Wen Feng Resume

1. Wen Feng
6902 Lakeview Blvd Apt. 20305 | Westland, MI 48185 | [phone] | [email protected]

Details-Focused M.S. in Industrial Engineering Candidate Seeking Growth-Oriented Role

DATA SCIENTIST | DATA MINING | BIG DATA | STATISTICS | ANALYTICS | OPERATION RESEARCH | PROGRAMMING
QUALITY ENGINEERING | QUALITY CONTROL | SIX SIGMA GREEN BELT | SYSTEMS DESIGN | LEAN MANUFACTURING | MARKETING

 Excellent learner who are interested in learning new tools and techniques.
 Out-of-the-box thinker who exhibits superior attention-to-detail and sharp skills to achieve goals.
 Ambitious self-starter who plans, prioritizes, and completes critical tasks within deadline-driven environments.

EDUCATION & TECHNICAL SUMMARY

Master of Science in Industrial Engineering (4.0 GPA, Graduated Apr. 2016)
UNIVERSITY OF MICHIGAN – DEARBORN, DEARBORN, MI

Industrial Engineering Studies – Degree Exchange (3.90 GPA, 2013 – 2014)
UNIVERSITY OF MICHIGAN – DEARBORN, DEARBORN, MI

Bachelor of Science in Industrial Engineering (3.90 GPA, Graduated 2013)
XI’AN JIAOTONG UNIVERSITY, XI’AN, CHINA

C | R | SAS | AMPL | WEKA | SPSS | MATLAB | Minitab | Data Analyst
SAP| SIMO | ARENA | Auto CAD | Six Sigma Green Belt Certificated (Mar. 2015 to Apr. 2018, License NO. 2227-5166)

COURSE PROJECTS

Letter Image Recognition (Fall 2015)
 16 numerical attributes of 20000 letter images were used as input factors to predict the belonging of the letter.
 Two methods (Neural Network and Support Vector Machines) with different parameters were applied to achieve the
pattern recognition goal.
 Performance level (accuracy and efficiency) were tested and SVM system who gave an accuracy of 97.46% and efficiency
of 2.86 seconds was chosen as the final model.

Reliability Analysis of Vehicle Side Impact Crashworthiness (Winter 2015)
 The crashworthiness of vehicle side impact was studied in two design conditions, baseline design and reliability-based
optimal design (RBDO) with 90 percent reliability.
 Comprehensively applied Monte Carlo simulation and Taylor Expansion methods to simulate the reliability of the actual
performance system.
 Effectively valuated the two methods in terms of their accuracy and efficiency. The MCS with large sample size is more
accurate but less efficient.

Model Quality Improvement Using Multivariate Statistical Analysis (Fall 2014)
 Effectively designed and built two different prediction models (Linear regression model and Sparse Coding Neural
Network model) using 37 input factors of car infrastructure to predict the intrusion responses at 7 different locations.
 Trained the two designs with 300 sets of designs and validated using 80 sets of designs. Based on the system
performance analysis, an ensemble system was proposed to make the intrusion prediction.
 The results of the ensemble system (MAE = 6.363) were much better than the results of any individual system (MAE =
10.703, 10.593, and 10.643).

2. Wen Feng
Resume – Page Two | [phone] | [email protected]

PROFESSIONAL SYNOPSIS

UNIVERSITY OF MICHIGAN – DEARBORN, DEARBORN, MI MAR. 2014 – APR. 2016

Research Assistant (Mar. 2014 – APR. 2016)

Assessment of Social Preference in Automotive Market using Generalized Multinomial Logistic Regression (submitted to Journal
of Marketing) (Mar. 2014 – Dec. 2015)
 Came up with a generalized multinomial logistic regression model to assess the social welfare function of each vehicle in
North America market by utilizing the knowledge of B-spline, discrete choice model and maximum likelihood.
 Built a logit model on predicating of the market share of the automobiles in North America market using the core
attributes selected by the likelihood ratio test.
 Obtained the utility plot as well as the relative importance of each attribute. In addition, explained all these results
including critical point in the utility plot and weights from the perspective of consumer.

V2V Based Online Detection, Modeling, and Prediction of Aggressive Driving Behavior (master thesis) (Sep. 2015 – APR. 2016)
 Developed a real-time V2V-based safety assistance tool to detect potential aggressive driver surrounding the subject
vehicle.
 Constructed a scenario-specific aggressive driving model that characterize aggressive drivers’ homogeneous behavior
using UMTRI datasets based on decision tree.
 Developed an online learning method using Bayesian approach to update these homogeneous decision trees to
characterize individual behavior of a detected vehicle.
 Developed an intelligent online prediction mechanism that can accurately predict the potential risky actions by the
aggressive drivers in the surroundings.

Teaching Assistant (Sep. 2014 – Apr. 2015)

Models of Operation Research (Graduate)
Introduction to Operation Research (Undergraduate)
Six Sigma and Statistical Methods for Process Improvement (Undergraduate)
 Planned and presented resourceful after-class tutoring sessions for these three large-scale classes.
 Responsible for their homework and exam grading as well as experiments guiding.

CEDAR POINT AMUSEMENT PARK, SANDUSKY, OH SEP. 2013 – APR. 2014

Senior Design Project | Team Member | Presenter
 Analyzed and reported the current issues in the maintenance shop, including layout, material flow, storage, etc.
 Reconfigured the layout and process in the maintenance shop utilizing knowledge of facility design and human factors,
improved the working efficiency and staffs’ safety.
 Analyzed the current inventory system and suggested an online integrated information system (SAP) to improve the
inventory management.

SHAANXI AUTOMOBILE GROUP, XI’AN, CHINA JUN. 2013 – AUG. 2013

Industrial Engineering Intern
 Played a vital role in directing the collection of working time and idle time for each worker in the heavy duty truck chassis
assembly line. Conducted motion study and time study for each assembly task.
 Reconfigured the material location and standard operation for each task and rescheduled the worker at each assembly
station based on the motion study and time study. Increased the efficiency of the assembly line by 8.2%.