Search

In 2018, the Department of Operations, Business Analytics, and Information Systems (OBAIS) in the Carl H. Lindner College of Business at the University of Cincinnati (UC) celebrated its 50th anniversary, and in 2019 the OBAIS department won the INFORMS UPS George D. Smith Prize. The OBAIS department has a long history of excellence in fostering academia-industry collaboration in the area of analytics as well as a track record of continued innovation. In this article, we summarize some of the history of the OBAIS department and describe many of the department’s innovations that enabled the department to win the 2019 Smith Prize. We provide an overview of the department’s curriculum of analytics-focused degree programs, and we explain how the UC Center for Business Analytics serves as the culmination of the department’s efforts to provide a supportive ecosystem in analytics for students, faculty, business leaders, and practitioners. We conclude with some lessons learned along our journey.

Scope Dietary fat composition is an important modulator of vascular function. Non-esterified fatty acids (NEFA) enriched in saturated fatty acids (SFA) are thought to reduce vascular reactivity by attenuating insulin signalling via vasodilator pathways (phosphoinositide 3-kinase (PI3K)/Akt/endothelial nitric oxide synthase (eNOS)) and enhancing signalling via pro-inflammatory pathways. Methods To examine the effects of fatty acids on these pathways, human aortic endothelial cells were incubated with single fatty acids, and mixtures of these fatty acids to mimic typical NEFA composition and concentrations achieved in our previous human study. RNA was extracted to determine gene expression using real-time RT-PCR and cell lysates prepared to assess protein phosphorylation by Western blotting. Results Oleic acid (OA, 100 µM) was shown to down regulate expression of the insulin receptor, PTEN and a PI3K catalytic (p110β) and regulatory (p85α) subunit compared to palmitic, linoleic and stearic acids (P 1177. Both concentration and composition of the SFA and SFA plus n-3 polyunsaturated fatty acids (PUFA) mixtures had significant effects on genes involved in the PI3K/Akt pathway. Greater up-regulation was found with 800 than 400 µM concentration (respective of concentrations in insulin resistant and normal individuals), whereas greater down-regulation was evident with SFA plus n-3 PUFA than SFA mixture alone. Conclusion Our findings provide novel insights into the modulation of the PI3K/Akt/eNOS pathway by single fatty acids and fatty acid mixtures. In particular, OA appears to promote signalling via this pathway, with further work required to determine the primary molecular site(s) of action.

The increasing vulnerability of the population from frequent disasters requires quick and effective responses to provide the required relief through effective humanitarian supply chain distribution networks. We develop scenario-robust optimization models for stocking multiple disaster relief items at strategic facility locations for disaster response. Our models improve the robustness of solutions by easing the difficult, and usually impossible, task of providing exact probability distributions for uncertain parameters in a stochastic programming model. Our models allow decision makers to specify uncertainty parameters (i.e., point and probability estimates) based on their degrees of knowledge, using distribution-free uncertainty sets in the form of ranges. The applicability of our generalized approach is illustrated via a case study of hurricane preparedness in the Southeastern United States. In addition, we conduct simulation studies to show the effectiveness of our approach when conditions deviate from the model assumptions.

