This was a course project carried out at the Center for Applied Nanobioscience – Biodesign Institute at Arizona State University. It involved applying the designed experiments methodology to study the Jenoptik machine for channel replication of a microstructure on a polymer using the hot embossing process. It had been observed that all the channels cannot be used due to defects occurring during the hot embossing process. This study characterized the process by identifying the most influential factors and their levels so as to minimize the defects in the output.
Course: IEE 572 – Design of Engineering Experiments
Software used: Design Expert
We attempted to create a Management Information System (MIS) in the context of environmental pollution. The Environmental Pollution Tracker collates information regarding pollution, manages the data and finally analyzes and presents the results in an easy to understand form. The three main objectives of this project were to
- Manage a large database
- Explore charting and graphing tools
- Implementation of a statistical model
We used Visual Basic, Microsoft Access and SQL to complete the project requirements. The project included identifying a suitable database, creating the database structure, developing an application to create graphs and charts to summarize the data and finally creating a time series model to forecast future pollutant levels based on past data.
Course: IEE 505 – Information Systems Engineering
Software used: Visual Basic
The IEE 582 course taught different response surface designs. The course used a simulated chemical process with three responses for the project. The aim was to optimize the responses given the five independent variables in the process. We used fractional factorial designs and central composite designs in a sequential series of experiments, minimizing the number of runs while achieving the goal of the process.
Course: IEE 582 – Response Surfaces / Process Optimization
Software used: Design Expert
This was a term project for the IEE 520: Statistical Learning for Data Mining class. The purpose of the project was to implement various data mining techniques learnt during the semester. It involved identifying the correct method / technique to classify an unbalanced dataset.
Course: IEE 520 – Statistical Learning for Data Mining
Software used: Weka
A number of experiments were run at the Applied Nanobioscience Center at the Biodesign Institute – Arizona State University to study a microfluidic mixing process. Since the data was already collected and available, regression analysis was used to analyze this dataset. The main objective of the study was to identify the important variables and thereby understand the system better. The system was a prototype hence it was essential that the process is understood better before trying to optimize the process. The results of the analysis helped shortlist the important variables and decide on the future experiments to be run.