TWO-DAY PRE-CONFERENCE WORKSHOP
DECEMBER 9-10, 2019
University of Delaware, Newark, Del.
Monday-Tuesday, 9 a.m.-5 p.m.
Registration fee is $715.50 before Oct. 28, $795 after
- Intensive two-day master class directly applicable to professionals working in railroad track engineering
- Technical deep dive into data mining and analytics techniques critical for today’s railroad engineers, data experts and industry leadership
- Practical case-study approach, informal workshop setting with expert faculty from UD’s Railway Engineering and Safety Program
Technology advances enable an ever-increasing collection of data. The technical challenge lies in how to best utilize the benefits and capitalize on the opportunities presented by data have having all the characteristics of big data, including volume, structure/unstructured, semi structured, among others.” The Application of Emerging Data Techniques in Railway Maintenance, a 2-day professional development course, focuses on data analysis techniques, and other “vanilla” machine learning techniques (SVM, Random Forest, Hierarchical Clustering, etc), signal processing techniques and practical case studies that are directly applicable to professionals working in Railroad Track Engineering. The interpretation of the analysis and visualization of the results will be highlighted.
Workshop attendees will be awarded 1.4 CEUs or 14 professional development hours for full participation.
The pre-conference workshop features key faculty from the University of Delaware’s Railway Engineering and Safety Program, dedicated to providing education and advanced research on railway engineering, safety, operations, and economics for the North American and international railway and transit industry.
- Nii O. Attoh-Okine, Professor of Civil Engineering, University of Delaware
- Joseph W. Palese, Senior Scientist, Railroad Research and Safety Program, University of Delaware
- Allan M. Zarembski, Professor and Director, Railroad Research and Safety Program, University of Delaware
- Click here for detailed faculty bios.
Big Data in Railroad Maintenance Planning
Hear updates from experts in academia and industry as well as network with professionals across the field.
BACKGROUND TO DATA ANALYSIS
- Different of Types of Railway Track Data
- Discrete Data
- Continuous Data (including time series data)
- Exploratory Data Analysis
- Basic Statistics
- Univariate and Multivariate Analysis (Linear and Non-Linear)
- Scatter Plots
- Correlation and Covariance
MACHINE LEARNING TECHNIQUES (I)
- Supervised Learning Techniques
- Unsupervised Learning Techniques
- Reinforcement Learning Technique’s
MACHINE LEARNING TECHNIQUES (II) [Data Mining]
- Descriptive Analysis
- Predictive Analysis
- Regression Analysis
DATA SCIENCE TECHNIQUES
- Feed forward Neural Networks
- Deep Learning and neural Networks
- Convolutional Neural Network
- Bayesian Theorem/Conditional Probability
- Bayesian Networks
- Naïve Bayes
- Gibbs Sampling/Metropolis Hasting Algorithm
- Markov Chain Monte Carlo
SIGNAL PROCESSING METHODS
- Time Series Methods
- Empirical Mode Decomposition
- Track geometry and ballast condition information
- Tie condition information
- Software Implementations – Examples Using R-studio and Ipython
- Rail defect vs. Geometry Defect Relationships
- Geometry Degradation Forecasting
- Tie failure analysis
Workshop registration includes all workshop materials, on-site workshop parking, and continental breakfast and lunch on both workshop days. Participants assume responsibility for travel arrangements and hotel accommodations and any meals not specified in the event agenda. The cost of the two-day pre-conference workshop is:
- Registration on or before October 28 — $715.50
- Registration after October 28 — $795.00
- Discounts may apply
- Click here for registration details.
For details, contact Dr. Allan M Zarembski, firstname.lastname@example.org, Professor and Director of the Railroad Engineering and Safety Program, University of Delaware.