Skip to content

Prizes and Competitions

UPS George D. Smith Prize

The George D. Smith Prize is aimed at strengthening ties between academia and industry by rewarding institutions of higher education for effective and innovative preparation of students to be good practitioners of operations research. The Prize is generously underwritten by UPS. Awarded for the first time in 2012, past winners are Carnegie Mellon University, H. John Heinz III College, Sauder School of Business, University of British Columbia – Center for Operations Excellence, MIT Leaders for Global Operations, Naval Postgraduate School, and Tauber Institute for Global Operations at University of Michigan.

The teams will present their work to the judges on Sunday, April 3.

The Smith Prize winner will be announced at the Edelman Gala on Monday, April 4. The 2022 winner will give their presentation on Tuesday, April 5 in the INFORMS Prizes & Special Sessions Track.

Daniel H. Wagner Prize for Excellence in Operations Research

The Daniel H. Wagner competition is held each fall at the INFORMS Annual Meeting. The 2021 Wagner Prize Reprise will take place on Monday, April 4 in the INFORMS Prizes & Special Sessions Track.

The 2021 Wagner Winner is a team from the the Wharton School, University of Pennsylvania and USC Marshall School of Business for their work, “Interpretable OR for High-Stakes Decisions: Designing the Greek COVID-19 Testing System.” It was presented by Hamsa Bastani from the Wharton School, University of Pennsylvania and Kimon Drakopoulos and Vishal Gupta from USC Marshall School of Business.

This prize emphasizes the quality and coherence of the analysis used in practice. Dr. Wagner strove for strong mathematics applied to practical problems, supported by clear and intelligible writing. The Wagner Prize recognizes those principles by emphasizing good writing, strong analytical content and verifiable practice successes. The competition is held and the winner is announced at the INFORMS Annual Meeting in the fall.

Past awardees include practitioners and researchers from Lehigh University and the Pennsylvania Department of Corrections, The Forestry Research Institute of Sweden, CDC, Ford, U.S. Coast Guard, Intel, IBM T. J. Watson Research, Schneider National, Boston University, University of Florida, and others.

Innovative Applications in Analytics Award (IAAA)

Brought to you by the Analytics Society of INFORMS, Kinaxis, and Adelphi University

The purpose this award is to recognize the creative and unique application of a combination of analytical techniques in a new area. The prize promotes the awareness and value of the creative combination of analytics techniques in unusual applications to provide insights and business value. The Analytics Section leadership would like to cordially thank all the members of the judging committee for their hard work in selecting these finalists.

The IAAA Finalists will present at the conference on Tuesday, April 5 in the INFORMS Prizes & Special Sessions Track from 9:10am-12:20pm EST. Judges will then review the rank the finalists from 12:30-1:30pm, after which the winner will be announced.

A Data-driven Optimization Approach to Solve the E-commerce Packaging Problem


Sharvendu Bhushan, Flipkart Internet Private Limited
Himanshu Gupta, Flipkart Internet Private Limited
Chandrasekhar K, Flipkart Internet Private Limited
Shanthan Kandula, Indian Institute of Management Ahmedabad
Srikumar Krishnamoorthy, Indian Institute of Management Ahmedabad
Rohan Nanaware, Flipkart Internet Private Limited
Primary Presenter: Debjit Roy, Indian Institute of Management Ahmedabad
Sandeep Sangwan, Flipkart Internet Private Limited

Determining the optimal packaging box assortment is crucial for e-commerce platforms. Typically, an assortment of fewer than 50 boxes is used to cover a few hundred thousand SKUs. Since the number of boxes is far less than the number of SKUs, boxes rarely fit the SKUs accurately, resulting in empty space. Typically, one-third volume of a typical package comprises air and filler material. The inefficient usage of space significantly increases operational costs and carbon footprint. In this project, we develop approaches to determine the optimal packaging box assortment in two phases. First, we employ a Mixed Integer Linear programming formulation that provides substantial gains in reducing the box assortment size, improving volumetric efficiency by 12%, reducing CO2 emissions by 8 thousand metric tons, and resulting in a cost savings of about 3.7M USD accrued over two years. In the subsequent phase, we develop an innovative hybrid optimization framework combining unsupervised learning, reinforcement learning, and tree-search. Specifically, the optimization problem is formulated as a sequential decision-making task called the box-sizing game. A neural network agent is then designed to learn to play the game and eventually solve the problem. This approach promises a further improvement of 5% in volumetric efficiency.

