Collaborative Work Published by Journal of Mechanical Design

Our paper titled “Integrating Sequence Learning and Game Theory to Predict Design Decisions under Competition”, is recently accepted by the Journal of Mechanical Design, Transactions of the ASME.

This paper presents an approach for predicting designers’ future search behaviors in a sequential design process under an unknown objective function by combining sequence learning with game theory. While the majority of existing studies focus on analyzing sequential design decisions from the descriptive and prescriptive point of view, this study is motivated to develop a predictive framework. We use data containing designers’ actual sequential search decisions under competition collected from a black-box function optimization game developed previously. We integrate the long short-term memory networks with the Delta method to predict the next sampling point with a distribution, and combine this model with a non-cooperative game to predict whether a designer will stop searching the design space or not based on their belief of the opponent’s best design. In the function optimization game, the proposed model accurately predicts 82% of the next design variable values and 92% of the next function values in the test data with an upper and lower bound, suggesting that a long short-term memory network can effectively predict the next design decisions based on their past decisions. Further, the game-theoretic model predicts that 60.8% of the participants stop searching for designs sooner than they actually do while accurately predicting when the remaining 39.2% of the participants stop. These results suggest that a majority of the designers show a strong tendency to overestimate their opponents’ performance, leading them to spend more on searching for better designs than they would have, had they known their opponents’ actual performance.

This study was performed in collaboration with Dr. ‪Alparslan Emrah Bayrak from Stevens Institute of Technology.

New Semester Starts in A Virtual World

Fall 2020 will be for sure an unforgettable semester. SiDi has been primarily operated virtually since March. Without doubt, SiDi kicked off the new semester with a virtual lab meeting and it was a nice moment to capture. We welcomed the new PhD student Sumaiya Sultana Tanu, who recently graduated to the University of Denver with double MS degrees in Mechanical Engineering and Information and Communication Technology. It is always exciting to see new SiDi members at the beginning of a semester! Sumaiya will be working on a research project focusing on Design for Market Systems with Fairness. This project is under a recently NSF-funded EPSCoR initiative on Data Analytics that are Robust and Trusted (DART): From Smart Curation to Socially Aware Decision Making. We very much look forward to collaborating with our collaborators in Computer Engineering and Marketing Science to conduct this interdisciplinary research project which can potentially generate broad and high impacts to the diverse society we are living in today.

SiDi will Participate in the NSF’s Established Program to Stimulate Competitive Research (EPSCoR)

University of Arkansas News: “A $20 million grant to the Arkansas Economic Development Commission’s Division of Science and Technology, in partnership with the University of Arkansas and eight other colleges and universities, will build a high-performance computing network for data analytics and bring together scientists and engineers from across the state to focus on data analytics. The five-year grant, from the National Science Foundation’s Established Program to Stimulate Competitive Research (EPSCoR), will be matched by $4 million from the state. With it, researchers will establish a program titled “Data Analytics That Are Robust and Trusted,” or DART. ”

SiDi had the opportunity to participate in and contribute to the proposal development. More than 30 scientists from the U of A will focus on three major areas of data science: 1) managing data sets that are too big or complex for traditional hardware and software, 2) insuring security and privacy of that data, and 3) developing machine learning and artificial intelligence models that better inform decisions and give insight into the underlying process. In this program, we will work with other research scientists to focus on one project that aims at developing new approaches to the design of market systems with a particular emphasis on the design of marketing strategies with fairness. See the full media coverage HERE.

SiDi Uses 3D Printer to Produce Protective Masks for Rehab Clinic

We recently joined a design team led by engineering professors Raj Rao, Wenchao Zhou and Zhenghui Sha at the University of Arkansas to use 3D printers to produce protective masks in response to the COVID-19 crisis. The team printed and delivered 50 protective masks to be used by a Northwest Arkansas rehabilitation clinic. The Ph.D. student Laxmi Poudel from SiDi Lab participated in this initiative and helped print about 35 masks. Prof. Zhou has produced two short videos, one showing how the masks are made and another showing clinicians how to assemble the masks. See the full story covered by UARK NEWS.


SiDi Receives Grant from the National Science Foundation to Conduct Collaborative Research

The University of Arkansas, Northwestern University and Ford Motor Company will form a joint force to conduct collaborative research on customer preferences modeling in product design and development. The objective is to investigate what product customers consider and what they eventually purchase using a hierarchical, multidimensional network-based design approach. Motivated by the need to model socio-technical interactions in engineering design, this research combines design theory with network science to explore three interrelated topics: 1) two-stage multidimensional network models for customer preference modeling that consider product associations and social influence; 2) dynamic network models for predicting the impact of multi-competitor strategic decisions, and 3) knowledge transfer to demonstrate generalizability and creation of shared data resources to benefit research community. See the NSF website for more details about this project.