Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

MD77

4 - Routing and Staffing when Servers are Strategic Ragavendran Gopalakrishnan, Cornell University, 220 Hollister Hall, Department of Civil and Environmental Engg., Ithaca, NY, 14853, United States, Sherwin Doroudi, Amy R. Ward, Adam Wierman Traditionally, research focusing on the design of routing and staffing policies for service systems has modeled servers as having fixed (possibly heterogeneous) service rates. However, service systems are generally staffed by people. Furthermore, people respond to workload incentives; that is, how hard a person works can depend both on how much work there is, and how the work is divided between the people responsible for it. In a service system, the routing and staffing policies control such workload incentives; and so the rate servers work will be impacted by the system’s routing and staffing policies. This observation has consequences when modeling service system performance, and our objective in this paper is to investigate those consequences. We do this in the context of the M/M/N queue, which is the canonical model for large service systems. First, we present a model for “strategic servers that choose their service rate in order to maximize a trade-off between an “effort costö, which captures the idea that servers exert more effort when working at a faster rate, and a “value of idlenessö, which assumes that servers value having idle time. Next, we characterize the symmetric Nash equilibrium service rate under any routing policy that routes based on the server idle time (such as the longest idle server first policy). We find that the system must operate in a quality-driven regime, in which servers have idle time, in order for an equilibrium to exist. The implication is that to have an equilibrium solution the staffing must have a first-order term that strictly exceeds that of the common square-root staffing policy. Then, within the class of policies that admit an equilibrium, we (asymptotically) solve the problem of minimizing the total cost, when there are linear staffing costs and linear waiting costs. Finally, we end by exploring the question of whether routing policies that are based on the service rate, instead of the server idle time, can improve system performance. n MD77 West Bldg 213A Joint Session PSOR/Practice Curated: Emergency Medical Services Sponsored: Public Sector OR Sponsored Session Chair: Pieter van den Berg, RSM, RSM, Rotterdam, 3062 PA, Netherlands 1 - A Scenario-based Ambulance Location Problem with Two Types of Servers Soovin Yoon, University of Wisconsin-Madison, 1415 Engineering Drive, Room 3261, Madison, WI, 53706, United States, Laura Albert Emergency medical service planning is challenging when there are heterogeneous patients and ambulances. Some of the challenges stem from uncertain demand and interdependencies between ambulances. We propose a data-driven approach that lifts distributional assumptions by sampling call arrival scenarios directly from the call log. A two-stage stochastic integer program is formulated to deploy and dispatch two types of servers to serve calls promptly and also match server types to varying patient needs. We conduct the numerical study with two real-world datasets to demonstrate the effectiveness of our model. 2 - Determining Ambulance Destinations in the Presence of Offload Delay Using an Markov Decision Process Mengyu Li, PhD Candidate, Dalhousie University, Halifax, NS, Canada, Peter Vanberkel Ambulance offload delay (AOD) is a prolongation between an ambulance arrival in the emergency department (ED) and transfer of patient care, typically due to ED crowding. We formulate an infinite horizon, discrete-time Markov decision process (MDP) model to determine when it is advantageous to send appropriate patients to out of region hospitals. Out of region hospitals have longer transport times but shorter offload times. The decision model considers patient acuity, travel distance, and AOD. A computational study is applied and a policy to return ambulances to service more quickly is found. This model can be used as a decision support tool to generate optimal ambulance patient allocation policy. 3 - Shift Schedule Optimization for Basic Life Support Ambulances Using Stochastic Programming Pieter van den Berg, RSM, Burgemeester Oudlaan 50, Rotterdam, 3062 PA, Netherlands, Theresia van Essen Many ambulance services have a fixed schedule of shifts for their vehicles. This defines the available capacity for each time of the day, which does not always match the demand for ambulances. We present a Stochastic Programming model to optimize the shift schedules of Basic Life Support (BLS) ambulances that are used for non-urgent patient transportation. By optimizing the schedule based on a large set of simulated scenarios, we find schedules that can improve the service provided to non-urgent patients. As emergency ambulances execute any patient transportation that cannot be served by a BLS ambulance, this also improves the coverage for emergency calls.

