Transportation systems underpin nearly every aspect of modern society, and the engineer who plans, designs, and operates them works with limited data, competing objectives, and consequences that are measured in safety, delay, and cost. This course builds the analytical toolkit that work requires, moving from regional travel forecasting down to the geometry of a single curve and the timing of a single signalized intersection. Methods are introduced in a consistent sequence: the physical or behavioral phenomenon first, then the model that represents it, then the calculation the engineer actually performs, then the conditions under which the model fails.
Learning objectives
By the end of the course, students will be able to:
Apply the four-step travel demand model to generate, distribute, split by mode, and assign trips on a network.
Analyze a traffic stream using flow, speed, and density relationships, shockwave theory, and deterministic and stochastic queuing models.
Compute road vehicle performance quantities, including braking distance and grade and rolling resistance, and use them to determine stopping sight distance.
Design vertical and horizontal alignments that satisfy sight distance and related geometric controls, and size a pavement structure for a projected traffic loading.
Determine signal timing at an intersection and evaluate intersection and freeway level of service.
Carry an analysis from data through model to a defensible engineering recommendation, and state the assumptions and limitations under which that recommendation holds.
Course outline
Module
Weeks
Topics
1. Transportation Planning
1–4
Travel demand forecasting and trip generation; trip distribution with the gravity model; mode choice with utility and logit models; traffic assignment (networks, shortest paths, link performance, user equilibrium, and system optimum); in-class practice session
2. Traffic Analysis
5–7
Traffic flow fundamentals; shockwave analysis; queuing theory (D/D/1, M/M/1, M/D/1, M/M/N); in-class practice session; mid-term exam
3. Vehicle Performance and Geometric Design
9–11
Road vehicle performance; stopping sight distance; road geometric design; horizontal sight distance and cross section; pavement system design; in-class practice session
4. Signal Control and Level of Service
12–15
Intersection control and signal fundamentals; delay and level of service at signalized intersections; signal coordination; freeway capacity and level of service; in-class practice session
Assessment: Assignments 40% (4 × 10%); in-class practice sessions 10% (4 × 2.5%); mid-term exam 25%; final exam 25%.
This course on Deep Learning for Civil Computing is crafted to bridge the significant gap between advanced computational methods and their practical applications in civil engineering. The curriculum offers an introductory exploration of essential machine learning principles and a range of deep learning models, including neural networks, reinforcement learning, and generative systems. With a strong emphasis on practical application, the course delves into how these technologies can be effectively utilized in critical civil engineering domains such as traffic management, infrastructure resilience, and sustainable development.
Learning objectives
By the end of the course, students will:
Gain a strong grasp of core deep learning concepts, including key architectures such as fully connected neural networks, convolutional neural networks, and recurrent neural networks.
Apply these techniques to analyze large-scale civil engineering datasets.
Design and implement machine learning models to solve practical engineering problems, with skills in model selection, training, validation, and interpretation.
Evaluate the limitations and risks of the models they build, including error asymmetry in safety-critical prediction, bias and distribution shift in training data, and the data governance implications of deployed sensing systems.
Course outline
Weeks
Topics
Hands-on practice
1–2
Course introduction; Python basics (Jupyter notebooks, Google Colab, syntax, NumPy) and advanced Python skills
Electric vehicles are entering the transportation system faster than the infrastructure that serves them, and the resulting problems cut across vehicle technology, travel behavior, the electrical grid, and public policy at once. This course treats that system as a whole and builds the quantitative tools to reason about it. Methods are introduced in a consistent sequence: the phenomenon first, then the model that represents it, then the code that runs it, then an honest evaluation of what the model supports and where its limits lie. Python is taught from zero, so the course is open to students from any background, and every technical week ends with a hands-on assignment on real electric vehicle data.
Learning objectives
By the end of the course, students will be able to:
Identify the components of an electrified transportation system and their interdependencies, and analyze current adoption and infrastructure trends from data.
Apply travel behavior and demand modeling concepts, and interpret electric vehicle energy, range, and charging profiles.
Build, calibrate, and interrogate models of electrified transportation in Python, including agent-based simulation, mathematical optimization, and integrated multimodal system models.
Formulate a transportation problem as a supervised, unsupervised, or reinforcement learning task, engineer features for it, and train and evaluate a model honestly against an appropriate baseline.
Map an electrified transportation system, identify its feedback loops, trade-offs, and stakeholders, and evaluate a real policy against equity, environmental justice, and grid resilience criteria.
Carry an analysis from data through model to a defensible recommendation, and communicate it to a technical and a non-technical audience.
Course outline
Week
Module
Topic
1
1. Foundations
Introduction and systems overview
2
1. Foundations
Travel behavior and demand modeling
3
1. Foundations
EV characteristics and charging behavior
4
2. Modeling
Python foundations for electrified transportation modeling
5
2. Modeling
Modeling charging demand and behavior (agent-based modeling in Mesa)
6
2. Modeling
Planning electric bus deployment (integer programming in PuLP)
7
2. Modeling
Integrated electrified mobility modeling
8
3. AI and ML
Fundamentals of AI and ML for transportation
9
3. AI and ML
Data analytics and feature engineering
10
3. AI and ML
Demand and charging prediction with ML (neural networks, RNNs, LSTMs)
11
3. AI and ML
AI for system optimization (reinforcement learning, Q-learning)
12
3. AI and ML
Multi-agent systems for complex problems
13
4. Integration
Systems thinking and policy-aware design
14
4. Integration
Policy brief and AI ethics
15
4. Integration
Final project
Assessment: Weekly assignments 70% (14 × 5%); final project 30%.
Student Feedback
Official University of Utah student course feedback for CVEEN 5565 (taught as CVEEN 5920, Machine Learning for Civil Computing).
5.0 / 5
Instructor recommendation
Spring 2025 and Spring 2026
4.8 / 5
Course recommendation
Spring 2025 and Spring 2026
6.0 / 6
Instructor effectiveness
Spring 2025
100%
Learning objectives clear
Spring 2025
Spring 2025: 10 of 14 students responded; subject means 4.28 (instructor recommendation) and 4.23 (course recommendation). Spring 2026: 5 of 10 students responded.
“The instructor is incredibly knowledgeable, and I always felt comfortable asking questions. He is resourceful in his responses, providing valuable insights. The syllabus was excellent and highly practical.”
Student, Spring 2025
“Very clear, helpful, knowledgeable. Draws on the whiteboard to explain complex concepts. Often able to break down a complex topic into understandable bits. Provided and explained code so that we could experiment with it on our own at our own pace.”
Student, Spring 2025
“The course is well-structured for students who want to learn machine learning and apply this technology to their own fields in civil and environmental engineering. The instructor is very knowledgeable and good at teaching.”
Student, Spring 2026
“It was an excellent course. I really liked it and it was worth all the effort and knowledge gained from it.”