Group Posts

In his dissertation, John E. Fontecha studies the mathematical design of policies aimed at increasing the adoption of specific behaviors/products. His dissertation comprises three main chapters. First, a systematic review of the research literature about incentive programs and policies that promote pro-environmental behavior and technology adoption among individual consumers. Second, a conceptual basis and methodological advancements for maximizing social influence through reach rather than diffusion as well as their application to energy-saving behavior are presented. Finally, a study aimed at bridging the divide between theory and practice in modeling collective behavior by accounting for time passage in threshold modeling and providing a mechanism to update the population's individuals about the actual fraction of adopters.

On April 28, 2021, Dr. Merve Bodur will present their research on two problems related routing and wavelength assignment in telecommunication networks. Each problem is modeled as a two-stage stochastic integer program and solved through a decomposition-based approach. Dr. Bodur is an Assistant Professor in the Department of Mechanical and Industrial Engineering at the University of Toronto. The seminar will begin at 2:00pm EST. Click here to join the meeting.

On April 7, 2021, Dr. Oleg Prokopyev will present his research on bilevel mixed integer optimization problems. In particular, this work focuses on a Stackelberg game where the follower is not completely rational, and may not solve their problem to optimality. Dr. Prokopyev is a Professor in the Department of Industrial Engineering at the University of Pittsburgh. The seminar will begin at 2:00pm EST. Click here to join the meeting.

We are happy to announce our GAMMA seminar series speakers for the Fall 2020 semester! Each seminar will feature current research from a graduate student, professor, or industry expert from outstide of the University at Buffalo. Click here to join each meeting.

October

• Mariana Escallón, Ph.D. Student, Northwestern University, October 1, 3:00 pm EST
• Zahed Shahmoradi, Ph.D. Student, University of Houston, October 8, 3:00 pm EST
• Robert Curry, Assistant Professor, United State Naval Academy, October 15, 3:00 pm EST
• Jamie González, Postdoctoral Fellow, HEC Monrtéal, October 22, 3:00 pm EST
• Anahita Khojandi, Associate Professor, University of Tennessee, October 29, 3:00 pm EST

November

• Joshua Margolis, Ph.D. Student, Clemson University, November 5, 3:00 pm EST
• Kelly Sullyvan, Associate Professor, University of Arkansas, November 19, 3:00 pm EST

December

• Daniel Duque, Research Scientist, Google, December 3, 3:00 pm EST
• Chrysafis Vogiatzis, Assistant Professor, University of Illinois at Urbana-Champaign, December 10, 3:00 pm EST

Since defending his dissertation last May, Ningji Wei has accepted a position as a postdoctoral reserach fellow at Heinz College at Carnegie Mellon University. He is collaborating with Dr. Peter Zhang where his current research involves affinely adjustable policies of robust optimization models.

In his dissertation, Cai Gao describes two important UAV routing problems: risk and reward asset routing problem (R2ARP) which and the close-enough traveling salesmen problem (CETSP). The solution approaches for these problems featured a mix of exact cutting plane approaches for MILP and MINLP, as well as an efficient Lin-Kernighan heuristic for larger instances.

Upon being conferred his M.S. with concentration in Operations Research from UB's ISE department, Luca is moving to the University of Pittsburgh to pursue a Ph.D. in Industrial Engineering. Here, he will be advised by Dr. Oleg Prokopyev, an academic "brother" of Dr. Walteros. Dr. Prokopyev leads work in interdiction optimization, with particular interest in applications addressing challenges such as incomplete information or sequential games. Good luck, Luca!

This award is presented to a graduate student who has shown their willingness to support others and that has dedicated significant time, effort and leadership to serve the university, school and/or department. This is an acknowledgement for John’s leadership and dedication to the ISE Graduate Student Association (ISE-GSA) and the Group for Applied Mathematical Modeling and Analytics (GAMMA) as well as other small acts of service that impact positively the campus.

This award is presented to a graduate student who has shown their willingness to support others and that has dedicated significant time, effort and leadership to serve the university, school and/or department. This is an acknowledgement for Rishabh’s leadership and dedication to the ISE department through the several activities that he conducts

This award is presented to a graduate student for demonstrating exceptional qualities as a teaching assistant or instructor through effective course material delivery, enthusiasm, responsibility, and professional demeanor. This is an acknowledgement for Rishabh and John’s qualities as instructors for IE320 Engineering Economy and IE306 Statistics for Engineers, respectively.

In this seminar, Dr. Andrés D. González (Assistant Professor at School of Industrial and Systems Engineering at University of Oklahoma) presents his research on the resilience of critical interdependent infrastructure networks. He introduces a set of mathematical models centered on the Interdependent Network Design Problem (INDP), concerned with defining the minimum-cost reconstruction strategy of a partially destroyed system of interdependent networks. He presents several and diverse applications of the INDP.

In this seminar, Dr. Ethan Malinowski (UB ISE, Ph.D.) presents an overview of personal industry experience with expectations and advice to current graduate students. He talks about topics ranging from work structure and environments to potential roles students could expect to occupy after completing their time in graduate school. He also takes the time to describe some unique applications of ORMS that he has done on general business analytics, transportation optimization in the context of middle and last mile transportation in an Ecommerce business, and private aviation sector.

In this workshop, John Fontecha shows some quick and nice tricks in Excel for basic formulation and data visualization.

Interdiction games are commonly used to model scenarios where two actors with conflicting objectives both aim to operate optimally. Dr. Ningji Wei defined an abstract game called the Boolean interdiction that captures the essence of an important subclass of such interdiction games and developed several theoretical tools to solve them. Ningji works under the supervision of Dr. Jose Walteros and he is broadly interested in combinatorial optimization, integer programming, graph interdiction, and stochastic processes.

