Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

MD50

4 - Multi-compartment Vehicle Routing Problem for Oil Products Distribution with Transshipment Yaxian Zhou, Dalian University of Technology, No 2 Linggong Road, Gaoxinyuan District, DaLian, 116023, China, Lijun Sun, Haiyang Shi, Xiangpei Hu This paper addresses the multi-compartment vehicle routing problem for oil products distribution with transshipment. A mixed distribution scheme combining separate distribution and co-distribution is proposed to reduce the total distribution cost. We solve the problem with the Clarke and Wright savings algorithm incorporating the exchange and insertion operators. Numerical results show that the mixed scheme is always better than the separate distribution used in practice. As a result, the mixed scheme is advisable in oil products distribution with transshipment. 5 - Parallel-batching Scheduling: Impact of Adding Machine Capacity Jun Xu, Northwestern Polytechnical University, Xi’an, China, Junqiang Wang Scheduling performance can be improved with extra cost by resource augmentation such as adding machine capacity. This paper investigates the impact of adding machine capacity on the scheduling objective function. We measure such capacity impact by the ratio and give the bounds on the capacity impact for the problem to minimize the makespan and total completion time. Furthermore, to trade off gains in scheduling performance against the losses in adding machine capacity extra cost, we propose a cost-effective approximate algorithm and analyze the corresponding worst-case ratio. Chair: Nohel Zaman, Blacksburg, VA, 24060, United States 1 - Effective Text Mining Techniques for Understanding Employee Opinions David Michael Goldberg, Virginia Tech, 880 West Campus Drive, Suite 1007, Blacksburg, VA, 24060, United States, Nohel Zaman Employee satisfaction is a vital component of workplace culture, and it has substantial impacts in firm performance. In this work, we propose and compare several text analytics methods for understanding employee perceptions of their work environment using data from online discussions. These techniques will allow firms to understand nuanced employee perceptions and to pursue remediation if necessary. 2 - The Effects of Online Consumer Reviews on Product Sales in Mobile Apps Industry: An Empirical Study Zhilei Qiao, Virginia Institute of Technology, 880 West Campus Drive, Blacksburg, VA, 24061, United States Prior studies show that online consumer reviews have significant impacts on consumers’ adoption decisions and firms’ sales. This study extends the previous research and examines the effect of online user reviews on new product sales in the mobile app industry. Our results show that the review valence has a significant positive effect on new paid products, whereas the review valence has a significant negative effect on free products. Furthermore, we find that in-app purchase option has a moderating effect on the relationship between review characteristics and product sales for free and paid products. These surprising results show that review ratings have strong persuasive effects on product sales. 3 - Comparing Text Analytics Methods for Understanding Hotel Reviews Nohel Zaman, Virginia Tech, 880 West Campus Drive, Suite 1007, Blacksburg, VA, 24060, United States, David Michael Goldberg The hotel industry is rife with various service issues, which lead to dissatisfied customers. Online user-generated content presents firms with a valuable opportunity to collect business intelligence and to understand these problems. This study focuses on comparing the performance of text analytics for detecting these instances in online media and determining how best to analyze different types of consumer complaints. n MD52 North Bldg 231C Joint Session SMA/Practice Curated: Social Media Analytics and Business Value Emerging Topic: Social Media Analytics Emerging Topic Session

n MD50 North Bldg 231A

Freestyle O.R. Supreme: Frame that Problem Sponsored: INFORMS Section on Practice (formerly CPMS) Sponsored Session Chair: Carrie Beam, University of Arkansas, Walnut Creek, CA, 94596, United States 1 - Freestyle OR Supreme Moderator Carrie Beam, University of Arkansas, Walnut Creek, CA, 94596, United States Join us at our Freestyle OR Supreme competition. Our ‘clients’ will present their OR/Analytics problem to our teams, who will have to formulate and present their OR/Analytics solution on the fly - with no Internet. This is great experience to frame and solve a problem quickly, and a great showcase for any budding consultants (or hiring companies) out there. Winners will be selected by the client, with possible influence from audience applause. Panelists James T. Williams, FICO, 2665 Long Lake Rd, Building C, Roseville, MN, 55113, United States Aaron Burciaga, Booz Allen Hamilton, 4305 Majestic Lane, Fairfax, VA, 22033, United States Sanjay Saigal, Graduate School of Management, UC Davis, 327 Oakwood Place, Menlo Park, CA, 94025, United States Kenneth J. Fordyce, Arkieva, 5460 Fairmont Drive, Wilmington, DE, 19808, United States in Memory of Joe Leung, Emerging Topic Session Chair: Yumei Hou, City University of New York, College of Staten Island 1 - Multitasking Scheduling Problem with Deteriorating Job and Efficiency Promotion Yan Wang, Northwestern Polytechnical University, Xi’an, China, Junqiang Wang In a human-related scheduling environment, operators have a need for multitasking to cope with the challenges like deterioration effect. Multitasking scheduling allows the processing of a primary job can be interrupted by waiting jobs that are available but unfinished. We consider the position-dependent deterioration of the primary job, and positive effect of the waiting jobs on efficiency promotion. This research trades off gains in multitasking against losses in deterioration effect. The related research methods and results will provide a scientific support to efficiently improve production scheduling performance, and further a new insight to better understand human-related behavior. 2 - Minimizing Total Completion Time on Flow Shop with Unavailability Constraint Hairong Zhao, Purdue University Northwest, 2200 169 Street, Hammond, IN, 46323, United States We consider the problem of minimizing total completion time on 2 stage flow shop with the unavailability constraint on the second machine. We show that some special cases are solvable optimally in polynomial time. For the general case, the problem is extremely difficult. We develop some meta-heuristics whose performance is evaluated by experimental results. 3 - Multicriteria Scheduling Subject to Limited Machine Availability Yumei Huo, City University of New York, College of Staten Island, 2800 Victory Boulevard, 1N 215, Staten Island, NY, 10314, United States We study the problems of multi-criteria scheduling subject to machine availability. As Panwalkar et al. points out, decision makers are often faced with the problem of satisfying several different groups of people simultaneously. On the other hand, machines may not be continuously available due to breakdown, preventive maintenance or processing unfinished jobs from a previous planning horizon. So it is natural to consider bi-criteria scheduling subject to the limited machine availability. Theoretically, the problems of multi-criteria scheduling subject to machine availability are more complicated. In this talk, we will present all the polynomial solvable problems. n MD51 North Bldg 231B Scheduling II Emerging Topic: Project Management and Scheduling,

232