Decision, Operations & Information Technologies

Permanent URI for this community

http://hdl.handle.net/1903/2230

Prior to January 4, 2009, this unit was named Decision & Information Technologies.

Browse

Now showing 1 - 20 of 110

Absorptive Capacity And Open Source Software Project Performance
(2007-12-04) Daniel, Sherae Lee; Agarwal, Ritu; Stewart, Katherine; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
The open source phenomenon is an exciting movement that is transforming traditional forms of software development. Some open source software (OSS) projects, such as Linux and Apache, are performing extremely well and rapidly replacing proprietary software in major corporations and governments. In addition to these highly publicized examples, there are legions of OSS projects that have not experienced a similar uptake. The purpose of this dissertation is to understand how and why some OSS projects are able to perform better than others. It explores antecedents of OSS project performance from a knowledge-focused perspective because software development is a knowledge-intensive activity. In particular, it examines the development and effects of absorptive capacity for an OSS project. Absorptive capacity captures the degree to which an organization is able to acquire and assimilate knowledge. In describing how OSS absorptive capacity is developed, this dissertation identifies characteristics and behaviors of project participants that indicate an OSS project's absorptive capacity. I underscore the importance of the characteristics and behaviors of two different sets of project participants in an OSS project: those in the Internet-based user community and those in the development group. To the extent that absorptive capacity influences OSS project performance, I argue that these characteristics and behaviors are critical for OSS project performance. Archival data about OSS projects that use the SourceForge platform are used to empirically test the model developed. This dissertation makes several contributions to theory and practice. The research informs project managers regarding the participants to target and behaviors to encourage that will lead to superior performance for their OSS project. In exploring the effect of absorptive capacity in an OSS project, this dissertation adds to the absorptive capacity literature by examining the interaction of two dimensions of this construct: knowledge acquisition and knowledge transfer. Finally, this dissertation extends the OSS literature by specifically exploring the effect of the Internet-based user community on OSS project performance.
Air Transportation System Performance: Estimation and Comparative Analysis of Departure Delays
(2006-11-22) Tu, Yufeng; Ball, Michael; Jank, Wolfgang; Decision and Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
The U.S. National Airspace System (NAS) is inherently highly stochastic. Yet, many existing decision support tools for air traffic flow management take a deterministic approach to problem solving. In this study, we focus on the flight departure delays because such delays serve as inputs to many air traffic congestion prediction systems. Modeling the randomness of the delays will provide a more accurate picture of the airspace traffic situation, improve the prediction of the airspace congestion and advance the level of decision making in aviation systems. We first develop a model to identify the seasonal trend and daily propagation pattern for flight delays, in which we employ nonparametric methods for modeling the trends and mixture distribution for the residual errors estimation. This model demonstrates reasonable goodness of fit, robustness to the choice of the model parameters, and good predictive capabilities. We emphasize that a major objective is to produce not just point estimates but estimates of the entire distribution since the congestion estimation models envisioned require delay distribution functions, e.g. to produce probability of certain delays or expected traffic levels for arbitrary time intervals. Local optima problems are typically associated with mixture distribution estimation. To overcome such problems, we develop a global optimization version of the Expectation Maximization algorithm, borrowing ideas from Genetic Algorithms. This optimization algorithm shows the ability to escape from local traps and robustness to the choice of parameters. Finally, we propose models to estimate the so called "wheels-off delays" for flights within the NAS while incorporating a dynamic update capability. Approaches are evaluated based on their ability to reduce variance and their predictive accuracy. We first show that how a raw histogram can be misleading when a trend is present and how variance can be reduced by trend estimation. Then, various techniques are explored for variance reduction. The multiple seasonal trends method shows great capability for variance reduction while staying parsimonious in parameters. The downstream ripple effect method further enhances the variance reduction capability and makes real-time prediction practical and accurate. A rolling horizon updating procedure is described to accommodate the arrival of new information. Finally different models are compared with the current model adopted by the ETMS systems and the predictive capabilities of all models are shown.
