Real-Time and Data-Driven Operation Optimization and Knowledge Discovery for an Enterprise Information System

An enterprise information system (EIS) is an integrated data-applications platform characterized by diverse, heterogeneous, and distributed data sources. For many enterprises, a number of business processes still depend heavily on static rule-based methods and extensive human expertise. Enterprises are faced with the need for optimizing operation scheduling, improving resource utilization, discovering useful knowledge, and making data-driven decisions.

This thesis research is focused on real-time optimization and knowledge discovery that addresses workflow optimization, resource allocation, as well as data-driven predictions of process-execution times, order fulfillment, and enterprise service-level performance. In contrast to prior work on data analytics techniques for enterprise performance optimization, the emphasis here is on realizing scalable and real-time enterprise intelligence based on a combination of heterogeneous system simulation, combinatorial optimization, machine-learning algorithms, and statistical methods.

On-demand digital-print service is a representative enterprise requiring a powerful EIS.We use real-life data from Reischling Press, Inc. (RPI), a digit-print-service provider (PSP), to evaluate our optimization algorithms.

In order to handle the increase in volume and diversity of demands, we first present a high-performance, scalable, and real-time production scheduling algorithm for production automation based on an incremental genetic algorithm (IGA). The objective of this algorithm is to optimize the order dispatching sequence and balance resource utilization. Compared to prior work, this solution is scalable for a high volume of orders and it provides fast scheduling solutions for orders that require complex fulfillment procedures. Experimental results highlight its potential benefit in reducing production inefficiencies and enhancing the productivity of an enterprise.

We next discuss analysis and prediction of different attributes involved in hierarchical components of an enterprise. We start from a study of the fundamental processes related to real-time prediction. Our process-execution time and process status prediction models integrate statistical methods with machine-learning algorithms. In addition to improved prediction accuracy compared to stand-alone machine-learning algorithms, it also performs a probabilistic estimation of the predicted status. An order generally consists of multiple series and parallel processes. We next introduce an order-fulfillment prediction model that combines advantages of multiple classification models by incorporating flexible decision-integration mechanisms. Experimental results show that adopting due dates recommended by the model can significantly reduce enterprise late-delivery ratio. Finally, we investigate service-level attributes that reflect the overall performance of an enterprise. We analyze and decompose time-series data into different components according to their hierarchical periodic nature, perform correlation analysis,

and develop univariate prediction models for each component as well as multivariate models for correlated components. Predictions for the original time series are aggregated from the predictions of its components. In addition to a significant increase in mid-term prediction accuracy, this distributed modeling strategy also improves short-term time-series prediction accuracy.

In summary, this thesis research has led to a set of characterization, optimization, and prediction tools for an EIS to derive insightful knowledge from data and use them as guidance for production management. It is expected to provide solutions for enterprises to increase reconfigurability, accomplish more automated procedures, and obtain data-driven recommendations or effective decisions.

Why the World Suddenly Cares About Global Supply Chains

The global value chain (GVC) concept has gained popularity as a way to analyze the international expansion and geographical fragmentation of contemporary supply chains and value creation and capture therein. It has been used broadly in academic publications that examine a wide range of global industries, and by many of the international organizations concerned with economic development. This note highlights some of the main features of GVC analysis and discusses the relationship between the core concepts of governance and upgrading. The key dynamics of contemporary global supply chains and their implications for global production and trade are illustrated by: (1) the consolidation of global value chains and the new geography of value creation and capture, with an emphasis on China; (2) the key roles of global supermarkets and private standards in agri-food supply chains; and (3) how the recent economic crisis contributes to shifting end markets and the regionalization of value chains. It concludes with a discussion of the future direction of GVC analysis and a potential collaboration with supply chain researchers. © 2012 Institute for Supply Management, Inc.

Effect of primary care physicians' use of estimated glomerular filtration rate on the timing of their subspecialty referral decisions.

