2020 ASME-CIE Hackathon: Identifying, Extracting, Analyzing Value from Large Unstructured Data Sets in Mechanical Engineering

Virtually November 14-15, 2020
In conjunction with IMECE 2020
Sponsored by
ASME Computers & Information in Engineering Division (CIE)
ASME Manufacturing Engineering Division (MED)
ASME IMECE / Advanced Manufacturing Track (AMT) &
ASME Technical Events and Content (TEC) Sector Council

ASME Manufacturing Engineering Division (MED) Centennial Celebration Endorsed Event

Please download the ASME Hackathon 2020 flyer here and share it with your friends!

Click to Register for the Hackathon

Important Dates:

*By registering for the Student Hackathon, you agree to allow your information to be shared with other registrants and volunteer leaders for the purpose of communicating event information and intra team communication.

Introduction and Agenda

The ASME-CIE 2020 Hackathon is the first Hackathon held by ASME and is expected to become one of the signature events of the American Society of Mechanical Engineering (ASME). The first Hackathon event was held on the 2020 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference. See more detail here. This second Hackathon event is co-located with the International Mechanical Engineering Congress and Exposition (IMECE). This second Hackathon is sponsored by the ASME Technical Events and Content (TEC) Sector Council, and co-funded by the ASME Computers & Information in Engineering Division (CIE), the ASME Manufacturing Engineering (MED) Division, and the IMECE / Advanced Manufacturing Track (AMT) with the goal to build multi-stakeholder (society-university-industry) relations and impact the quality and quantity of data-skilled mechanical engineers.

Download the complete agenda here!

Download the slide for kick-off meeting here!

Date and Time (Eastern Time) Agenda
DAY 1, November 14 2:00 – 3:00 pm Online check-in and virtual platform testing
3:00 – 4:15 pm Hackathon kick-off and introduction of topic areas
4:15 – 5:15 pm Team formation and meeting with your mentors
5:00 – 8:30 pm Problem formulation and proposal submission (Tutorial and Q&A session will run in parallel from 6:30 to 7:30 pm)
8:30 – 10:30 pm Hackathon continues; pitching ideas based on the preliminary proposal/project plan (Tutorial and Q&A session will run in parallel from 9:00 to 10:00 pm)
DAY 2, November 15 10:00 – 10:15 am Day 2 kick-off and judging criteria recap
10:15 – 1:00 pm Hackathon pitch time (Tutorial and Q&A session will run in parallel from 11:00 to 12:00 am)
1:30 – 4:30 pm Hackathon pitch time (Tutorial and Q&A session will run in parallel from 3:00 to 4:00 am)
4:30 – 5:00 pm Project submission
5:15 – 7:30 pm Project presentations
7:30 – 8:00 pm Judge discussion and deliberation
8:15 – 9:00 pm Awards and closing ceremony


Note: Teams will be judged in each problem tropic area, and awardees will be selected separately.


Both students and non-students (e.g., researchers from national labs, professionals from industry, etc.) are welcome to attend the Hackathon and experience the exciting competitions. Participants can register for the event as (1) a student or (2) a non-student. Note: non-student groups include any group with a non-student member, even in the case when several members of the group are students. Per the funding objective, non-student participants are not eligible for travel awards and prizes. Each group category will be independently assessed and will receive their own rankings within the group category.


IMECE Hackathon Problem Statement

Big Data is defined as “large volumes of high velocity, complex and variable data that require advanced techniques and technologies to enable the capture, storage, distribution, management, and analysis of the information.” However, it is reported that majority of the data collected (more than 80%) is unstructured data in the form of image, video, audio, undefined text and numbers. This is true in many mechanical engineering subfields where sensors are ubiquitous and digitization is pervasive, for example, when analyzing Amazon reviews to elicit customer preferences in support of engineering design, and using images of 3D printing in support of manufacturing prognosis. While the value of unstructured data is evident by the vigor and velocity with which new tools are being created in the private sector to extract this hidden value, in mechanical engineering, the question of how to leverage the power of unstructured data to benefit product design and development, manufacturing and complex systems engineering is still yet fully answer.