Algorithms for Online Advertising Portfolio Optimization and Capacitated Mobile Facility Location
(2017) Sahin, Mustafa; Raghavan, Subramanian; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In this dissertation, we apply large-scale optimization techniques including column generation and heuristic approaches to problems in the domains of online advertising and mobile facility location. First, we study the online advertising portfolio optimization problem (OAPOP) of an advertiser. In the OAPOP, the advertiser has a set of targeting items of interest (in the order of tens of millions for large enterprises) and a daily budget. The objective is to determine how much to bid on each targeting item to maximize the return on investment. We show the OAPOP can be represented by the Multiple Choice Knapsack Problem (MCKP). We propose an efficient column generation (CG) algorithm for the linear programming relaxation of the problem. The computations demonstrate that our CG algorithm significantly outperforms the state-of-the-art linear time algorithm used to solve the MCKP relaxation for the OAPOP. Second, we study the problem faced by the advertiser in online advertising in the presence of bid adjustments. In addition to bids, the advertisers are able to submit bid adjustments for ad query features such as geographical location, time of day, device, and audience. We introduce the Bid Adjustments Problem in Online Advertising (BAPOA) where an advertiser determines base bids and bid adjustments to maximize the return on investment. We develop an efficient algorithm to solve the BAPOA. We perform computational experiments and demonstrate, in the presence of high revenue-per-click variation across features, the revenue benefit of using bid adjustments can exceed 20%. Third, we study the capacitated mobile facility location problem (CMFLP), which is a generalization of the well-known capacitated facility location problem that has applications in supply chain and humanitarian logistics. We provide two integer programming formulations for the CMFLP. The first is on a layered graph, while the second is a set partitioning formulation. We develop a branch-and-price algorithm on the set partitioning formulation. We find that the branch-and-price procedure is particularly effective, when the ratio of the number of clients to the number of facilities is small and the facility capacities are tight. We also develop a local search heuristic and a rounding heuristic for the CMFLP.
ANALYSIS OF DISTRIBUTION-FREE METHODS FOR REVENUE MANAGEMENT
(2008-10-14) Gao, Huina; Ball, Michael; Karaesmen, Itir; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Revenue management (RM) is one area of research and practice that has gained significant attention in the past decade. The practice originated in the airline industry, where the idea was to maximize revenues obtained from a fixed amount of resources through differentiation/segmentation and strategic use of pricing and capacity. While many of the research models take into account uncertainty, the uncertainty is modeled using random variables and known probability distributions, which is often difficult to estimate and prone to error for a variety of reasons. For instance, demand patterns can fluctuate substantially from the past,and characterizing demand from censored data is challenging. This dissertation focuses on the multifare single resource (leg) problem in RM.We consider the "limited information" case where the demand information available consists of lower and upper bounds rather than a characterization of a particular probability distribution or stochastic process. We first investigate the value of the amount and type of information used in solving the single-leg RM problem. This is done via extensive computational experiments. Our results indicate that new robust methods using limited information perform comparably to other well-known procedures. These robust policies are very effective and provide consistent results, even though they use no probabilistic information. Further, robust policies are less prone to errors in modeling demand. Results of our preliminary computations justify the use of robust methods in the multi-fare single-leg problem. We next apply this distribution-free approach to a setting where progression of demand is available through time-dependent bounds. We do not make any further assumptions about the demand or the arrival process beyond these bounds and also do not impose a risk neutrality assumption. Our analytical approach relies on competitive analysis of online algorithms, which guarantee a certain performance level under all possible realizations within the given lower and upper bounds. We extend the robust model from a problem using static information into a dynamic setting, in which time-dependent information is utilized effectively. We develop heuristic solution procedures for the dynamic problem. Extensive computational experiments show that the proposed heuristics are very effective and provide gains over static ones. The models and computations described above assume a single airline, disregarding competition. As an extension of robust decision-making, in the third part of this dissertation, we analyze a model with two airlines and two fare classes where the airlines engage in competition. The model does not use any probabilistic information and only the range of demand in each fare-class is known. We develop a game-theoretic model and use competitive analysis of online algorithms to study the model properties. We derive the booking control policies for both centralized and decentralized models and provide additional numerical results.
