INFORMS 2021 Program Book

INFORMS Anaheim 2021

WC35

3 - Artificial Intelligence and Autonomy Trevor Bihl, Air Force Research Laboratory, Wheelersburg, OH, United States Autonomous systems are in development with promises to revolutionize everything from cars to commerce to government. To be autonomous, systems must sense and extract meaning from their surroundings, and take actions to achieve their goals. This involves a considerable amount of intelligence and the fields of autonomy and artificial intelligence (AI) are necessarily heavily intertwined. This talk will present a discussion of what is autonomy, what is AI, how they are related, and what the future holds.

WC35 CC Room 210A In Person: Optimizing Ride-Sourcing Vehicle Routing, Tolling and Information Provision General Session Chair: Kenan Zhang, Northwestern University, Evanston, IL, 60208- 0001, United States 1 - A Co-optimization Approach for Compensating Toll Facility Concessionaires for Reduced Tolls during Roadway Maintenance Sohrab Mamdoohi, George Mason University, 10570 Main St Apt 413, Fairfax, VA, 22030-7108, United States, Elise Miller-Hooks, Jonathan Gifford Execution of improvement activities, such as maintenance and repair, along roadways reduces roadway capacity and, thus, increases traffic congestion. This study investigates the potential to reduce such resulting congestion through an option to reimburse a toll facility concessionaire for decreasing or suspending tolls during improvement activity execution. The problem of determining optimal toll prices, concessionaire remuneration and improvement action schedules is formulated through concepts of bilevel mixed integer programming and equilibrium modeling. 2 - A Game Theorectic Approach for Ride-hail Vehicle Routing Kenan Zhang, Northwestern University, Evanston, IL, 60208-0001, United States, Nie Yu This study proposes a game-theoretic approach to modeling the routing behaviors of drivers in a spatiotemporal ride-hail market. Driven by profit, idle drivers move across local markets for passenger search and their search strategies are modeled as Markov decision process (MDP). Since the probability of meeting a passenger in a local market is determined by the number of idle drivers, each driver has to optimize his own search strategy given others’ strategies. The collective vehicle routing behaviors lead to an MDP congestion game. We characterize the congestion game for two common ride-hail modes, namely, street-hail and e-hail, and develop a solution algorithm to solve the equilibrium. WC37 CC Room 210C In Person: Advances in Analytics for Military and Security Applications I General Session Chair: Trevor Bihl, Air Force Research Laboratory, Wheelersburg, OH, 45694, United States 1 - Equitable Assignment of U.S. Marine Corps Reserve Recruiting Gary Lazzaro, Permanent Military Professor, United States Naval Academy, Annapolis, MD, United States We focuses on the assignment of new reservist Marines to jobs at Reserve Centers by Marine Corps Recruiting Command (MCRC). Recruiting Substations acquire new reservists to be stationed at Reserve Centers to fill specific job openings. Manual assignment of jobs to Recruiting Substations takes MCRC weeks to complete. We create a novel application of the classic assignment problem with additional constraints for maximum travel distance and recruiter workload to automate the process. Our model displays the 6,011 job openings for FY2021, their assigned Recruiting Substation, and the distance between the Recruiting Substation and Reserve Center in miles. 2 - Meta-heuristic Optimization Methods for Quaternion-valued Neural Networks Jeremiah Bill, Air Force Institute of Technology Real-valued neural networks have demonstrated promising, and often striking, results across a broad range of domains. This has driven a surge of applications utilizing high-dimensional datasets. While many techniques exist to alleviate issues of high dimensionality, they all induce a cost in terms of network size or computational runtime. This work examines the use of quaternions in neural networks. The constructed networks demonstrate the ability of quaternions to encode high-dimensional data in an efficient manner while reducing the number of total trainable network parameters compared to their real-valued equivalents. Finally, this work introduces a novel training algorithm using a meta-heuristic approach that bypasses the need for a quaternion chain rule and analytic quaternion loss or activation functions.

Wednesday, 1:30PM 2:30PM

Wednesday Keynote 01 CC Ballroom A /Virtual Theater 1 Keynote: Boundary-Expanding OR/OM Research Keynote Session 1 - Boundary-Expanding OR/OM Research Rachel Q. Zhang, The Hong Kong University of Science & Technology, Dept of IEEM, Clear Water Bay, Kowloon, Hong Kong OR and OM have brought about significant improvements to operations in diverse domains, including military, manufacturing and service, and the knowledge economy. Every technological advance in the modern world has been met with the pursuit of new models by the OR/OM community, often providing fundamental understanding of and significant improvements to its deployment. In this talk, the speaker will share her experience in pursuing research in the boundaries of operations and finance, wireless communications and blockchains, including the inspirations, execution, challenges and lessons learned. Pursuing such projects is not without risk, but is an effective way for a researcher to reinvent him/herself and have a fulfilling career. Wednesday Keynote 02 CC Ballroom B /Virtual Theater 2 Edelman Reprise: UN World Food Programme Keynote Session” 1 - The world’s leading humanitarian organization, and 2020 Nobel Peace Prize Laureate, is saving and changing lives by delivering food assistance in emergencies and working with communities to improve nutrition and build resilience. In 2020, WFP assisted nearly 100 million people across 88 countries. Analytics has underpinned WFP’s management of its vast and complex humanitarian operations, helping it reach more people, respond faster in emergencies, and realize significant savings that are used to improve lives and empower communities. Wednesday Keynote 03 CC Ballroom C /Virtual Theater 3 Keynote: Operational Data Driven Interventions to Decrease Adverse Events Associated with Opioid Overdose Keynote Session 1 - Operational Data Driven Interventions to Decrease Adverse Events Associated with Opioid Overdose Mahesh Nagarajan, University of British Columbia, Sauder School Of Bus. 2053 Main Mall, Vancouver, BC, V6T 1Z2, Canada In this talk, we present a systematic data driven approach to decrease adverse events associated with overdose episodes. We take a three-fold approach. First, we examine pathways that result in opioid use and devise protocols to decrease the number of new users. Second, we predict adverse occurrence of adverse episodes among current users and adopt timely interventions that will decrease the likelihood and severity of an event. Third, we focus on the care pathways for existing users and use simple operational techniques to increase the system’s capacity as well as improve outcomes.

166