BACKGROUND: Primary care providers' suboptimal recognition of the severity of chronic kidney disease (CKD) may contribute to untimely referrals of patients with CKD to subspecialty care. It is unknown whether U.S. primary care physicians' use of estimated glomerular filtration rate (eGFR) rather than serum creatinine to estimate CKD severity could improve the timeliness of their subspecialty referral decisions. METHODS: We conducted a cross-sectional study of 154 United States primary care physicians to assess the effect of use of eGFR (versus creatinine) on the timing of their subspecialty referrals. Primary care physicians completed a questionnaire featuring questions regarding a hypothetical White or African American patient with progressing CKD. We asked primary care physicians to identify the serum creatinine and eGFR levels at which they would recommend patients like the hypothetical patient be referred for subspecialty evaluation. We assessed significant improvement in the timing [from eGFR < 30 to ≥ 30 mL/min/1.73m(2)) of their recommended referrals based on their use of creatinine versus eGFR. RESULTS: Primary care physicians recommended subspecialty referrals later (CKD more advanced) when using creatinine versus eGFR to assess kidney function [median eGFR 32 versus 55 mL/min/1.73m(2), p < 0.001]. Forty percent of primary care physicians significantly improved the timing of their referrals when basing their recommendations on eGFR. Improved timing occurred more frequently among primary care physicians practicing in academic (versus non-academic) practices or presented with White (versus African American) hypothetical patients [adjusted percentage(95% CI): 70% (45-87) versus 37% (reference) and 57% (39-73) versus 25% (reference), respectively, both p ≤ 0.01). CONCLUSIONS: Primary care physicians recommended subspecialty referrals earlier when using eGFR (versus creatinine) to assess kidney function. Enhanced use of eGFR by primary care physicians' could lead to more timely subspecialty care and improved clinical outcomes for patients with CKD.

The providing resources to enhance African American patients' readiness to make decisions about kidney disease (PREPARED) study: protocol of a randomized controlled trial.

BACKGROUND: Living related kidney transplantation (LRT) is underutilized, particularly among African Americans. The effectiveness of informational and financial interventions to enhance informed decision-making among African Americans with end stage renal disease (ESRD) and improve rates of LRT is unknown. METHODS/DESIGN: We report the protocol of the Providing Resources to Enhance African American Patients' Readiness to Make Decisions about Kidney Disease (PREPARED) Study, a two-phase study utilizing qualitative and quantitative research methods to design and test the effectiveness of informational (focused on shared decision-making) and financial interventions to overcome barriers to pursuit of LRT among African American patients and their families. Study Phase I involved the evidence-based development of informational materials as well as a financial intervention to enhance African American patients' and families' proficiency in shared decision-making regarding LRT. In Study Phase 2, we are currently conducting a randomized controlled trial in which patients with new-onset ESRD receive 1) usual dialysis care by their nephrologists, 2) the informational intervention (educational video and handbook), or 3) the informational intervention in addition to the option of participating in a live kidney donor financial assistance program. The primary outcome of the randomized controlled trial will include patients' self-reported rates of consideration of LRT (including family discussions of LRT, patient-physician discussions of LRT, and identification of a LRT donor). DISCUSSION: Results from the PREPARED study will provide needed evidence on ways to enhance the decision to pursue LRT among African American patients with ESRD.

Using Behavioral Insights For a Better Understanding of Economic Decisions (Regarding Savings, Inequality and Dishonesty)

This dissertation consists of three essays on behavioral economics, with a general aim of enriching our understanding of economic decisions using behavioral insights and experimental methodology. Each essay takes on one particular topic with this general aim.

The first chapter studies savings behavior of the poor. In this project, partnering with a savings product provider in Kenya, we tested the extent to which behavioral interventions and financial incentives can increase the saving rate of individuals with low and irregular income. Our experiment lasted for six months and included a total of twelve conditions. The control condition received weekly reminders and balance reporting via text messages. The treatment conditions received in addition one of the following interventions: (1) reminder text messages framed as if they came from the participant’s kid (2) a golden colored coin with numbers for each week of the trial, on which participants were asked to keep track of their weekly deposits (3) a match of weekly savings: The match was either 10% or 20% up to a certain amount per week. The match was either deposited at the end of each week or the highest possible match was deposited at the start of each week and was adjusted at the end. Among these interventions, by far the most effective was the coin: Those in the coin condition saved on average the highest amount and more than twice as those in the control condition. We hypothesize that being a tangible track-keeping object; the coin made subjects remember to save more often. Our results support the line of literature suggesting that saving decisions involve psychological aspects and that policy makers and product designers should take these influences into account.

The second chapter is related to views towards inequality. In this project, we investigate how the perceived fairness of income distributions depends on the beliefs about the process that generated the inequality. Specifically, we examine how two crucial features of this process affect fairness views: (1) Procedural justice - equal treatment of all, (2) Agency - one's ability to determine his/her income. We do this in a lab experiment by varying the equality of opportunity (procedural justice), and one's ability to make choices, which consequently influence subjects’ ability to influence their income (agency). We then elicit ex-post redistribution decisions of the earnings as a function of these two elements. Our results suggest both agency and procedural justice matter for fairness. Our main findings can be summarized as follows: (1) Highlighting the importance of agency, we find that inequality resulting from risk is considered to be fair only when risk is chosen freely; (2) Highlighting the importance of procedural justice, we find that introducing inequality of opportunity significantly increases redistribution, however the share of subjects redistributing none remain close to the share of subjects redistributing fully revealing an underlying heterogeneity in the population about how fairness views should account for inequality of opportunity.