APPLYING OPERATIONS RESEARCH MODELS TO PROBLEMS IN HEALTH CARE
(2015) Price, Stuart Patrick; Golden, Bruce; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Intensity- modulated radiation therapy is a form of cancer treatment that directs high energy x-rays to irradiate a tumor volume. In order to minimize the damage to surround-ing tissue the radiation is delivered from multiple angles. The selection of angles is an NP-hard problem and is currently done manually in most hospitals. We use previously evaluated treatment plans to train a machine learning model to sort potential treatment plans. By sorting potential treatment plans we can find better solutions while only evalu-ating a fifth as many plans. We then construct a genetic algorithm and use our machine learning models to search the space of all potential treatment plans to suggest a potential best plan. Using the genetic algorithm we are able to find plans 2% better on average than the previously best known plans. Proton therapy is a new form of radiation therapy. We simulated a proton therapy treatment center in order to optimize patient throughput and minimize patient wait time. We are able to schedule patients reducing wait times between 20% and 35% depending on patient tardiness and absenteeism. Finally, we analyzed the impact of operations research on the treatment of pros-tate cancer. We reviewed the work that has been published in both operations research and medical journals, seeing how it has impacted policy and doctor recommendations.
APPROPRIATING VALUE FROM INFORMATION TECHNOLOGY IN HEALTHCARE
(2011) Goh, Jie Mein; Agarwal, Ritu; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
The value potential of information technology (IT) in healthcare settings has generated considerable optimism yet, significant questions remain unanswered. This dissertation employs the lens of social structure to investigate the value of information technology in healthcare situated in two distinct contexts: hospitals, that exemplify the traditional institutional form for the delivery of healthcare services, and online patient communities that represent new organizational forms enabled by IT. It seeks to address the following fundamental research questions "What is the impact of information technology in healthcare settings? How does social structure influence the appropriation of the value of information technology in healthcare?" Each of the two contexts is investigated in a separate essay, drawing upon distinct bodies of literature and using both qualitative and quantitative analytical methods. Essay 1: Evolving Work Routines: A Theory of Successful Adaptation to Information Technology in Healthcare The first essay investigates the impact of healthcare technologies such as electronic medical record systems in the traditional hospital environment. It traces the development of changes in social structure before and after an IT implementation. Using a longitudinal field study, the process of how information technology and routines interact is deconstructed. A theory of the co-evolution of routines and technology is proposed and described. Essay 2: The Social Value of Online Health Communities The second essay examines the impact of health information technology in the form of online patient communities by uncovering the social structure of the community. Using data collected from a popular online patient community, I identify the generative processes using support patterns between patients within the community. I find that online patient communities yield social value through information and emotional support to patients by enabling the transfer of support between patients with differential needs. Results also provide descriptive insights into the attributes of patients that contribute to variation in the provision of support within such online patient communities. The two studies in this dissertation make theoretical and empirical contributions. They shed light on the impact of information technology in healthcare, and further inform us about the appropriation of HIT value from a social structure perspective.
ASSESSING QUALITY IN HIGH-UNCERTAINTY MARKETS: ONLINE REVIEWS OF CREDENCE SERVICES
(2016) Lantzy, Shannon; Stewart, Katherine; Viswanathan, Siva; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In economics of information theory, credence products are those whose quality is difficult or impossible for consumers to assess, even after they have consumed the product (Darby & Karni, 1973). This dissertation is focused on the content, consumer perception, and power of online reviews for credence services. Economics of information theory has long assumed, without empirical confirmation, that consumers will discount the credibility of claims about credence quality attributes. The same theories predict that because credence services are by definition obscure to the consumer, reviews of credence services are incapable of signaling quality. Our research aims to question these assumptions. In the first essay we examine how the content and structure of online reviews of credence services systematically differ from the content and structure of reviews of experience services and how consumers judge these differences. We have found that online reviews of credence services have either less important or less credible content than reviews of experience services and that consumers do discount the credibility of credence claims. However, while consumers rationally discount the credibility of simple credence claims in a review, more complex argument structure and the inclusion of evidence attenuate this effect. In the second essay we ask, “Can online reviews predict the worst doctors?” We examine the power of online reviews to detect low quality, as measured by state medical board sanctions. We find that online reviews are somewhat predictive of a doctor’s suitability to practice medicine; however, not all the data are useful. Numerical or star ratings provide the strongest quality signal; user-submitted text provides some signal but is subsumed almost completely by ratings. Of the ratings variables in our dataset, we find that punctuality, rather than knowledge, is the strongest predictor of medical board sanctions. These results challenge the definition of credence products, which is a long-standing construct in economics of information theory. Our results also have implications for online review users, review platforms, and for the use of predictive modeling in the context of information systems research.