The ASME-CIE Hackathon attempts to provide an open mechanism for researchers to explore new statistical and machine-learning techniques appropriate for the use of unstructured text, images, audio etc. in design, manufacturing and systems engineering, and on the other hand, to develop new educational pathways to train the next generation of data-skilled mechanical engineers. The participants will have the opportunity to learn and experience various data visualization, data mining, and machine learning methods to develop automated processes for:

All IMECE 2020 participants are welcome to attend the Hackathon. Particularly, the second 2020 ASME-CIE Hackathon problems are highly related to the following session topics. More details can be found in the Hackathon Problems.


Hackathon is a teamwork. You do not need to have all the skills – that’s what TEAMWORK is for! Please join us if you have:

Hackathon Team and Presentation

Thanks and Recognition

The following individuals have offered advice, suggestions, support, and help in different ways for the ASME-CIE Hackathon. We appreciate everyone’s commitment to this exciting event for our students and our community.


Machine Monitoring: Generating a Data-Driven Surrogate Model for Machine Damage Accumulation
  • 1st place: MGP - Sam Lim at Georgia Institute of Technology, Daniel Lee at Purdue University, and Vaishnavi Addala at MIT.
  • 2nd place: Decision Trio - Kristen Edwards, Lyle Regenwetter, and Amin Heyrani Nobari at MIT.
  • 3rd place: Utah - Keven Carlson at University of Utah, and Karl Garbrecht.
Smart Manufacturing: In-Process Data Mining for Powder-Bed Fusion Additive Manufacturing
  • 1st place: UMMS - Zhuo Wang at University of Michigan-Dearborn, and Wenhua Yang Chandan at Mississippi State University.
  • 2nd place: GRAIL - Zhibo Zhang at University at Buffalo, and Chandan Kumar Sahu at Clemson University.
  • 3rd place: XiaoXiaoSun - Yinshuang Xiao at University of Arkansas, Qingyu Xiao, and Xiaotong Sun at University of Arkansas.

Meme Challenge

2020 Mechanical Engineering Memester Extraordinaire (MEME) First Place: Zhenghui Sha at University of Arkansas.

2020 Mechanical Engineering Memester Extraordinaire (MEME) Second Place: Patrick Manser at Florida State University.

Other interesting works


Machine Monitoring

Download the complete training dataset here
Download the introduction slide for problem 1 here
Download the reference for problem 1 here
Dataset description: Data is provided for several different machines, including three Bridgeport mills, one drill press, and one lathe. Each machine has several sensors, and each of these sensors collects data such as peak velocity, RMS velocity, peak acceleration, and temperature. A damage accumulation value is also computed from these data. Data is logged approximately every 10 minutes.

Smart Manufacturing

Download the complete training dataset here
Download the introduction slide for problem 2 here
Download the references for problem 2 here
Dataset description:An experimental L-PBF build was conducted on the Additive Manufacturing Metrology Testbed (AMMT) at National Institute of Standards and Technology (NIST). The AMMT is a fully customized metrology instrument that enables flexible control and measurement of the L-PBF process. A high-speed melt pool monitoring camera was used to capture melt pool images. The galvo mirror system and the beam splitter allow the high-speed camera to observe the laser melting spot at every location the laser scans and melts material. Emitted light from the melt pool is imaged through a 850 nm (40 nm bandwidth) bandpass filter on to the camera sensor. On the AMMT, both the galvo and laser command are updated by field programmable gate array (FPGA) at 100 KHz. The digital commands are developed to specify the motion of the galvo scanner of the L-PBF system. It is transformed into a time series of scanner positions and laser power as control commands.

The dataset used in this problem is “20190711-HY-RHF” pertaining to an AM experiment performed on the AMMT by Ho Yeung on July 11, 2019. The powder is nickel superalloy 625(IN625). The experiment uses continuous-varying laser power to scan multiple rectangle single layer parts on a bare build plate (i.e. no metal powder). Each part is a 3 mm x 2 mm rectangle with same scan geometry, but different laser power transient profiles that range from 145 W to 195 W. The 120 pixel × 120 pixel (at 8 μm/pixel) in-situ melt-pool images were captured at 20000 frames per second.


