Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

TE76

3 - Column Generation Algorithm for Preventive Maintenance Scheduling with Hierarchical Set-ups Yongpei Guan, University of Florida, 303 Weil Hall, P.O. Box 116595, Gainesville, FL, 32611, United States, Eric C. Blair We study a preventive maintenance (PM) scheduling problem where there are multiple PM activities that must be completed within some time-variant cycle limit, along with a group of set-up activities which must be performed in conjunction with certain PM activities based on a tree-like dependency graph. The objective is to minimize total cost of performing set-up and PM activities over a finite horizon. Relaxing the dependency constraints, we show the problem with only cycle limit constraints has the integrality property. We develop a column generation framework with pricing problems solvable in polynomial time exploiting this structure, and show computational results of the approach. Joint Session PSOR/MIF: Non-profit Operations in Humanitarian and Disaster Response Logistics Sponsored: Minority Issues Sponsored Session Chair: Trilce Encarnacion, Rensselaer Polytechnic Institute, Troy, NY, 12180, United States 1 - A Minimal Technology Routing System for Food Banks Mobile Food Pantry Programs Eduardo Perez, Assistant Professor, Texas State University, Roy F. Mitte Complex, 749 N. Comanche St., San Marcos, TX, 78666, United States, Emilia Ochoa, Jesus Jimenez The Mobile Pantry Program directly serves clients in areas of high need with the goal of supplementing other hunger-relief agencies. Through a Mobile Pantry, a truckload of food is distributed to clients in pre-packed boxes or through a market-style distribution where clients choose to take what they need. A novel routing system is developed to handle the daily routing of a limited number of vehicles serving varying number of delivery points. A case study is presented that shows how the new routing system can combine multiple visiting sites in one route to increase throughput rates. 2 - Assessing Food Supply Chain Vulnerability Using Consumer Behavior and Mitigation/Compliance Strategies Jessye Bemley Talley, Morgan State University, Baltimore, MD, 21251, United States As the food supply chain continues to grow globally, it is important to have strategies (i.e. recall) and tools to keep products safe when purchased by the end consumer. However, the guidelines that indicate the best time to implement a recall is uncertain. It is also unclear on the best mode of messaging to communicate with consumers about these contamination events. An agent-based simulation approach is used to address the following: (1) consumer behavior, (2) messaging strategies, and (3) timing of mitigation strategies to reduce consumer illnesses. 3 - Making Non-profit and Service-providing Agencies Smarter with Big Data Analytics Charalampos Chelmis, University at Albany, State University of New York, NY, United States, Daphney-Stavroula Zois, Wonhyung Lee Currently, those in need of help often do not know how to locate or access service providers. Likewise, service-providing agencies often work in silos. Response becomes even more problematic when a problem demands the coordination of service providers, volunteers, and government structures, and after business hours, when the communication channels that can aid people in need become sparse. In this work, we report our ongoing project towards developing analytic solutions to help non-profit and community-based organizations respond effectively to the needs of their communities. We report our preliminary findings, and discuss key challenges and future plans. 4 - Role of Non-established Groups in the Aftermath of Hurricanes Harvey, Irma, and Maria Trilce Encarnacion, Rensselaer Polytechnic Institute, 110 8th Street, Jonsson Engineering Center # 4049, Troy, NY, 12180, United States, Jose Holguin-Veras, Diana Gineth Ramirez-Rios, Johanna Amaya Research has already established that the civic society plays a major role in all aspects of the response to extreme events, from search and rescue and relief distribution, to rebuilding efforts. This paper presents preliminary findings of the role of civic society and other non-established groups is the disaster response efforts to Hurricanes Harvey, Irma, and Maria impacted the US and the Caribbean in 2017. n TE76 West Bldg 212C

5 - Value Function Approximation for Last-mile Distribution in Humanitarian Relief Robert Alexander Cook, University of Alabama, Northport, AL, 35473, United States, Emmett J. Lodree This study describes a Value Function Approximation approach for solving a Markov Decision Problem in which we distribute stochastically-arriving donations to disaster survivors. Donations accumulate over time at collection sites and are periodically transported to a relief center where the donations are distributed to beneficiaries over a finite horizon. n TE77 West Bldg 213A Matching and Recommendation Systems with Social Impact Sponsored: Public Sector OR Sponsored Session Chair: Tina Rezvanian, Northeastern University,Boston, MA, 02115- 4420, United States Co-Chair: Ergun Ozlem 1 - On-demand Volunteers Through a Hierarchical Approach with Personalized Recommendations and Choice A new hierarchical approach enables on-demand volunteers by recasting the platform’s role as one providing personalized recommendations (i.e., a menu of multiple requests) to ad hoc volunteers. Choice can increase participation (capacity) and resource utilization. A novel bilevel optimization framework captures the interdependent outcomes of volunteer selections. The computationally expensive mixed integer linear bilevel problem is transformed into a single level problem by proposing logical expressions. Insights on what influences the optimal number of choices and performance are explored. 2 - Maintaining Diversity in Student Exchange M. Utku Unver, PhD, Boston College, Boston, MA, United States, Umut Dur In Europe, annually 200,000 college students study in different countries through Erasmus Student Exchange. Although this exchange is successful in many dimensions, it has two major drawbacks. First, the number of students exchanged is not country-balanced. Unbalanced exchange causes a financial burden for countries that are importing more students than their exports. Second, certain countries are exchanging students mainly between each other, limiting diversity, one of the main goals of the exchange. We propose the diversity-motivated top- trading-cycles-and-chains mechanism to solve these problems by satisfying constrained efficiency, strategy-proofness, and fairness. 3 - Centralized Mechanisms for Two-sided Stable Matching with Societal Impacts Tina Rezvanian, Boston, MA, United States Agents often are not aware of all their options or making applications for all positions can be costly. We leverage the staff assignment problem in the World Food Programme where a perfect assignment of staff and job positions is required. Perfect matching under partial information is incompatible with ensuring stability. To address this issue, we design a multi-period mechanism that negotiates on behalf of self-interested agents and identifies an increased-cardinality matching that guarantees a minimum level of stability while maximizing social welfare for all agents across multiple periods. 4 - Constructing Ordinal Preference Lists with Ties Using Spectral Clustering Matthew Williams, Northeastern University, Boston, MA, United States We study the matching of military cadets to assignments over two periods, modeling the problem as a weighted, rank-maximal capacitated house allocation problem with ties (CHAT). To ensure every instance of the problem admits an agent-complete matching, we propose obtaining agent-specified ranks of the categorical attributes of assignments. Using spectral clustering, with these weights as input, we construct complete ordinal preference lists, with ties, for each agent. Lastly, we propose a generalization of CHAT to include gender constraints for further study. Jennifer A. Pazour, Rensselaer Polytechnic Institute, 110 8th Street, CII 5217, Troy, NY, 12180, United States, Seyed Shahab Mofidi

406