A Configurational Approach to Examining the Influence of Information Technology Management and Governance on Organization Performance
(2019) Aljazzaf, Salman; Mithas, Sunil; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Information technology (IT) is becoming an increasingly crucial part of modern organizations. This dissertation includes two essays that examine how effective IT management and decision-making structure are associated with better organizational performance. The first essay examines the complementarity between IT management and human resource (HR) management capabilities and discusses the mechanisms through which these two capabilities jointly lead to better organizational performance. The unique contribution of this study is the use of direct measures of IT management and HR management capabilities to estimate their joint impact on organizational performance. Furthermore, I disaggregate HR capability into two specific dimensions: (1) work systems such as employee performance management systems and hiring and promotion systems, and (2) employee learning and development. The main results confirm the complementarity between IT management and both HR management dimensions, and show that work systems more positively moderates the impact of IT management on organizational performance based on financial and market measures. The study is supplemented with a configurational analysis that examines the complex relationships between the organizational capabilities and explain how the complementarity between IT management, work systems, and employee learning varies across sectors and relies also on the presence and absence of other capabilities such as leadership and strategic planning. The study compares the results of the conventional and configurational methods and highlights the unique insights derived from each approach. The second essay discusses the optimal IT reporting structure in a firm, that is, whether the IT head should report to the chief executive officer or some other executive. This study proposes that there are several factors that determine the optimal IT reporting structure such as firm size, industry, IT investment intensity, and whether IT is viewed as strategic to the firm. The study argues that the relationship between these factors and the optimal IT reporting structure is too complex to be represented by linear models that rely on the correlation-based approach. Instead, there is a need to study configurations that lead to better performance based on different combinations of firm-level and industry-level conditions. The study uses a novel configurational approach and a corresponding method, the fuzzy-set qualitative comparative analysis, to determine the optimal IT reporting structure of different configurations. The study results shed light on the complex relationship between IT reporting structure and the conditions defining various firm configurations. Together the two essays provide new insights on how successful IT management and governance structure lead to organizational success.
Coordinating Demand Fulfillment With Supply Across A Dynamic Supply Chain
(2006-04-25) Chen, Maomao; Ball, Michael; Decision and Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Today, technology enables companies to extend their reach in managing the supply chain and operating it in a coordinated fashion from raw materials to end consumers. Order promising and order fulfillment have become key supply chain capabilities which help companies win repeat business by promising orders competitively and reliably. In this dissertation, we study two issues related to moving a company from an Available to Promise (ATP) philosophy to a Profitable to Promise (PTP) philosophy: pseudo order promising and coordinating demand fulfillment with supply. To address the first issue, a single time period analytical ATP model for n confirmed customer orders and m pseudo orders is presented by considering both material constraints and production capacity constraints. At the outset, some analytical properties of the optimal policies are derived and then a particular customer promising scheme that depends on the ratio between customer service level and profit changes is presented. To tackle the second issue, we create a mathematical programming model and explore two cases: a deterministic demand curve or stochastic demand. A simple, yet generic optimal solution structure is derived and a series of numerical studies and sensitivity analyses are carried out to investigate the impact of different factors on profit and fulfilled demand quantity. Further, the firm's optimal response to a one-time-period discount offered by the supplier of a key component is studied. Unlike most models of this type in the literature, which define variables in terms of single arc flows, we employ path variables to directly identify and manipulate profitable and non-profitable products. Numerical experiments based on Toshiba's global notebook supply chain are conducted. In addition, we present an analytical model to explore balanced supply. Implementation of these policies can reduce response time and improve demand fulfillment; further, the structure of the policies and our related analysis can give managers broad insight into this general decision-making environment.
A DATA ANALYTICAL FRAMEWORK FOR IMPROVING REAL-TIME, DECISION SUPPORT SYSTEMS IN HEALTHCARE
(2010) Yahav, Inbal; Shmeuli, Galit; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In this dissertation we develop a framework that combines data mining, statistics and operations research methods for improving real-time decision support systems in healthcare. Our approach consists of three main concepts: data gathering and preprocessing, modeling, and deployment. We introduce the notion of offline and semi-offline modeling to differentiate between models that are based on known baseline behavior and those based on a baseline with missing information. We apply and illustrate the framework in the context of two important healthcare contexts: biosurveillance and kidney allocation. In the biosurveillance context, we address the problem of early detection of disease outbreaks. We discuss integer programming-based univariate monitoring and statistical and operations research-based multivariate monitoring approaches. We assess method performance on authentic biosurveillance data. In the kidney allocation context, we present a two-phase model that combines an integer programming-based learning phase and a data-analytical based real-time phase. We examine and evaluate our method on the current Organ Procurement and Transplantation Network (OPTN) waiting list. In both contexts, we show that our framework produces significant improvements over existing methods.
