Analytics Team Research Articles

Petroleum Engineering - The Best Profession

Special Section: The Value and Future of Petroleum Engineering Short of choosing a life partner, deciding on a profession may be the most important decision any of us ever makes. And there are many who would debate which of these takes precedence. I consider myself amazingly fortunate in having made perfect choices in both categories. Anyone who has met my wife, Dina, can vouch for the first one. This is my opportunity to talk about the second. When people ask me what I think about petroleum engineering as a profession, I answer, “Petroleum engineering is not just a great profession; it’s the best profession.” Here is why. Name another profession that is fun, intellectually and physically challenging, personally and financially rewarding, and critical to true global sustainability on both a macro and micro scale. I can’t think of one. Beyond the basics of skills, aptitude, education, and interest, the way we choose our professions has changed dramatically over the past 50 to 60 years. The goal for many of our parents, and even some of us, depending on our age, was to join a company where we could make a good living, rise in the ranks, and work from graduation to retirement. That world no longer exists. Today the question of what to consider when choosing a profession consumes magazines, books, and the internet. A 2018 online Harvard Business Review (HBR) article reported on what motivates people at work, beyond obvious basics such as salary, time off, and benefits. To find answers, HBR’s people analytics team collaborated with Facebook. Based on surveys conducted twice a year asking what employees value most, the team identified three big buckets of motivators: career, community, and cause. As the article said, “It turns out we’re all hoping to find a what, a who, and a why.” Career—The What According to HBR, career is about work. It’s about autonomy, about using your strengths, and promoting learning and development. It’s at the heart of intrinsic motivation. It’s what you do throughout your working life, and may span many jobs and many companies. Engineers share many traits, perhaps the most common of which is our love of solving problems. Petroleum engineering is the pinnacle of problem solving. For the 7.2 billion people on our planet, we have to solve the problem of how to deliver affordable energy from oil and gas, and how to do so safely and economically. We must solve the problem of operating in ways that don’t threaten our planet and its future generations. And, we must do all of this in a way that helps our employers achieve acceptable return on investment.

Abstract TP487: Data Dashboard: Making the Electronic Medical Record Work for You

Background: Stroke care is becoming increasingly complex in the 21st century. In order to assure optimal patient outcomes, our health network developed a stroke care pathway based on evidence-based practices and expert consensus. Overall goals for the project include increased adherence to practice standards and defined process metrics. This project is part of a larger institutional initiative for specific disease states. An integral aspect of the project was the development and implementation of a dashboard allowing for a global view of the stroke population as well as the ability to drill down to the patient or provider level. Purpose: The purpose of the dashboard is to use data to evaluate trends and develop data driven interventions. This tool is available to key stakeholders including administration, quality, program leadership and care providers. Methods: The pathway initiative is a collaboration between quality, clinical, and data analytic teams. It used a multidisciplinary approach and included input from services throughout the care continuum. Clinical pathways for stroke alert, reperfusion, neurointerventional care, and nursing care were developed and recently launched. The dashboard auto-populates from discreet fields within the electronic medical record. It is an informational tool that allows for data transparency through the use of interactive charts and graphs which provides temporal trending. Results: Utilization of this approach supports system level data trending and analysis. Preliminary results demonstrate improved provider engagement, reduction in variability in several care processes, and data transparency through accessible real-time reporting. Conclusions: Disease specific dashboards using automatic data feeds from the electronic medical record allow for evaluation of program efficiencies, population trends, and provider level data without additional manual abstraction or analysis. Evaluating data in real time allows for identification of variations in practice, real-time interventions targeting identified variances, as well as determining the effectiveness of interventions on quality and cost of care.

Pilot Study - Composite Risk Score for Pre-Transplant Risk Assessment in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

