This Excel VBA tool opens the google map website and search specific point-to-point route travel time, extract text from the website, and filter out travel time in numbers. 

This is another web scraping example. The tool access to Energy Supply Logistics website and extract the detailed information of each petroleum product terminal. Without the tool, one must click and copy/paste manually, which requires a lot of time. 

This Matlab tool identifies multiple data sources by their filenames in a designated folder, re-arrange data by time stamps and user-input time interval, and then calculate several performance measurements with the display of 3D bar graphs. 

This Python program is designed to 1) download a file from our internal traffic data server (SFTP server), 2) extract the tar file, 3) extract and filter a data of one day for a specific Detector by user input, 4) write a text file for the filtered data, 5) plot a time vs speed figure, 6) email the figure to multiple recipients by user input.

This PDF report was generated automatically with information of time stamp, location of incident, incident type, and travel speed on the background image of Google map. The incident data used in this example is not real, but provided as an example of illustration. The report is also emailed to a designated person. e.g. traffic operation manager, automatically.

This Matlab tool uses a self-learning algorithm to match license plates (vehicles) in different locations. The association matrix that represents the probability of character-to-character combination is continuously updated as data comes in. 

The Java program simulates vehicle flows at microscopic level. It considers drivers' perception zones, which are affected by various factors, such as vehicle speeds, accelerations, adjacent vehicles' position/speeds, obstacles, and etc. 

This Matlab tool was developed for one of my dissertation chapter.The basic idea of this study is to identify and update error distribution of multi-source raw data measurements by combining all available information. A modified Approximate Bayesian Computation method was incorporated. The estimation is conducted real-time based, and the learning process will try to keep improving the accuracy of estimation.

This Matlab tool was developed to estimate a capacity of petroleum product terminals for my project. At this moment, I can't share all the details since it's on-going research. Basically, it utilizes an image processing techniques (circle detection) and several other methods to improve the results. 

This Matlab data analysis tool was developed to simulate various scenarios for assessing impact of sampling bias, time interval, segment length, and distribution types of traffic data aggregation. In my dissertation, there are more details of the sampling bias issue, including a mathematical proof and numerical examples. 

This Matlab tool extracts detailed information of drivers behaviors, i.e. lane position, steering angle, acceleration, deceleration, and etc, from driving simulator. Students are asked to follow certain instructions to examine their reactions based on various events, like crossing pedestrians, texting, and calling. I also made an automatic generation of probability density estimate (distribution) and calculated scores for our engineering events, for fun.