic Design unit recordings will be mixed with the lectures from that unit. The list of all videos can be found here, and the links for each recording will be added to the weekly overview pages directly.
Programming
Programming in Python and use of a command line interface will be required to complete most of the workshops, and some (not all) homework exercises. We will build on concepts covered in MUDE and use GitHub (not GitLab) to receive and submit assignments. We will also introduce VS Code as an IDE rather than Jupyter Lab. Workshops that are part of the exercise portfolio will require a Markdown-formatted report to be included in the repository.
Assignments
The term assignments is used generally to refer to the following:
- Homework: something you do at home that is meant to prepare you for the in-class sessions. Not required, but (obviously) strongly recommended.
- Workshops: the in-class assignments we do on Friday. Three of them will be submitted for feedback (see Assessment page for information on how this contributes to your module grade)
- Exercise Portfolio: the collection of three Workshops that will be submitted for feedback.
Assignments will be provided via GitHub, and you will receive a link on the relevant page of this website for each assignment. You will need a GitHub account to access the assignments. If for some reason you do not want to create a GitHub account, please contact Robert directly.
Grading
The three workshops will each be given a grade of 0, 5, 7 or 9, along with qualitative feedback, via a Markdown file in your repository. The three grades will be averaged to determine the final grade for your Exercise Portfolio, and depending on your overall progress, the final value may be rounded up or down.
You are generally expected to return answers in Markdown reports, similar to those used in MUDE. A good report includes judicious use of Markdown formatting (Latex equations, bold/italic font, etc), figures and tables to illustrate the answers to questions, as needed.
In general, the grades correspond to:
- 0 points: did not submit, or did almost nothing in submission
- 5 points: made an attempt, but a significant portion is incomplete, or a significant misunderstanding of the material is apparent. Formatting of report is difficult to read and analysis is hard to follow, such that significant time is spent interpreting what you meant to say. Does not use figures, tables or Markdown formatting to illustrate results, or uses a minimal number ineffectively.
- 7 points: some errors or misconceptions, but generally acceptable results. Formatting of report is acceptable but not polished and/or concise. Figures and tables may be included, but formatting or purpose is not clear.
- 9 points: thorough understanding of material and interpretation of results is obvious. Report is concise and well-written, such that one can easily find the results and main conclusions of the group.
You are welcome to use an AI tool to help you write code or answers to the questions. However, please state that you do so (no points will be taken off). However, note that if you do not state that you used an AI tool, and you return code or answers that is obviously written by AI and it has not been proof-read to be relevant for the question asked, your grade will be reduced one or two levels, depending on severity. For example, if we ask you to define the assumption for a method and you return an AI paragraph that describes the importance of assumptions for reliability methods, your grade will be lowered one level. If you do this multiple times on an assignment, perhaps two levels.
Groups
You can do the assignments individually, or in groups. You can also choose your groups, but keep in mind that feedback will be provided by group, not by individual. For each assignment that is turned in, you will be asked to report your group members to Robert via email. Please include the following information in your email:
- name of group
- name of each group member
- GitHub username of each group member
- email of each group member
Unless you send a new email for the next assignment that is turned in, we will assume your group has remained unchanged.
Although the assignments are meant to be completed in class, this may not always be the case depending on your schedule. In case you have issues with your group members completing an assignment outside of class, or for any other reason, contact Robert immediately.
Resources
Online Textbook
The Risk and Reliability book will be used (there is also a link at the top right of this website). This book will be updated weekly throughout Q4. Updates will be announced in-class. Note that there is a significant overlap in the content of this book with the MUDE book.
Workbook
The Workbook is an online book (similar to the textbook above) that will be used to present various materials that don’t quite fit in the textbook, for example: exercises, workshop assignments, solutions, miscellaneous teaching material, etc. This book will be updated throughout Q4. Updates will be announced in-class..
Other Material
This unit replaced a 1st-year MSc course CIE4130 Probabilistic Design in the old curriculum. The lecture notes (14MB) can be downloaded here. In particular, the chapters on structural reliability and system reliability may be interesting.
For those who want more background in the topics of this unit, two excellent textbooks are recommended for the component and system reliability part of the course. The first has a more thorough treatment of theory (including probability theory and some multivariate probability distributions), whereas the second is a more simple and practical treatment.
Der Kiureghian, A. (2022). Structural and System Reliability. Cambridge: Cambridge University Press. doi:10.1017/9781108991889
Here is the book website. Hard copies can be borrowed from Robert and also found in the TU Delft Library.
Moss, R. E. S. (2020). Applied Civil Engineering Risk Analysis. New York: Springer International Publishing. doi:10.1007/978-3-030-22680-0
Here is the book website.
Additional reference material will be added here and organized on an as-needed basis.