Objective An internal pre-transplant risk assessment tool or Risk Monitor was developed by the Baylor Blood & Marrow Transplant (BMT) Data Management and Analytics team to assist with optimization of patient selection. The tool assigns a total risk score for allogeneic transplants based on attributes such as disease status, disease risk index (DRI), patient age, Karnofsky performance status (KPS), hematopoietic cellular therapy co-morbidity index (HCT-CI), and CMV status, along with psychosocial factors such as caregiver support, coping skills or motivation, health literacy, medication compliance, socioeconomic viability, distance of residence from the transplant center, as well as others. A retrospective pilot study was conducted to predict which attributes have the highest correlation with the primary outcome variable, post-transplant patient mortality. Method The Risk Monitor was developed using Microsoft Access (Microsoft®, Redmond, WA). A score of 0 (no risk) or 1 (risk) was assigned for each variable and an aggregate score was calculated for each patient (Figure 1). The data was analyzed using Waikato Environment for Knowledge Analysis (Weka). Weka is an open-source data mining tool consisting of a collection of machine learning algorithms for data mining tasks. It contains tools for data preparation, classification, regression, clustering, association rules mining, and visualization. Results We evaluated the total risk score for 97 patients that had undergone an allogeneic stem cell transplant (SCT) at Baylor University Medical Center in 2016. The total risk score for the patients varied between 0 and 9. Weka provided us with a list of independent attributes with rankings based on the intensity of correlation with the dependent variable, early mortality. Evaluation of the data set suggests that one attribute (Total Score) has the highest correlation with mortality, while other individual attributes (KPS, CMV status, DRI, HCT-CI and psychosocial factors such as caregiver support, coping skills, and health literacy) demonstrated a moderate to high correlation (Figure 2). Conclusion The Risk Monitor can serve as a useful tool for screening adult patients prior to allogeneic transplant. In addition, the tool may help to identify patients who need additional screening and social support to minimize the possibility of early mortality after transplantation. The Access database has been transitioned to a web-based (SQL Server/PHP/JavaScript) application to overcome certain Access-related issues such as client software installation, file corruption and interface update issues. Furthermore, this will enable easy cross platform usage, enhance security, and allow more concurrent users.

A New Model for Integrating Big Data into Phases of Decision-Making Process

Big data is becoming a key factor within organizations since the application of Big Data in modern business provides information in real-time that allows organizations to take a faster and smarter decision. Therefore, big data is widely used in many domains of modern society to achieve progress in these domains. This study aims to harness the opportunity of Big Data to enhance the decision-making process in the organization. The objective of this paper is to develop a theoretical model that integrates big data into the decision-making process to improve the decision-making process in the organization. The proposed model relies on three basic elements, namely: (1) decision-making process that includes four phases: intelligence, design, choice, and implementation phase;(2) big data analytics to collect, store, manage, and analyze the huge amount of diverse data to extract business value; and (3) decision-modeling using Decision Model and Notation standard to support the communication between the decision makers and Big Data analytical team during the phases of the decision-making process. The interaction between these three elements enhances the decision-making process to make smarter and faster driven-data decisions that make a real impact in the organization.

Organizational Design of Big Data and Analytics Teams

Abstract Although many would argue that the most important factor for the success of a big data project is the process of analyzing the data, it is more important to staff, structure and organize the participants involved to ensure an efficient collaboration within the team and an effective use of the tool sets, the relevant applications and a customized flow of information. A main challenge of big data projects originates from the amount of people involved and that need to collaborate, the need for a higher and specific education, the defined approach to solve the analytical problem that is undefined in many cases, the data-set itself (structured or unstructured) and the required hard- and software (such as analysis-software or self-learning algorithms). Today there is neither an organizational framework nor overarching guidelines for the creation of a high-performance analytics team and its organizational integration available. This paper builds upon (a) the organizational design of a team for a big data project, (b) the relevant roles and competencies (such as programming or communication skills) of the members of the team and (c) the form in which they are connected and managed.

TL;DR: Focus on the Relationships, and Partnerships Will Come

During our tenure on Google's People Analytics team, we have participated in dozens of research partnerships. Some have been wildly successful, resulting in changes to our own organizational practices or policies as well as academic publications. Other partnerships have been less fruitful, largely because both science and organizations are messy and unpredictable. But no matter the outcome, nearly all of our partnerships have come about in the same way—a way overlooked by the authors in their article (Lapierre et al., 2018).

A Semantic Data Model: Meaning Making from Data Structures in the SQL Server

Information systems designs are increasingly concerned with entity relationships and technical programmatic approaches to solutions architecture as opposed to semantic based, business focused information architecture that places business definitions at the centre of the information system design and implementation. The disconnect between information technology and business is perpetuated by an overly prescriptive information technology technical design method that fails to incorporate qualitative and normative aspects of business, where information is structured and delivered according to business. The paper will discuss various decision support and semantic approaches to information design and delivery and argue that the traditional modes of solution delivery do not include meaning making of the data elements which are essential to business information reporting and analytics. The meaning making aspect identified is linked to data dictionary or business data glossary that allows for the discovery of semantic meaning from the SQL Server. Using Christian Fürber’s methodology on semantic programming, the analytics team developed a semantic model that enabled detailed definition of fields and the discovery of information using semantic search functionality embedded in the SQL Server. The project provided semantic data framework that provided business with the capability for semantic reconciliation and data sets that were further integrated with Tableau visualization and SQL auto processes.

Improving quality care through IT/analytics for team-based care interventions.