We've compiled a list of resources, including tutorials, ML tools, libraries, and ideas just for you.

Machine Learning Developement Tools
1. PyTorch 2. TensorFlow
3. scikit-learn 4. R
5. MATLAB 6. Keras
7. Theano 8. Caffe
File Sharing & Cloud Stores
1. Dropbox 2. Google Drive
3. Box 4. Amazon S3
5. FileZilla
Project Management
1. Github 2. Trello
3. ASANA 4. Slack
Virtual Communication
1. Skype 2. Cisco WebEx
3. BlueJeans 4. Zoom
Text Editors
1. Sublime Text 2. Atom
3. Notepad++ 4. TextWrangler


Dr. Christopher McComb (Pennsylvania State University)

Assistant Professor

Dr. Christopher McComb is an Assistant Professor of Engineering Design and Mechanical Engineering at Penn State, and the director of the Technology and Human Research in Engineering Design (THRED) Group. In his research, he draws on perspectives from engineering, design, and psychology to study how humans act while solving problems. Within this regime, he explores the spectrum between human teams and computational agent teams, creating advanced design algorithms and cognitive support tools. In addition to design research, he is interested in machine learning, STEM education, design automation, and design for the developing world. https://cmccomb.com/

Dr. Zhenghui Sha (University of Arkansas)

Assistant Professor

Dr. Zhenghui Sha is an Assistant Professor of Mechanical Engineering at University of Arkansas. He is interested in decision-making in engineering systems, design complex systems and complex networks, game theory and behavioral experimentation, decision-based enterprise-driven design, artificial intelligence in design, crowd-sourced design, and open-source product design and development. https://sidilab.net/sidi-people/faculty/

Dr. Faez Ahmed (Massachusetts Institute of Technology)

Assistant Professor

Dr. Faez Ahmed is an Assistant Professor of Mechanical Engineering at MIT. He is interested in AI-driven design problems. https://faezahmed.com/about/

Dr. Yan Lu (National Institute of Standards and Technology)

Senior Research Scientist

Dr. Yan Lu is a member of the System Integration Division. Her research interests at NIST include smart manufacturing system reference architecture design, production operation and optimization, and additive manufacturing modeling and design optimization. https://www.nist.gov/people/yan-lu

Dehao Liu (George Institute of Technology)

PhD Candidate

Dehao Liu is a Ph.D. student in the School of Mechanical Engineering at Georgia Institute of Technology, advised by Dr. Yan Wang in the Multi-Scale Systems Engineering Research Group. His research interest is to investigate the Process-Structure-Property relationship during metal additive manufacturing (MAM) process using multiphysics simulation and physics-constrained machine learning. https://dehaoliu.github.io/

Dr. Anh Tran (Sandia National Laboratories)

Senior Member of Technical Staff

Dr. Anh Tran is a senior member of technical staff at Optimization and Uncertainty Quantification, Sandia National Laboratories. His research interests are broad but mainly in the context of computational materials science, uncertainty quantification, optimization (inverse problems), and machine learning (including deep/shallow learning) applications.

Dr. Zhuo Yang (National Institute of Standards and Technology)

Postdoctoral Researcher

Dr. Zhuo Yang is a postdoctoral researcher in System Integration Division at National Institute of Standards and Technology.

Dr. Dazhong Wu (University of Central Florida)

Assistant Professor

Dr. Dazhong Wu is an assistant professor in mechanical and aerospace engineering at University of Central Florida. https://mae.ucf.edu/person/dazhong-wu/

Dr. Binyang Song (Pennsylvania State University)

Postdoctoral Researcher

Dr. Binyang Song is a postdoctoral researcher in School of Engineering Design at Pennsylvania State University.


International Mechanical Engineering Congress and Exposition (IMECE)

Follow us!

A fork of BigDataForSanDiego