n MD78 West Bldg 213B Resilient Infrastructure and Community Networks Sponsored: Public Sector OR Sponsored Session Chair: Kash Barker, University of Oklahoma, OK, 73019, United States 1 - Social-vulnerability Driven Infrastructure Network Component Importance Measures Deniz Berfin Karakoc, University of Oklahoma, 202 W. Boyd, Room 448, Carson Engineering Center, Norman, OK, 73019, United States Critical infrastructures are often described as interdependent in nature, vulnerable to multiple hazards, and vital for societies. To plan a comprehensive preparedness plan, we identify the critical components that have the largest impact on the performance of these networks and on the society. In this work, we propose two different component importance measures to quantify the effect of disruption and the impact of restoration on components over the resilience of interdependent infrastructure networks. We integrate multiple social vulnerability measures using a multi-criteria decision analysis technique and illustrate our approach with three infrastructure networks in Shelby County, TN. 2 - Restoring Community Structures in Interdependent Infrastructure Networks Kash Barker, University of Oklahoma, 202 West Boyd, Room 124, Norman, OK, 73019, United States, Yasser A. Almoghathawi Community structures exist in many infrastructure networks where each network is partitioned into sets of densely connected components with sparse connections between them. However, infrastructure networks depend on one another for their proper functionality, which potentially make them highly vulnerable to disruptions. This work studies the restoration problem of community structures in interdependent infrastructure networks following a disruption. A restoration model is proposed with the objective of enhancing the resilience of the system of interdependent networks considering their interdependencies. The proposed model is illustrated through a set of interdependent networks. 3 - Measuring Multi-dimensional Network Resilience Christopher Zobel, Virginia Tech, 880 West Campus Drive, Blacksburg, VA, 24061-0235, United States, Milad Baghersad This paper discusses a new multi-dimensional measure of critical infrastructure network resilience that takes the relative resilience of different network components into account. The new measure allows for explicitly considering the tradeoffs between system loss and recovery time, and it also allows for applying relative weights to different sub-components in order to represent varying degrees of criticality within the network. 4 - Infrastructure Resilience Evaluation in Multiple Public Transportation Systems when Disruptions Occur, the Case of New York City Gabriela Noemi Gongora Svartzman, Stevens Institute of Technology, 1 Castle Point On Hudson, Hoboken, NJ, 07030, United States, Jose Emmanuel Ramirez-Marquez, Kash Barker Public transportation systems can be subject to unforeseen events, ranging from sick passengers up to weather conditions. The novelty of this research resides in evaluating the resilience of public transportation as multi-modal, interdependent systems. The usual behavior of different modes of transportation (buses, subway, bicycles) is modeled through discrete events simulations, and compared against their real data when a range of disruptions occur. The aggregated commuting times, the type of area affected (e.g. residential or commercial) and the possible transportation alternatives are calculated. Visualizations are created to analyze New York City as a case study. 5 - Stochastic Optimization for Maintenance Decisions in Transportation Networks under Seismic Hazard Jorge-Mario Lozano, Universidad de los Andes, Bogota, 111711, Colombia, Camilo Gómez, Jack W. Baker Transportation networks are particularly vulnerable to seismic hazards and critical for cities’ sustainability. Hence, deciding whether to retrofit a bridge in the present, or wait to repair it after an earthquake, is important to ensure reasonable travel times and limit the disruption of city dynamics. In this paper, we present a risk management optimization approach for the San Francisco Bay Area for a hazard-consistent set of earthquake scenarios. We use a regularized continuous Benders decomposition strategy, which provides advantages with respect to existing combinatorial approaches in terms of convergence, as well as shadow price analysis and prioritization capabilities.

242