Informs Annual Meeting 2017

WC13

INFORMS Houston – 2017

4 - Impact of Business Intelligence in Organizational Resilience Cristina Ruiz Martín, PhD Student, INSISOC, Valladolid, Spain, cruiz@eii.uva.es, Felix Villafañez, Adolfo Lopez Paredes, Gabriel Wainer The aim of this work is to demonstrate that Business Intelligence (BI) is a key issue for organizations to achieve the capabilities and attributes of resilient organizations. In the last century, organizations focused on performance, but in today’s changing environment, this is not enough: continued existence has become a key challenge. One of the main areas of Organizational Resilience research is to identify the characteristics of such organizations. BI aims to help organizations to develop sound strategies in order to survive and thrive taking advantage of data analysis. Organizations can benefit from a combined application of BI and Organizational Resilience. 332C Decision Support Systems Contributed Session Chair: Majed Al-Shawa, Strategic Actions, Waterloo, ON, Canada, malshawa@strategic-actions.com 1 - Repair Scheduling using Generalized Additive Models Grant Lents, United Technologies Research Center, 411 Silver Lane, East Hartford, CT, 06108, United States, lentsgn@utrc.utc.com This work focuses on the scheduling of repairs for parts in a repair organization with multiple sites using machine learning techniques. The goal is to predict an ordinal score for each part based on the stage in which the part is between receiving and shipping and how close the part is to its promised turnaround time. The relationship between the score and explanatory variables was non-linear and a Generalized Additive Model (GAM) was fit to the data across all sites. In the second stage, a 2D assignment algorithm was used to group the sites and assign one GAM model to a group of sites. Our work provided site managers with a interpretable tool for consistent scheduling of the repairs for their parts. 2 - Evaluating and Selecting Data Center Fire Protection System with Analytical Hierarchy Process Xiaojun (Gene) Shan, University of Houston-Clear Lake, Houston, TX, United States, shan@uhcl.edu, Noel S. Ellis Fire protection systems are critical in protecting data center assets. This paper applies analytical hierarchy process (AHP) to evaluation and selection of a fire protection system for a data center. Expert opinions from multiple data center professionals was used to select the most favorable fire protection system. The selection alternatives were chosen from options that are accepted by the Life Safety Code of the National Fire Protection Association (NFPA). Results show that pre-action water sprinkler system is the preferred fire protection system. This study provides insights for data center managers in constructing new data centers or renovating of existing ones. 3 - Environmental, Health, and Safety Department Reorganization and Prioritization: using Analytical Hierarchy Process to Mitigate Scope Creep Xiaojun (Gene) Shan, University of Houston-Clear Lake, Houston, TX, United States, shan@uhcl.edu, William Barton Personnel changes often create a void in experience and knowledge of client processes, decreasing efficiency of client-oriented organizations. The objective of this paper is to demonstrate the use of analytical hierarchy process to establish core priorities of personnel tasks. This is novel to the industry because it allows users to prioritize daily tasks and to recalibrate priorities annually. Examples are given on how to use this tool to prioritize tasks, as the company goals and daily tasks change. The structure of the reorganized department centralizes tasks and reduces duplication. This framework is illustrated with a case study of an Environmental Health and Safety (EHS) department. 4 - Optimal Number of Choices in Rating Contexts Sam Ganzfried, Assistant Professor, Florida International University, Miami, FL, United States, sam.ganzfried@gmail.com In many settings people give numerical scores to entities from a small discrete set, e.g., attractiveness from 1-5 on dating sites and papers from 1-10 for conferences. We study the problem of understanding when using a different number of options is optimal. We study several natural processes for score generation. One may expect that using more options always improves performance, but we show that this is not the case, and that using fewer choices—even just two—can surprisingly be optimal. Our results suggest that using fewer options than typical could be optimal in certain situations. This would have many potential applications, as settings requiring entities to be ranked by humans are ubiquitous. WC13

5 - Decision Making under Uncertainty Srinivasa Prasanna, IIIT-Bangalore, 26/C, Hosur Road, Electronics City, Opposite Infosys Technologies, Bangalore, 560100, India, gnsprasanna@iiitb.ac.in, Abhilasha Aswal In this presentation, we explore a novel framework for systematically analyzing alternative assumptions about uncertainty and for finding robust strategies under deep uncertainty. We present a tractable robust optimization approach where we model uncertainty as polyhedral sets that can be generated from historical data. This representation of uncertain data models has economic meaning and is intuitive to a decision maker, especially in applications like supply chains, finance, transportation etc. 6 - Is it Worth Fighting a Patent Troll? using the Constrained Rationality Framework to Model and Analyze the Showdown Between RIM and NTP, as an Example Majed Al-Shawa, Strategic Actions, Waterloo, ON, Canada, malshawa@strategic-actions.com The RIM vs. NTP holds many features of a classical strategic business conflict that real product innovators found themselves facing. We use Constrained Rationality, a formal value-driven strategic decision and conflict analysis framework with robust multi-agent decision support approach to: model the RIM vs. NTP strategic conflict; analyze the players options and strategies including the possible cooperation between RIM and NTP to settle, and the possible coalitions between NTP and RIM’s competitors; and then elicit the most stable equilibrium end states of this conflict, and similar ones. We, finally, compare the produced stability and coalition analysis with how the conflict actually ended. 332D Managing Capacity and Demand to Improve Efficiency in Healthcare Delivery Sponsored: Manufacturing & Service Oper Mgmt, Healthcare Operations Sponsored Session Chair: Jonathan Helm, Indiana University, Bloomington, IN, 47405, United States, helmj@indiana.edu 1 - The Use of Teletriage in Healthcare Demand Management Ozden Cakici, American University, 4400 Massachusetts Ave, Washington, DC, 20016, United States, cakici@american.edu, Alex Mills Teletriage (TT) is offered by many healthcare providers to help patients get advice over the phone. This decision support should benefit patients and improve demand management in healthcare, but TT is not widely used despite a growing push for telemedicine. We use an MDP to model a patient’s choice of providers during an illness episode where the illness severity is partially observable to the patient and when teletriage is subject to errors. Allowing TT may improve the patients’ outcomes and reduce the rate of arrivals to the emergency department. We extend our analysis to patients who are gain seekers and loss averse. Interestingly, these patients may actually be better off without the option to use TT. 2 - Online Appointment Scheduling Problem with a Rolling Horizon Approach Esmaeil Keyvanshokooh, University of Michigan, Ann Arbor, 2144 Glencoe Hills Drive, Apt 9, Ann Arbor, MI, 48108-1020, United States, keyvan@umich.edu, Cong Shi, Mark P. Van Oyen We study an online appointment scheduling problem in which patients come one by one into a medical center in an online fashion. Upon arrival of each patient, the scheduler should choose both a surgeon and a day over a specific planning horizon without knowing the full information of the subsequent coming patients. We design an online optimization algorithm for this purpose and compute a suitable competitive ratio. 3 - Patient Centered Capacity Planning for Primary Care Appointment Scheduling Ali Dogru, akdogru@crimson.ua.edu, Sharif Melouk Motivated by patient-centered medical home principles, we develop a capacity planning model for appointment scheduling. Considering seasonal demand, patient preferences, no-shows, and lateness, we use stochastic optimization to determine the number of service providers and appointment slot durations to minimize unmet demand, patient waiting, and overage time. Experimentation provides managerial insights. Keywords: Capacity Planning, Appointment Scheduling, Stochastic Optimization, Patient-Centered Medical Home WC14

508