316 Background: Team based care is essential to achieve coordinated, high-quality care. We recognized a responsibility to identify new treatment patients to provided targeted, team based interventions. Previous attempts at identifying new patients were difficult, manual, and delayed. Our aim was to provide personnel with the appropriate means to identify new patients for services such as patient education, financial navigation, social work assessments and outreach calls for care coordination. As a result of multiple barriers and challenges, only 51% of our new patients were being identified. This project fit well in the larger context of the transformation of health care delivery through our participation in the Oncology Care Model, which places an emphasis on team based care. Methods: We used the Plan-Do-Study-Act continuous improvement model to develop an automated data set to identify new treatment patients across the oncology service line and leveraged our Information Technology (IT) Analytics team. A report was established based utilizing logic to identify the first appointment scheduled within the department and a secondary tab to look back at new patients within a three week time frame. Information is housed within a business intelligence software system that is easy for end users to see and understand. The initial report was validated for accuracy at an identification rate of 93% and the 3 week “look back” report gives the ability to identify the remaining patients. Results: We have seen marked improvements metrics as a result of the utilization of this report. Our Financial Navigators have been able to improve out of pocket estimate calls to 98% in April of 2018. Social Work initial patient assessments have improved by 40%, post chemotherapy outreach calls from RN Care Coordinators have improved 12%, and we have seen a 20% increase in the number of documented oncology patient education encounters. Conclusions: This project demonstrates the value of using IT/Data Analytics for continuous quality improvement projects related to patient centered team based care. The cross collaboration of stakeholders in data analysis should be explored as a mechanism to transform care through key services and interventions.

Standardizing and optimizing institutional-level documentation in the electronic medical record era.

155 Background: Although electronic medical records (EMR) offer numerous potential benefits, there are significant challenges to balance high quality patient care and clinical efficiency with the rising demands of required documentation. Our objective was to propose and implement a strategy to standardize and optimize EMR documentation. Methods: An EMR taskforce was created within our institution’s radiation oncology (RO) department consisting of representatives of the clinical staff and EMR coders. To optimize clinical documentation, disease-site specific consult templates were created using EMR standard data elements and personalized drop down menus (Smartlists). Meetings were held with the billing department for template review and with the analytics team to begin automated data extraction into research databases. Quality metrics such as the average number of clinic visits billed at levels 4 or 5 (B45) and the average number of hours per visit (AHV) were compared in different clinics within a 6-month pre- and post- template implementation time period. Results: Within 3 months of establishing the taskforce, 32 RO disease-specific templates were created and distributed for usage in clinic. Optimized template features include consistent cancer staging, standardized oncologic history entry, usage of macros for patients’ symptoms for multiple provider verification, imbedded disease-specific protocol list to improve patient screening and enrollment, and multiple Smartlists for common treatment recommendations. In the GI Clinic, B45 increased from 49.3 to 61.4 per month and similar trends were seen in the Thoracic Clinic (B45, 51.4 to 85.2 per month). The AHV decreased in both the GI (21 to 10 hours) and Thoracic (22 to 13 hours) clinics, showing a significant added value to use of standard documentation. Conclusions: EMR systems can be challenging to navigate; however, they also present a unique opportunity to optimize clinical and research efficiency if used properly. The implementation of standard documentation in our department led to significant improvement in clinical efficiency. Further efforts are being made to adopt these changes on an institutional-wide level.

Pharmacy contributions to improving care and sustaining change in a comprehensive cancer center.

150 Background: Delivery of care in an outpatient hematology/oncology clinic is complex and requires the coordination of numerous services for timely and efficient treatment of patients. The Norris Cotton Cancer Center in Lebanon, NH, began a comprehensive assessment of clinic and infusion operations in 2012 and developed a system-wide initiative to address inconsistent performance in infusion, clinic and pharmacy services. Methods: Cancer center staff and leadership were trained on lean six-sigma methodology to initiate system-wide improvements through targeted project work. We collaborated with quality improvement experts and our analytics team to obtain and analyze data from our EMR. Then we established process measures for all operational areas and a key performance indicator (KPI) for hem/onc infusion: “on time performance.” The pharmacy contributed to on-time performance by focusing improvement initiatives on turn-around time (TAT) for clinical reviews and order verification, TAT for drug preparation, and initiating a process for pre-making chemotherapy. Results: "On-time performance” was initially defined as the percentage of patients whose first medication began within 45 minutes of their scheduled infusion start time. The baseline of 63% in 2012 was improved to 86% in 2018. Through sustained process improvement work and concerted efforts to further improve the patient experience and utilization of resources, an interdisciplinary team targeted a 15 minute on-time performance measurement; this baseline of 28% was improved to 47% in less than a year. In 2012 the pharmacy set a goal of reviewing and verifying orders within 30 minutes, and completing drug preparation within 45 minutes of order verification. The baseline percentage of orders reviewed within 30 minutes was 47% and this increased to 93% in 2018. Our baseline percentage of doses prepared within 45 minutes was 37% and the current median time for drug preparation across all orders is 45 minutes. The pharmacy currently pre-makes approximately 11% of doses. Conclusions: The cancer center pharmacy increased efficiency and improved on-time performance in delivery of care by using data-driven continuous improvement methods.