The phosphatidylinositol (PI) cycle is central to eukaryotic cell signaling. Its complexity, due to the number of reactions and lipid and inositol phosphate intermediates involved makes it difficult to analyze experimentally. Computational modelling approaches are seen as a way forward to elucidate complex biological regulatory mechanisms when this cannot be achieved solely through experimental approaches. Whilst mathematical modelling is well established in informing biological systems, many models are often informed by data sourced from different cell types (mosaic data), to inform model parameters. For instance, kinetic rate constants are often determined from purified enzyme datain vitroor use experimental concentrations obtained from multiple unrelated cell types. Thus they do not represent any specific cell type nor fully capture cell specific behaviours. In this work, we develop a model of the PI cycle informed byin-vivoomics data taken from a single cell type, namely platelets. Our model recapitulates the known experimental dynamics before and after stimulation with different agonists and demonstrates the importance of lipid- and protein-binding proteins in regulating second messenger outputs. Furthermore, we were able to make a number of predictions regarding the regulation of PI cycle enzymes and the importance of the number of receptors required for successful GPCR signaling. We then consider how pathway behavior differs, when fully informed by data for HeLa cells and show that model predictions remain relatively consistent. However, when informed by mosaic experimental data model predictions greatly vary. Our work illustrates the risks of using mosaic datasets from unrelated cell types which leads to over 75% of outputs not fitting with expected behaviors.Authors summaryComputational models of cellular complexity offer much in terms of understanding cell behaviors and in informing experimental design, but their usefulness is limited in them being built with mosaic data not representing specific cell types and tested against limited experimental outputs. In this work we demonstrate an approach using quantitative proteomic datasets and temporal experimental data from a single cell type (platelets) to inform kinetic rate constants and protein concentrations for a mathematical model of a key signaling pathway - the phosphatidylinositol (PI) cycle; known for its central role in numerous cell functions and diseases. After using our model to make novel predictions regarding how aspects of the pathway are regulated, we demonstrate its versatile nature by utilising proteomic data from other cell types to generate similar predictions for those cells while highlighting the pitfalls of using mosaic data when constructing computational models.

In a tele-ICU center with flexible staff scheduling, interruptions to tasks with lower priority are common, often prolonging the processing time of these tasks. Dedicated staff scheduling, however, can reduce such interruptions. While dedicated staff scheduling is generally perceived to be less efficient than flexible systems according to traditional queueing theory, the reductions in interruptions call this result into question. To investigate when dedicated staff scheduling could be more efficient than flexible systems for tele-ICUs, we build a discrete-event simulation model to analyze the operational process of admissions and interventions at a real tele-ICU center. Our simulation results show that if the dedicated nurse (intensivist) can process admissions at least 31% (26%) faster than the flexible nurse (intensivist), the dedicated staff scheduling will be preferred. Thus, the tele-ICU administrators should choose the scheduling approach (flexible versus dedicated) based on how often interruptions are likely to occur and their anticipated effect on prolonging the medical staff’s processing times of the interrupted tasks. Our simulation model can also help tele-ICU administrators determine a suitable staffing level by providing the estimated waiting times of admissions and interventions under varying staffing levels.

We propose a heuristic approach to address the general class of optimization problems involving the capacitated clustering of observations consisting of variable values that are realizations from respective probability distributions. Based on the expectation-maximization algorithm, our approach unifies Gaussian mixture modeling for clustering analysis and cluster capacity constraints using a posterior regularization framework. To test our algorithm, we consider the capacitated p-median problem in which the observations consist of geographic locations of customers and the corresponding demand of these customers. Our heuristic has superior performance compared to classic geometrical clustering heuristics, with robust performance over a collection of instance types.

This work investigates how fairness concerns influence supply-chain decision making, while examining the effect of private production-cost information and touching on issues related to bounded rationality. We conduct laboratory work utilizing a supply-chain dyad with an upstream supplier feeding a downstream retailer under a simple wholesale-price contract. We perform human–computer (H–C) experiments where human subjects play the role of the supplier paired with the computerized retailer, as well as human–human (H–H) experiments where human subjects play the role of both supplier and retailer. These experiments allow us to isolate other effects like bounded rationality from the effects of fairness concerns on supply-chain decision making. We find that, compared to standard analytical model, the bounded rationality slightly reduces overall supply chain profit without changing its distribution between the supplier and the retailer, while fairness concerns lead to greater supply-chain profits and a more balanced supply-chain profit distribution. We further illustrate that under private cost information, the retailer's fairness concern is suppressed by the lack of reciprocity from not being able to observe her rival's profit information, but that the supplier's fairness concern from altruism persists. Based on our experimental results, we modify classical supply-chain models to include utility functions that incorporate both bounded rationality and fairness concerns. The estimated other-regarding coefficients are significantly lower under private information than under public information for the H–H experiments, and we find no evidence of inequity aversion for the H–C experiments.

