Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

TC84

2 - New Models and Algorithms for Operating Room Scheduling Problems

2 - Design Cost and the Birth of User Entrepreneurs Ohchan Kwon, Harvard Business School, Wyss House, Soldiers Field, Boston, MA, 02163, United States

Yang Wang, Associate Professor, Northwestern Polytechnical University, 127 West Youyi Road, Xi’an, 710072, China, Xue Yang, Haichao Liu, Haibo Wang We propose new continuous and discrete mathematical programming models and design dedicated algorithms for solving operating room scheduling problems. The throughput of surgical patients, overtime of operating rooms and satisfaction of patients are optimized. Our algorithms incorporate new coding/decoding rules to represent feasible scheduling and dynamic programming guided scheduling strategy to significantly reduce computational time. We perform extensive experiments on both simulated and real-life hospital data to verify the merit of this study. 3 - Teaching Healthcare Process Management using Discrete Event Simulation Chester G. Chambers, Asst. Professor, Johns Hopkins University, 100 International Drive, Baltimore, MD, 21202, United States In this case study we use Discrete Event Simulation to teach about management of patient flow in an outpatient clinic. Students are provided with simulation models built using actual clinic data. Students explore issues involving patient punctuality, parallel activities, pre-processing, and appointment scheduling. This case has been used successfully with a wide array of students to teach about ways to develop process measures, experiment with process improvements, and evaluate potential results from interventions. The case serves as a course capstone to tie together elements of experimentation, process analysis, and process improvement. 4 - Optimal Experimental Design for a Partially Observable Simple Birth Process Ali Eshragh, Senior Lecturer, The University of Newcastle, Newcastle, 2308, Australia Our goal is to estimate the rate of growth, lambda, of a population governed by a simple birth process. We may choose n time points at which to count the number of individuals present. But due to detection difficulties, we are able only to observe each individual independently with fixed probability p. We discuss the optimal times at which to make our n observations in order to maximise the Fisher Information for the birth rate lambda. Finding an analytical form of the Fisher Information appears intractable. Nonetheless, we utilise the concept of generating functions to develop a new algorithm to maximise the Fisher Information. Our numerical results reveal the efficiency of this new algorithm. 5 - Clinic Scheduling with Patient Re-entrant Haolin Feng, Sun Yat-sen University, Lingnan College, Guangzhou, 510275, China, Mark Lawley, Michelle M. Alvarado, Stephen Steidle We consider a clinic scheduling with same-day patient re-entrants. The motivation is Mohs Micrographic Surgery (MMS), a surgical method for skin cancer excisions. It repetitively removes and examines one skin layer at a time until a cancer free layer is found. Current scheduling practice results in long in- clinic waiting and doctor overtime due to the stochastic nature of excision and pathology. We develop a model for MMS clinic scheduling to improve patient experience and clinic revenues. The model captures the key characteristics of the surgery-pathology and stochastic re-entrants. Theoretical results and numerical study based clinical data are provided to demonstrate the benefit of the method. n TC84 Hyatt, Russell Practice- Innovation/ Entrepreneurship I Contributed Session Chair: Wenbo Jiang, Northwestern Polytechnical University, 127 West Youyi Road, Beilin District, Xi’an Shaanx, 710072, China 1 - Advancing Digital Transformation: Integrating the Problem Space, the Solution Space, and the Quality Space for Successful Digital Transformation Munir Majdalawieh, Associate Professor, Zayed University, Mirdif, Dubai, 19282, United Arab Emirates Many c-suite executives are facing various challenges such as rigid processes, risk avoidance, digital disruption, interdepartmental conflict, disinterest in new ideas and complacency to move forward. These challenges require a vastly different approach to transform companies. The objective of this paper is to propose an integrated framework for problem space (design thinking), solution space (lean startup, Agile, SDLC), quality space (total quality management, six sigma) and change management space methods for successful digital transformation.

This paper examines how a reduction in design cost affects the entrepreneurial entry in the context of the video game industry. I exploit a sudden business model innovation of a major game engine software company, which provides the low- cost licensing option to some user innovators if their existing contents are compatible, but not others. I find that when the licensing cost decreases dramatically, user innovators are more likely to release independent games based on their prior innovation. The effects are stronger if the innovators’ idea is of high-quality. 3 - Coexistence of Quality and Innovation an Automotive Perspective Donna L. Bell, Wayne State University (Detroit, MI), Redwood City, CA, 94063, United States, Ratna Babu Chinnam, Julia Gluesing In a time when the consumer electronics industry is getting new products to market at a rapid rate, automotive manufacturers must identify ways of getting new products and features to customers faster and with high quality to maintain or increase market share. We provide an analysis of the interviews and surveys completed by professionals of a global automotive company in understanding the impact that quality requirements have on innovation and the advanced product design process. 4 - Too Much is Not a Good Thing: The Inverted U-shaped Relationships of Average Tie Strength, Structural Holes, Intraorganizational Knowledge Search and Transfer Wei Wang, PhD Student, Xi’an Jiaotong University, No.28, Xianning West Road, Xi’an, Shaanxi, 71, Xi’an, 710049, China, Xiaoming Sun, Antonio Capaldo, Wentian Cui In this paper, we predict both average tie strength and structural holes exert inverted U-shaped effects on intraorganizational knowledge search and transfer by inventors and further suggest above relationships are flattener in key inventors compared to normal ones. Analysis of American patent data of 33 largest pharmaceutical firms worldwide from 1975 to 2014 offers support for our conjectures, except for the relationship of average tie strength and intraorganizational knowledge search, which is a gradually diminishing positive curve instead. We thus advance network research on knowledge flow by confirming the existence of curvilinear relationships and adding inventor attributes. 5 - Green Entrepreneurial Orientation for Enhancing Firm Performance: A Dynamic Capability Perspective Wenbo Jiang, PHD, Northwestern Polytechnical University, Chang’an Compus, Xi, 710129, China Despite much attention has been focused on the importance of green entrepreneurial orientation, its impact on environmental and financial performance remains unclear. Drawing on dynamic capability theory, we hypothesized that green entrepreneurial orientation has positive influences on two dimensions of firm performance, and green technology dynamism and knowledge transfer and integration play moderating roles in the relationship between green entrepreneurial orientation and firm performance. We tested the research hypotheses using data from 264 Chinese firms. This study enhances our understanding of green entrepreneurial orientation, as a dynamic capability applied in the firm. n Tuesday General Posters North Exhibit Hall E Tuesday Poster Session Poster Session Chair: Neng Fan, University of Arizona, Tucson, AZ, 85721, United States Co-Chair: Junming Yin, University of Arizona, Tucson, AZ, 85721, United States Co-Chair: Burcu B. Keskin, University of Alabama, Tuscaloosa, AL, 35406, United States 1 - Training Data Construction for Recommender Systems Hanisha Vemireddy, Iowa State University, Ames, IA, 50011, United States, Sigurdur Olafsson Recommendation systems provide predictions of user preferences based on prior ratings by the user. For users with only a few ratings, the quality of the recommendations may be low. Addressing this shortcoming, we develop an approach that uses modified rating for similar users to improve recommendations for the target user. Tuesday, 12:30PM - 2:30PM

338