Data Analytics and Mathematical Models to Facilitate Disease Prevention in the U.S.
(2020) Apergi, Lida Anna; Baras, John; Golden, Bruce; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
The U.S. is leading in healthcare expenditures worldwide, but health outcomes in the U.S. are not reflective of the level of spending. Prevention plays a crucial role in improving the health of individuals in the U.S., since it helps people live longer and healthier lives. Preventive services include actions that prevent diseases from ever occurring, detect diseases at an early stage, and manage diseases that have already been diagnosed. In this dissertation, we use data analytics and mathematical modeling techniques to better understand factors that influence disease prevention and help provide efficient solutions. In the first part of this dissertation, we study two problems of disease prevention at the public health level. First, we investigate the impact of state-level vaccination exemption policy and of the highly publicized Disneyland measles outbreak on MMR vaccination rates of young children. At the same time, we highlight the impact that the choice of socioeconomic factors can have on measurement results. We estimate the impact of these policies using multiple linear regression. Furthermore, we study the sensitivity of the results by examining a number of different approaches for the selection of socioeconomic control variables. Second, we utilize big data to estimate the additive cost of chronic diseases and study their cost patterns. We model the cost based on a cost hierarchy; that is, the cost of each condition is modeled as a function of the number of other more expensive chronic conditions the individual has. Using large scale claims data, we identify members that suffer from one or more chronic conditions and estimate their healthcare expenditures. Through our analysis, we categorize the chronic conditions into different expenditure groups based on the characteristics of their cost profiles. In the second part of this dissertation, we study two problems of disease prevention at the healthcare provider level, focusing in the area of cardiology. First, we study the adoption of conversational agent technology by patients with heart failure. Conversational agents can help patients with heart failure to manage their condition and provide frequent feedback to their healthcare providers. We analyze data from two studies, with each study focusing on a different type of conversational agent. We compare the two types of conversational agent technologies in terms of patient engagement, and investigate which patient characteristics are important in determining the patient engagement. Second, we tackle the problem of outpatient scheduling in the cardiology department of a large medical center. The outpatients have to go through a number of diagnostic tests and treatments before they can complete the final procedure. We develop an integer programming model to schedule appointments that are convenient for the outpatients by minimizing the number of visits that the patients have to make to the hospital and the time they spend waiting in the hospital. Furthermore, we investigate whether scheduling outpatients in groups can lead to better schedules for the patients.
DATA VISUALIZATION OF ASYMMETRIC DATA USING SAMMON MAPPING AND APPLICATIONS OF SELF-ORGANIZING MAPS
(2005-03-17) Li, Haiyan; Golden, Bruce L.; Decision and Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Data visualization can be used to detect hidden structures and patterns in data sets that are found in data mining applications. However, although efficient data visualization algorithms to handle data sets with asymmetric proximities have been proposed, we develop an improved algorithm in this dissertation. In the first part of the proposal, we develop a modified Sammon mapping approach that uses the upper triangular part and the lower triangular part of an asymmetric distance matrix simultaneously. Our proposed approach is applied to two asymmetric data sets: an American college selection data set, and a Canadian college selection data set which contains rank information. When compared to other approaches that are used in practice, our modified approach generates visual maps that have smaller distance errors and provide more reasonable representations of the data sets. In data visualization, self-organizing maps (SOM) have been used to cluster points. In the second part of the proposal, we assess the performance of several software implementations of SOM-based methods. Viscovery SOMine is found to be helpful in determining the number of clusters and recovering the cluster structure of data sets. A genocide and politicide data set is analyzed using Viscovery SOMine, followed by another analysis on the public and private college data sets with the goal to find out schools with best values.