Small World AI


Primary Presenter: Haoran Zhang, LocationMind Inc.
Zhilng Guo, LocationMind Inc.
Dou Huang, LocationMind Inc.
Xiaodan Shi, Center for Spatial Information Science, The University of Tokyo
Peiran Li, Center for Spatial Information Science, The University of Tokyo
Jinyu Chen, Center for Spatial Information Science, The University of Tokyo
Yuhao Yao, Center for Spatial Information Science, The University of Tokyo
Qing Yu, Key Laboratory of Road and Traffic Engineering of the Ministry of Education, Tongji University
Wenjing Li, Center for Spatial Information Science, The University of Tokyo
Yanxiu Jin, Center for Spatial Information Science, The University of Tokyo
Zhiheng Cheng, Center for Spatial Information Science, The University of Tokyo
Xudong Shen, Center for Spatial Information Science, The University of Tokyo
Wenyi Lu, Center for Spatial Information Science, The University of Tokyo
Ning Xu, China University of Petroleum-Beijing, Beijing Key Laboratory of Urban oil and Gas Distribution Technology

Integration of mobility information with geospatial information data gives it the potential to predict and draw insights that identify broader impacts in temporal, spatial, and demographic perspectives. However, developing a relationship between available static datasets and dynamic human mobility remains a critical challenge. Moreover, the need for real-world data to estimate, predict and digitally replicate real-world scenarios for stakeholders in decision-making to design a safe, resilient, and energy-efficient sustainable future is at an all-time high. We developed the Small World AI (SWAI), solving this challenge. SWAI stands for Spatial Multimodal ALL-World Artificial Intelligence. Within the growing ambit of Machine Learning and AI, we introduce Spatial Multimodality. We combine two areas of advanced Geo-analytics- Geospatial datasets creation and Multilayered AI configuration to analyze collected datasets to generate insights. SWAI has provided insights in various fields, especially for the COVID-19 situation in Japan. Geospatial data has played an important role in estimating the threats and impact of the pandemic and further aided decision-making. We are confident that such a dual-prong approach helps reduce redundancy in data, increases synergies, and eases replication of the technology anywhere globally.

Online Non-Parametric Regression for Sales Forecast amid a Pandemic


David Simchi-Levi, Massachusetts Institute of Technology
Michelle Wu, Massachusetts Institute of Technology
Primary Presenter: Ruihao Zhu, Purdue University
Rui Sun, Amazon
Felipe Aragao, AB InBev
Obrad Scepanovic, AB InBev
Ivo Montenegro, AB InBev
Nitesh Bhardwaj, AB InBev
Tapas Ray, AB InBev

Motivated by collaboration between AB InBev and the MIT Data Science Lab, we consider the problem of forecasting sales under the COVID-19 pandemic. Our approach combines non-parametric regression, game theory, and pandemic modeling to develop an online non-parametric regression method. Specifically, the method takes the future COVID-19 cases estimates as an input, and outputs the level of calibration for AB InBev’s baseline sales forecast. Our algorithm sequentially predicts the label (i.e., the level of calibration) of a random covariate (i.e., the number of active COVID cases) given past observations and the generative process of future covariates. To provide robust performance guarantee, we derive a computationally efficient algorithm that minimizes regret over all possible scenarios and prove that the algorithm is rate optimal. We demonstrate its performance on AB InBev’s datasets of three different markets. The numerical experiments show that our method is capable of reducing AB InBev’s forecasting error in terms of WMAPE by more than 37%.