In this paper we propose a framework for shift-level container scheduling and resource allocation decisions at a cross-dock facility. The Multi-Mode Resource-Constrained Cross-Dock Scheduling Problem (MRCDSP) approach minimizes material flow and schedules inbound and outbound containers to dock-doors such that the total processing time is minimized subject to the resource constraints at the cross-dock. While container scheduling and resource allocation problems at cross-dock facilities have been studied previously in isolation, our work is the first to consider a complete view of cross-dock operations providing optimal container to dock-door allocation, and a makespan minimizing schedule of containers to the cross-dock. We present a comprehensive framework that includes identification of container clusters to reduce the problem size, a container-to-dock-door assignment algorithm, and a container clusters scheduling model that is solvable for practically sized problems. In a comparative numeric study based on data simulating a cross-dock facility, our approach is shown to outperform current practice, reducing the average time required for processing a set of containers by 37% and reducing the weighted-distance material traveled within the cross-dock by 45%.

Systems biology and modeling approaches require quantitative data-rich temporal experiments to better understand the dynamics and regulation of the components of the signaling pathways that governs cell biology and physiology. Here we present a modified Western blotting method to rapidly analyze and accurately quantify protein copy number, and their respective phosphorylation states at specific sites over detailed time courses.

Systems biology and modeling approaches require quantitative data-rich temporal experiments to better understand the dynamics and regulation of the components of the signaling pathways that governs cell biology and physiology. Here we present a modified Western blotting method to rapidly analyze and accurately quantify protein copy number, and their respective phosphorylation states at specific sites over detailed time courses.

In this paper we analyse resource allocation decisions in a Long-Term Home Care (LTHC) program. LTHC programs provide various long-term in-home care services to patients who have been discharged fr...

Intensive Care Units (ICUs) are frequently the bottleneck in a hospital system, limiting patient flow and negatively impacting profits. This article examines admission control policies for a surgical ICU where patients arrive in batches. This problem is formulated as a Markov Decision Process (MDP) with an objective function that allows for varying degrees of emphasis on efficiency versus equity. Equity concerns are driven by a combination of surgery type and operating surgeon and are captured in a robust manner in the proposed models. A simple and efficient heuristic solution method related to our MDP formulation is proposed that provides a performance guarantee. The proposed admissions policy is applied to a real setting motivated by the cardiothoracic surgical ICU at Mount Sinai Medical Center in New York; the results demonstrate that the ICU can achieve large equity gains with no efficiency losses.

Supplemental Digital Content is available in the text., Objectives/Design: Little is known on how to best prioritize various tele-ICU specific tasks and workflows to maximize operational efficiency. We set out to: 1) develop an operational model that accurately reflects tele-ICU workflows at baseline, 2) identify workflow changes that optimize operational efficiency through discrete-event simulation and multi-class priority queuing modeling, and 3) implement the predicted favorable workflow changes and validate the simulation model through prospective correlation of actual-to-predicted change in performance measures linked to patient outcomes. Setting: Tele-ICU of a large healthcare system in New York State covering nine ICUs across the spectrum of adult critical care. Patients: Seven-thousand three-hundred eighty-seven adult critically ill patients admitted to a system ICU (1,155 patients pre-intervention in 2016Q1 and 6,232 patients post-intervention 2016Q3 to 2017Q2) Interventions: Change in tele-ICU workflow process structure and hierarchical process priority based on discrete-event simulation. Measurements and Main Results: Our discrete-event simulation model accurately reflected the actual baseline average time to first video assessment by both the tele-ICU intensivist (simulated 132.8 ± 6.7 min vs 132 ± 12.2 min actual) and the tele-ICU nurse (simulated 128.4 ± 7.6 min vs 123 ± 9.8 min actual). For a simultaneous priority and process change, the model simulated a reduction in average TVFA to 51.3 ± 1.6 min (tele-ICU intensivist) and 50.7 ± 2.1 min (tele-ICU nurse), less than the added simulated reductions for each change alone, suggesting correlation of the changes to some degree. Subsequently implementing both changes simultaneously resulted in actual reductions in average time to first video assessment to values within the 95% CIs of the simulations (50 ± 5.5 min for tele-intensivists and 49 ± 3.9 min for tele-nurses). Conclusions: Discrete-event simulation can accurately predict the effects of contemplated multidisciplinary tele-ICU workflow changes. The value of workflow process and task priority modeling is likely to increase with increasing operational complexities and interdependencies.