DATA-DRIVEN ESSAYS: ROLE OF PRICING AND RETURNS
(2022) Hemmati, Sahar; Elmaghraby, Wedad J; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In this dissertation, we identify leavers in retailers’ operations such as their shipping policy design as well as markdown strategy which can significantly impact their product returns and the costs associated with it. We study how a high free shipping thresholds can induce shoppers to place orders with an intention to return a part of the order later. Using an empirical approach, which exploits natural experiments, we confirm that high threshold and/or high shipping fees induce a substantial order padding behavior, which leads to lower sales revenue, after adjusting for returns. However, we find that such behavior can be largely alleviated with frictions to returns. We propose that retailers looking to design their shipping policy should correctly account for return environment features. We also explore the link between a retailer’s markdown pricing strategies and its impact on customers’ purchase and return behavior. We specifically study the distinction between regular price markdowns and bundle price markdowns and the key contributors to such distinctions, and how they contribute to sales and customers’ return behavior. Capturing this notion and the heterogeneous impact of bundle and regular discounts on merchandise with different substitutability or complementarity as well as correlation between items at return stage, we offer recommendations on how a retailer should approach the design of their markdown strategies.
DATA-DRIVEN OPTIMIZATION AND STATISTICAL MODELING TO IMPROVE DECISION MAKING IN LOGISTICS
(2019) Sinha Roy, Debdatta; Golden, Bruce; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In this dissertation, we develop data-driven optimization and statistical modeling techniques to produce practically applicable and implementable solutions to real-world logistics problems. First, we address a significant and practical problem encountered by utility companies. These companies collect usage data from meters on a regular basis. Each meter has a signal transmitter that is automatically read by a receiver within a specified distance using radio-frequency identification (RFID) technology. The RFID signals are discontinuous, and each meter differs with respect to the specified distance. These factors could lead to missed reads. We use data analytics, optimization, and Bayesian statistics to address the uncertainty. Second, we focus on an important problem experienced by delivery and service companies. These companies send out vehicles to deliver customer products and provide services. For the capacitated vehicle routing problem, we show that reducing route-length variability while generating the routes is an important consideration to minimize the total operating and delivery costs for a company when met with random traffic. Third, we address a real-time decision-making problem experienced in practice. In one application, routing companies participating in competitive bidding might need to respond to a large number of requests regarding route costs in a very short amount of time. In another application, during post-disaster aerial surveillance planning or using drones to deliver emergency medical supplies, route-length estimation would quickly need to assess whether the duration to cover a region of interest would exceed the drone battery life. For the close enough traveling salesman problem, we estimate the route length using information about the instances. Fourth, we address a practical problem encountered by local governments. These organizations carry out road inspections to decide which street segments to repair by recording videos using a camera mounted on a vehicle. The vehicle taking the videos needs to proceed straight or take a left turn to cover an intersection fully. Right turns and U-turns do not capture an intersection fully. We introduce the intersection inspection rural postman problem, a new variant of the rural postman problem involving turns. We develop two integer programming formulations and three heuristics to generate least-cost vehicle routes.
Decisions under Uncertainty in Decentralized Online Markets: Empirical Studies of Peer-to-Peer Lending and Outsourcing
(2010) Lin, Mingfeng; Viswanathan, Siva; Lucas, Hank; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Recent developments in information technologies, especially Web 2.0 technologies, have radically transformed many markets through disintermediation and decentralization. Lower barriers of entry in these markets enable small firms and individuals to engage in transactions that were otherwise impossible. Yet, the issues of informational asymmetry that plague traditional markets still arise, only to be exacerbated by the "virtual" nature of these marketplaces. The three essays of my dissertation empirically examine how participants, many of whom are entrepreneurs, tackle the issue of asymmetric information to derive benefits from trade in two different contexts. In Essay 1, I investigate the role of online social networks in mitigating information asymmetry in an online peer-to-peer lending market, and find that the relational dimensions of these networks are especially effective for this purpose. In Essay 2, I exploit a natural experiment in the same marketplace to study the effect of shared geographical ties on investor decisions, and find that "home bias" is not only robust but also has an interesting interaction pattern with rational decision criteria. In Essay 3, I study how the emergence of new contract forms, enabled by new monitoring technologies, changes the effectiveness of traditional signals that affect a buyers' choice of sellers in online outsourcing. Using a matched-sample approach, I show that the effectiveness of online ratings and certifications differs under pay-for-time contracts versus pay-for-deliverable contracts. In all, the three essays of my dissertation present new empirical evidence of how agents leverage various network ties, signals and incentives to facilitate transactions in decentralized online markets, form transactional ties, and reap the benefits enabled by the transformative power of information technologies.