The third chapter is on morality. In this project, we study whether religious rituals act as an internal reminder for basic moral principles and thus affect moral judgments. To this end, we conducted two survey experiments in Turkey and Israel to specifically test the effect of Ramadan and Yom Kippur. The results from the Turkish sample how that Ramadan has a significant effect on moral judgments to some extent for those who report to believe in God. Those who believe in God judged the moral acceptability of ten out of sixty one actions significantly differently in Ramadan, whereas those who reported not to believe in God significantly changed their judgments only for one action in Ramadan. Our results extends the hypothesis established by lab experiments that religious reminders have a significant effect on morality, by testing it in the field in the natural environment of religious rituals.

This thesis is part of a broader collaborative research agenda with both colleagues and advisors. The programming, analyses, and writing, as well as any errors in this work, are my own.

Essays on the Dynamic Decisions of Homeowners and Retailers

Urban problems have several features that make them inherently dynamic. Large transaction costs all but guarantee that homeowners will do their best to consider how a neighborhood might change before buying a house. Similarly, stores face large sunk costs when opening, and want to be sure that their investment will pay off in the long run. In line with those concerns, different areas of Economics have made recent advances in modeling those questions within a dynamic framework. This dissertation contributes to those efforts.

Chapter 2 discusses how to model an agent’s location decision when the agent must learn about an exogenous amenity that may be changing over time. The model is applied to estimating the marginal willingness to pay to avoid crime, in which agents are learning about the crime rate in a neighborhood, and the crime rate can change in predictable (Markovian) ways.

Chapters 3 and 4 concentrate on location decision problems when there are externalities between decision makers. Chapter 3 focuses on the decision of business owners to open a store, when its demand is a function of other nearby stores, either through competition, or through spillovers on foot traffic. It uses a dynamic model in continuous time to model agents’ decisions. A particular challenge is isolating the contribution of spillovers from the contribution of other unobserved neighborhood attributes that could also lead to agglomeration. A key contribution of this chapter is showing how we can use information on storefront ownership to help separately identify spillovers.

Finally, chapter 4 focuses on a class of models in which families prefer to live

close to similar neighbors. This chapter provides the first simulation of such a model in which agents are forward looking, and shows that this leads to more segregation than it would have been observed with myopic agents, which is the standard in this literature. The chapter also discusses several extensions of the model that can be used to investigate relevant questions such as the arrival of a large contingent high skilled tech workers in San Francisco, the immigration of hispanic families to several southern American cities, large changes in local amenities, such as the construction of magnet schools or metro stations, and the flight of wealthy residents from cities in the Rust belt, such as Detroit.

How did varying levels of intensity of the 2005 Pakistan earthquake differentially affect women’s fertility decisions and children’s health outcomes?

This paper uses a difference in difference model to investigate the impact of a large scale and high mortality 2005 earthquake in Pakistan on women’s fertility decisions and children’s health outcomes. Using a nationally representative, cross sectional DHS data from 2006 and geographical data from USGS, this paper investigates how variation in earthquake intensity levels can differentially impact total fertility for women and the likelihood of children suffering from diseases such as diarrhea, Acute Respiratory Infections (ARI) and fever. The post-earthquake results demonstrate a statistically significant increase in total fertility for areas closer to the epicenter of the earthquake, within a 100km radius of the rupture surface and at higher altitudes. Similarly, for children who were in-utero at the time of the earthquake, the probability of having early symptoms of ARI or fever was much smaller in lower earthquake intensity zones compared to the highest intensity zone.

Adults with autism spectrum disorders exhibit decreased sensitivity to reward parameters when making effort-based decisions.

BACKGROUND: Efficient effort expenditure to obtain rewards is critical for optimal goal-directed behavior and learning. Clinical observation suggests that individuals with autism spectrum disorders (ASD) may show dysregulated reward-based effort expenditure, but no behavioral study to date has assessed effort-based decision-making in ASD. METHODS: The current study compared a group of adults with ASD to a group of typically developing adults on the Effort Expenditure for Rewards Task (EEfRT), a behavioral measure of effort-based decision-making. In this task, participants were provided with the probability of receiving a monetary reward on a particular trial and asked to choose between either an "easy task" (less motoric effort) for a small, stable reward or a "hard task" (greater motoric effort) for a variable but consistently larger reward. RESULTS: Participants with ASD chose the hard task more frequently than did the control group, yet were less influenced by differences in reward value and probability than the control group. Additionally, effort-based decision-making was related to repetitive behavior symptoms across both groups. CONCLUSIONS: These results suggest that individuals with ASD may be more willing to expend effort to obtain a monetary reward regardless of the reward contingencies. More broadly, results suggest that behavioral choices may be less influenced by information about reward contingencies in individuals with ASD. This atypical pattern of effort-based decision-making may be relevant for understanding the heightened reward motivation for circumscribed interests in ASD.