We apply service-operations-management concepts to improve the efficiency and equity of voting systems. Recent elections in the United States and elsewhere have been plagued by long lines, excessive waiting times, and perceptions of unfairness. We build models for the waiting lines at voting precincts using both traditional steady-state queueing methods and simulation models. We develop solution methods to allocate voting machines optimally to precincts. Our objective functions consider both the efficiency and the equity of the voting system. We compare our allocation algorithm to several competing methods, including those used in practice. We examine several different strategies for improving voting operations on both the demand and the capacity side of voting systems, and we present a complete case study of applying our method to data from the 2008 election for Franklin County, Ohio. We conclude that our method is superior to existing polices in terms of efficiency and equity and that it is robust in terms of uncertainties regarding turnout rates on Election Day. We also suggest several operational improvements to the voting process drawn from the service-operations literature.

Providing equal access to public service resources is a fundamental goal of democratic societies. Growing research interest in public services (e.g., health care, humanitarian relief, elections) has increased the importance of considering objective functions related to equity. This article studies discrete resource allocation problems where the decision maker is concerned with maintaining equity between some defined subgroups of a customer population and where non-closed-form functions of equity are allowed. Simulation optimization techniques are used to develop rigorous algorithms to allocate resources equitably among these subgroups. The presented solutions are associated with probabilistic bounds on solution quality. A full-factorial experimental design demonstrates that the proposed algorithm outperforms competing heuristics and is robust over various inequity metrics. Additionally, the algorithm is applied to a case study of allocating voting machines to election precincts in Franklin County, Ohio. [...

Objective: To develop queuing and simulation-based models to understand the relationship between ICU bed availability and operating room schedule to maximize the use of critical care resources and minimize case cancellation while providing equity to patients and surgeons. Design: Retrospective analysis of 6-month unit admission data from a cohort of cardiothoracic surgical patients, to create queuing and simulation-based models of ICU bed flow. Three different admission policies (current admission policy, shortest-processing-time policy, and a dynamic policy) were then analyzed using simulation models, representing 10 yr worth of potential admissions. Important output data consisted of the “average waiting time,” a proxy for unit efficiency, and the “maximum waiting time,” a surrogate for patient equity. Setting: A cardiothoracic surgical ICU in a tertiary center in New York, NY. Patients: Six hundred thirty consecutive cardiothoracic surgical patients admitted to the cardiothoracic surgical ICU. Interventions: None. Measurements and Main Results: Although the shortest-processingtime admission policy performs best in terms of unit efficiency (0.4612 days), it did so at expense of patient equity prolonging surgical waiting time by as much as 21 days. The current policy gives the greatest equity but causes inefficiency in unit bed-flow (0.5033 days). The dynamic policy performs at a level (0.4997 days) 8.3% below that of the shortest-processing-time in average waiting time; however, it balances this with greater patient equity (maximum waiting time could be shortened by 4 days compared to the current policy). Conclusions: Queuing theory and computer simulation can be used to model case flow through a cardiothoracic operating room and ICU. A dynamic admission policy that looks at current waiting time and expected ICU length of stay allows for increased equity between patients with only minimum losses of efficiency. This dynamic admission policy would seem to be a superior in maximizing case-flow. These results may be generalized to other surgical ICUs. (Crit Care Med 2013; 41:414–422)