Design and Operations on the Supply Side of Online Marketplaces
(2020) Zhang, Wenchang; Elmaghraby, Wedad J; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Online platforms like eBay, Upwork, Airbnb, and Uber have transformed their markets, and many more are about to emerge. The rise of platforms has become one of the predominant economic and social developments of our time. Moreover, it has created many opportunities and challenges for both practitioners and researchers. My dissertation focuses on the design and operations on the supply side of online marketplaces. In particular, I study supply-side levers (e.g., listing policy and information provision policy) in different marketplace context (e.g., auction marketplace and service platform), with the consideration of strategic behavior of market participants and various friction involved in transactions (e.g., participation cost, information asymmetry, and supply adjustment friction). The first essay investigates how a one-sided liquidation auction marketplace maximizes its revenue by managing the supply-side market thickness under an exogenous supply inflow. The second essay examines the operational impacts of service platforms’ information disclosure regarding service providers’ qualities and revealing their mechanisms. The last essay studies whether two-sided marketplaces benefit or suffer from sellers’ quantity competition under unanticipated demand shocks. We further show that marketplaces can maneuver the competition in favorable directions by manipulating the supply adjustment friction. Overall, the findings from the three essays show that marketplaces’ operational levers on the supply side have significant effects on the strategies of all participants, which impacts the marketplaces’ operational performance. The dissertation offers both theoretical insights on the mechanisms of the studied supply-side levers and practical implications on how these levers should be designed and implemented.
Design of Online Auction System with Alternative Currencies
(2004-05-05) Deshpande, Vainateya Suresh; Lucas, Henry; Decision and Information Technologies
The University of Maryland has one of the most popular Basketball programs in the region. About 35,000 students seek 4,000 free student tickets allocated for every home game. An auction-based system provides a procedure to achieve and equitable and fair distribution of a high-demand resource. In an auction-based system, goods being sold end up with the person who values them the most. This is a very desirable scenario for a ticket distribution system that aims at maximizing attendance for home games. People who bid high have high values for the tickets and are more likely to attend a game than someone who receives a ticket through a random draw. The thesis lays out the framework for an auction based system to distribute home game tickets.
DESIGNING INFORMATION STRATEGIES FOR DIGITAL PLATFORMS: FINDINGS FROM LARGE-SCALE RANDOMIZED FIELD EXPERIMENTS
(2019) Shi, Lanfei; Viswanathan, Siva; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
The rise of digital platforms has transformed our economy and reshaped consumer behaviors and experiences. While practitioners and researchers have a growing interest in understanding digital platforms, there is still a dearth of research on how platforms can design effective information strategies to mitigate fundamental issues such as information asymmetry and search frictions by leveraging granular data. My dissertation seeks to fill this gap. Specifically, by focusing on significant real-world problems on digital platforms, I aim to examine IT-enabled and analytics-driven information strategies and study the impact of these strategies on the users as well as on the platforms themselves. In collaboration with two different online platforms, I design and conduct three randomized field experiments to investigate the impact of informational interventions and provide actionable suggestions. In Essay 1, I examine incentive strategies for motivating effective mobile app adoptions, by comparing monetary incentives against informational incentives. I find that the usage after app adoption depends on how customers are motivated, and only information induced adoption leads to long-term increase in purchases. In Essay 2, I investigate the role of “verification” when it is made optional, and find that it serves as a very effective signaling device, especially in markets that lack other mechanisms such as reputation systems. I also find that users on the two sides of online platform use the same signal very differently, and that this is attributable to the difference in the credibility of their primary signaling-attribute of each side, viz. income in males and beauty in females. In Essay 3, I examine the effectiveness of three different recommendation systems in two-sided matching platforms with a focus on how the provisioning of potential candidates’ preference information impacts focal user’s decision-making and matching outcomes. I find that compared to “people you might prefer”, users act strategically towards “people who might prefer you” and “people who you might prefer and who might prefer you” by actively reaching out to less desirable candidates, which leads to improved outcomes. In short, the three studies present new empirical evidence of how platforms can leverage information as a tool to design effective incentives, signaling mechanisms and recommender systems to facilitate users’ decision-making, transactions and matching.