An Optimization and Artificial Intelligence Based Cooperative Mucilage Cleaning Approach via a Mixed Fleet of Autonomous Unmanned Surface Vehicles and Drones


Primary Presenter: Mumtaz Karatas, National Defence University, Turkish Naval Academy, Department of Industrial Engineering
Levent Eriskin, National Defence University, Turkish Naval Academy, Department of Industrial Engineering
Mustafa Erol, IWROBOTX Co.Ltd., TechnoPark

Addressing a real-life problem faced in the surrounding seas of Turkey, in this study, we propose a cooperative mucilage cleaning approach based on optimization and artificial intelligence, and the utilization of autonomous Unmanned Surface Vehicles (USVs) and drones. The proposed approach (i) uses a machine learning algorithm to create regional mucilage distribution and density maps for the area where mucilage is spotted, (ii) implements a partitioning procedure to generate sub-areas that represent islands of mucilage zones for assigning cleaning vehicles, (iii) solves a combinatorial optimization model called the “Multi-vehicle Mucilage Cleaning Problem” for generating optimal routes and assignments for USVs and drones, and for determining the location of mucilage discharge station. Considering the extent and urgency of the mucilage disaster, our approach will bolster the efforts made for alleviating the ill effects of the mucilage. The initial results of our approach indicate that it is possible to reduce the operational cleaning costs significantly. Being the first one in the literature which incorporates a heterogeneous fleet of multiple maritime and air vehicles cooperatively in an intelligent framework, our holistic cleaning methodology has a generic structure which allows applying it on a variety of operations such as oil spill cleaning.

Data-Driven Surgical Tray Optimization to Improve Operating Room Efficiency


Vinayak Deshpande, Kenan-Flagler Business School, The University of North Carolina at Chapel Hill
Nishanth Mundru, Kenan-Flagler Business School, The University of North Carolina at Chapel Hill
Primary Presenter: Sandeep Rath, Kenan-Flagler Business School, The University of North Carolina at Chapel Hill
Martyn Knowles, UNC Rex Hospital
David Rowe, Operative Flow Technologies
Benjamin C. Wood, UNC Rex Hospital

Surgical procedures account for 60% of the operating cost of a hospital. About 15% of these operating costs relate to surgical instruments and supplies. Hospitals spend several million dollars a year on instrument sterilization, tray assembly, and repurchase costs. However, less than 30% of instruments supplied to surgery are used. OpFlow, a healthcare software company, has developed a novel, proprietary solution for cost and time-efficient data collection, storage, and analysis of instrument usage data directly from the operating room.

With this data, we formulate a data-driven mathematical optimization model for surgical tray configuration and assignment to reduce the costs of unused instruments. Our solution methodology scales to thousands of surgeries, thousands of instruments, and hundreds of surgical trays. Our model-based approach improved tray configuration and assignment, leading to a 54% reduction in unused instruments per surgery compared to the status quo. Implemented tray configurations were monitored closely over several months at a partner hospital, and the results demonstrated significant savings in costs without compromising instrument availability. We estimate projected annual cost savings of $1.39m in instrument-related costs at the hospital. Trays optimized by Opflow Inc. have been implemented at several hospitals, saving several million dollars in surgical instrument processing and repurchase costs.

Using Analytics to Detect Blight and Enforce Codes in Cities


Erik Johnson, Culverhouse College of Business, The University of Alabama
Brandon Kaesteler, Habitat for Humanity of Tuscaloosa
Scott Holmes, City of Tuscaloosa
Walt Maddox, City of Tuscaloosa
Robert Hogue, City of Springfield
Richard Benanti, City of Springfield
Gavin Baum-Blake, City Detect
Ezra Coutre, City Detect
Comer Jennings, City Detect
James Mulvey, City Detect
Primary Presenter: James J. Cochran, Culverhouse College of Business, The University of Alabama

Blight code enforcement monitoring is a major municipal expense, and it depends on reactive processes such as community reporting or inspections of properties by code enforcement officers. By the time violations are identified, the blight has often become more extensive and more difficult to remediate. Furthermore, enforcement officers must perform several return trips to monitor remediations. Our process of automated blight detection and code enforcement is proactive and frees code enforcement resources to focus on remediation. Our monitoring detects nuisance properties before they reach a ‘tipping point’, provides code enforcement officers with specific universally defined code violations to help alleviate perceived bias, and helps prioritize remediation.

We have worked for a year with the city of Tuscaloosa on implementation, mounting intelligent cameras to garbage trucks. The photographs taken from automatically defined target points are uploaded to the cloud, where our AI system identifies International Property Maintenance Code violations. Based on city priorities, we assign a blight score to each property, then combine these measures with other city level data to produce reports for code enforcement officers, city planners, and remediation partners (such as Habitat for Humanity, who we work with to identify damaged roofs). We recently completed a successful pilot in Springfield, IL, and are initiating a pilot in Irondale, AL.