Three minutes before his defense, Ningji Wei and his co-authors Jose L. Walteros and Foad Mahdavi Pajouh got their paper accepted in the Informs Journal On Computing (IJOC). Congratulations!

In this seminar, Daniela Angulo (instructor at Industrial Engineering at Universidad de los Andes) presents her research on the risk assessment of diseases associated to HPV vaccination and cost-effective evaluation of public policies for cervical cancer in Colombia.

In this interview, GAMMA gave us the wonderful opportunity to ask Dr. Walteros some difficult questions, just like the ones on his exams. Students could submit a question through the google form before the interview or through the chat Zoom chat during it. We got to know his second last name, that he is an Industrial Engineer since even before he was born, and why his twitter is “incorrect”.

In this workshop, Demetrios Papazaharias shows how to use and implement column generation to solve the cutting stock problem utilizing Gurobi. Column Generation is an efficient algorithm for solving linear programs with a huge number of variables, as many that it could be impossible to store them all in a computer. Materials for this workshop can be found here.

UB ISE highlights GAMMA. This highlight is the result of the efforts that GAMMA has been making in two fronts: 1) by extending bridges to the research conducted in other places around the world through the GAMMA seminars in which UB students and faculty gather to listen the presentation of our invited speakers; and 2) by sharing knowledge and skills through the GAMMA hands-on in which students can learn from simple tricks in Excel to High Performance Computing tools.

In this seminar, Dr. Jorge Sefair (Assistant Professor at School of Computing, Informatics, and Decision Systems Engineering at Arizona State University) presents his research on finding optimal coordinated routes and schedules for multiple vehicles traveling in a network. He discusses the hardness of this problem and presents an exact formulation for its solution. He also devises an exact solution algorithm based on a network decomposition that exploits the sparsity of the optimal solution and illustrates the performance of his methods on real and randomly generated networks.

In this seminar, Hamidreza Validi (Ph.D. student at Industrial Engineering and Management School at Oklahoma State University) presents his research on models for imposing contiguity when designing political districts. He first reviews two existing models for imposing contiguity and then proposes two new ones, and analytically compares them in terms of their strength and size. He conducts an extensive set of numerical experiments to evaluate their performance.

In this seminar, Daniel Otero (Ph.D. student at Industrial and Operations Engineering at University of Michigan) presents his research on Stochastic Dynamic Programming applied in the airline and healthcare sectors. He provides insights regarding the models' structural properties and how to overcome the drawbacks associated with the countless parametrization approaches and the increase of the problem size to deliver optimal strategies.

In this workshop sponsored for ISE Department, Arpit Rana (M.S., UB ISE Graduate and Analytics Engineer at ACV Auctions) presents the second session of Data Visualization in Tableau.

In this workshop sponsored for ISE Department, Arpit Rana (M.S., UB ISE Graduate and Analytics Engineer at ACV Auctions) presents the basics for Data Visualization in Tableau. These two-sessions of GAMMA hands-on workshop and material are for anyone who is interested in working with Tableau to produce high quality and interactive data visualizations.

In this seminar, Jorge Huertas (Instructor at Industrial Engineering at Universidad de Los Andes) presents his research on evacuations dynamics. He presents three projects on a real case study on the analysis of evacuation dynamics in a university campus with special topography.

We are excited to highlight the work of Dr. Maria Torres, Applied mathematical model for rainwater harvesting using Green Infrastructure, has been published in Environmental Modelling & Software.

Cai Gao did an internship in Los Alamos National Lab (LANL), New Mexico from 06/2019 to 12/2019. He worked on a project of Developing a Novel Interpretable Machine Learning Model with LANL scientist Dr. Hassan Hijazi. While classical machine learning algorithms heavily rely on unconstrained optimization techniques, this internship project mainly focuses on developing a constrained-based approach, called Hypersphere Classifier, to help guide and accelerate machine learning process. Our preliminary results demonstrate, in the case of 2-D object shape recognition, i.e., handwritten digit classification, this new classifier can reach 92.0% accuracy while visualizing the whole learning process.

The paper On the Distance Between Random Events on a Network by Ningji Wei, Dr. Jose Walteros, and Dr. Rajan Batta has been published in the journal Networks. In the paper, they studied several statistical properties regarding the distance between events that occur on random locations along the edges of a given network. The results can be applied to network design and analysis, facility location selection, and emergency response systems design.

In a competition with over 150 entries, Luca Wrabetz and Dr. Walteros entered their research project “Minimum Cost Set-Cover Blocker”. The work explores the computational aspects of blocking set-covers in leader-follower type scenarios, which has applications in a wide range of fields such as defense, wireless sensor networks, and bioinformatics. The poster was accepted as a finalist in the competition, and following a presentation to the judges, was awarded 3^rd place.

Several members of GAMMA presented their work at the INFORMS 2019 Annual Meeting. Luca Wrabetz competed in the student poster competition and won third place for his poster “Minimum Cost Set-Cover Blocker Problem”. Dr. Walteros was an honorable mention at the JFIG Paper Competition for his paper “Why is maximum clique often easy in practice?”. This paper was recently submitted and accepted in Operations Research (link). In addition, Ningji Wei presented his working paper “A Resiliency Analysis of Information Distribution Policies over Mobile Ad Hoc Networks”. Demetrios Papazaharias presented his work titled “Extended Formulations for Simple Graph Partitioning on Sparse Graphs”.