DEVELOPING MULTIMODAL LEARNING METHODS FOR VIDEO UNDERSTANDING
(2024) Sun, Mingwei; Zhang, Kunpeng; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In recent years, the field of deep learning, with a particular emphasis on multimodal representation learning, has experienced significant advancements. These advancements are largely attributable to groundbreaking progress in areas such as computer vision, voice recognition, natural language processing, and graph network learning. This progress has paved the way for a multitude of new applications. The domain of video, in particular, holds immense potential. Video is often considered the most potent form of digital content for communication and the dissemination of information. The ability to effectively and efficiently comprehend video content could prove instrumental in a variety of downstream applications. However, the task of understanding video content presents numerous challenges. These challenges stem from the inherently unstructured and complex nature of video, as well as its interactions with other forms of unstructured data, such as text and network data. These factors contribute to the difficulty of video analysis. The objective of this dissertation is to develop deep learning methodologies capable of understanding video across multiple dimensions. Furthermore, these methodologies aim to offer a degree of interpretability, which could yield valuable insights for researchers and content creators. These insights could have significant managerial implications.In the first study, I introduce an innovative network based on Long Short-Term Memory (LSTM), enhanced with a Transformer co-attention mechanism, designed for the prediction of apparent emotion in videos. Each video is segmented into clips of one-second duration, and pre-trained ResNet networks are employed to extract audio and visual features at the second level. I construct a co-attention Transformer to effectively capture the interactions between the audio and visual features that have been extracted. An LSTM network is then utilized to learn the spatiotemporal information inherent in the video. The proposed model, termed the Sec2Sec Co-attention Transformer, outperforms several state-of-the-art methods in predicting apparent emotion on a widely recognized dataset: LIRIS-ACCEDE. In addition, I conduct an extensive data analysis to explore the relationships between various dimensions of visual and audio components and their influence on video predictions. A notable feature of the proposed model is its interpretability, which enables us to study the contributions of different time points to the overall prediction. This interpretability provides valuable insights into the functioning of the model and its predictions. In the second study, I introduce a novel neural network, the Multimodal Co-attention Transformer, designed for the prediction of personality based on video data. The proposed methodology concurrently models audio, visual, and text representations, along with their intra-relationships, to achieve precise and efficient predictions. The effectiveness of the proposed approach is demonstrated through comprehensive experiments conducted on a real-world dataset, namely, First Impressions. The results indicate that the proposed model surpasses state-of-the-art methods in performance while preserving high computational efficiency. In addition to evaluating the performance of the proposed model, I also undertake a thorough interpretability analysis to examine the contribution across different levels. The insights gained from the findings offer a valuable understanding of personality predictions. Furthermore, I illustrate the practicality of video-based personality detection in predicting outcomes of MBA admissions, serving as a decision support system. This highlights the potential importance of the proposed approach for both researchers and practitioners in the field. In the third study, I present a novel generalized multimodal learning model, termed VAN, which excels in learning a unified representation of \textbf{v}isual, \textbf{a}coustic, and \textbf{n}etwork cues. Initially, I utilize state-of-the-art encoders to model each modality. To augment the efficiency of the training process, I adopt a pre-training strategy specifically designed to extract information from the music network. Subsequently, I propose a generalized Co-attention Transformer network. This network is engineered to amalgamate the three distinct types of information and to learn the intra-relationships that exist among the three modalities, a critical facet of multimodal learning. To assess the effectiveness of the proposed model, I collect a real-world dataset from TikTok, comprising over 88,000 videos. Extensive experiments demonstrate that the proposed model surpasses existing state-of-the-art models in predicting video popularity. Moreover, I have conducted a series of ablation studies to attain a deeper comprehension of the behavior of the proposed model. I also perform an interpretability analysis to study the contributions of each modality to the model performance, leveraging the unique property of the proposed co-attention structure. This research contributes to the field by proffering a more comprehensive approach to predicting video popularity on short-form video platforms.
Dual-Based Local Search for Deterministic, Stochastic and Robust Variants of the Connected Facility Location Problem
(2011) Bardossy, Maria G.; Raghavan, Subramanian; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In this dissertation, we propose the study of a family of network design problems that arise in a wide range of practical settings ranging from telecommunications to data management. We investigate the use of heuristic search procedures coupled with lower bounding mechanisms to obtain high quality solutions for deterministic, stochastic and robust variants of these problems. We extend the use of well-known methods such as the sample average approximation for stochastic optimization and the Bertsimas and Sim approach for robust optimization with heuristics and lower bounding mechanisms. This is particular important for NP-complete problems where even deterministic and small instances are difficult to solve to optimality. Our extensions provide a novel way of applying these techniques while using heuristics; which from a practical perspective increases their usefulness.

Browse

Browsing Decision, Operations & Information Technologies by Title

Results Per Page

Sort Options