The Effect of Incentives in Non-Routine Analytical Team Tasks - Evidence From a Field Experiment

Despite the prevalence of non-routine analytical team tasks in modern economies, little is known about how incentives influence performance in these tasks. In a field experiment with more than 3000 participants, we document a positive effect of bonus incentives on the probability of completion of such a task. Bonus incentives increase performance due to the reward rather than the reference point (performance threshold) they provide. The framing of bonuses (as gains or losses) plays a minor role. Incentives improve performance also in an additional sample of presumably less motivated workers. However, incentives reduce these workers' willingness to explore original solutions.

Facilitating Patient Recruitment Process for Research

IntroductionThe Assessing outcomes of enhanced Chronic disease Care through patient Education and a value-based formulary Study (ACCESS) conducted from the University of Calgary trial is seeking 4700 low-income Albertans over the age of 65 years at high risk for cardiovascular morbidity and mortality. Recruitment efforts using advertising, conventional methods including posters and brochures in pharmacies have been challenging. The use of admail was attempted but fewer than 260 people (out of nearly 122,000 letters mailed) were enrolled.
 Objectives and ApproachThe objective was to determine if linking data collected by Alberta Health Service (AHS) could identify eligible patients and facilitate recruitment for the study.
 We extracted cohorts of data based ICD codes. These patient’s data were linked with Admission, Discharge and Transfer (ADT) and Master Patient Index (MPI) data to pull patient’s names, addresses and postal codes. Deceased and previously contacted patients were eliminated. The final patient name-list from the Analytics team was merged with a notification letter from Research Administration and sent by the data communication team to candidate patients. Interested patients contacted the researchers. Once informed consent was obtained, the data communication team sent the study questionnaire to the patients directly.
 Results30,343 eligible patients were identified in Calgary and 23,305 in Edmonton. Out of 13825 people contacted, 304 people were enrolled into the study – a significantly higher rate than using other mail-based methods.
 Conclusion/ImplicationsBy linking various health administrative data, we assisted researchers to identify potential participants who would otherwise be inaccessible and geographically dispersed across Alberta. This effectively facilitated the recruitment process and enabled patients from across the province to participate with minimal investments.

Interactive Data Visualization of Patient Experience and Inpatient Datasets using Tableau Desktop

IntroductionData visualization is a valuable means for reporting and interpreting large datasets. It allows users to present key messages from the data in a simple way. Although many industries have adopted and embedded data visualization within their analytic teams, healthcare has only recently begun to realize this potential.
 Objectives and ApproachThe objective of this interactive presentation is to introduce Tableau (visualization software) and to provide quick, impactful examples of its use. Visualizations will be created using a free version of Tableau Desktop; available to students and academics. Two blinded datasets encompassing hospital discharges in Alberta from April 2014 to March 2016 will be used. The first dataset will contain approximately 50,000 patient experience surveys, as completed by patients within 6 weeks of their hospital discharge. The second dataset will contain inpatient records from the Discharge Abstract Database (DAD). Visualizations will be created using the individual and combined datasets.
 ResultsFollowing a brief description of each dataset and its respective elements, a variety of interactive visualizations will be created in real-time. From the patient experience dataset, we will be able to quickly determine which hospitals have the highest overall rating from their patients. We will then display the results from all survey questions from a single hospital; allowing for a determination of areas where care is delivered well, and to provide opportunities for improvements. From the DAD, we will highlight hospital length of stay, and its relation with gender, age group, geography, and clinical condition. In the final portion of the presentation, both datasets will be linked to examine the relationships between survey responses, patient demographics, and clinical characteristics.
 Conclusion/ImplicationsData visualization has great potential in healthcare. From this presentation, attendees will receive an introduction to its use using practical, real-world examples. The dynamic visualizations in this presentation will be created in mere minutes; a small fraction of the time typically spent by analysts to create static, paper-based reports.

Type 2 diabetes management among older American Indians: beliefs, attitudes, and practices

ABSTRACT Objective: The purpose of this study was to examine beliefs, attitudes, and practices of older American Indians regarding their type 2 diabetes mellitus (T2DM) management. T2DM is one of the leading causes of morbidity and mortality among American Indians. American Indians are more than twice as likely to have T2DM and have over three times a T2DM mortality rate as Whites. Design: Study participants were older members of a federally recognized tribe who had T2DM. A low-inference qualitative descriptive design was used. Data were collected through semi-structured in-depth qualitative interviews with a mixed inductive, deductive, and reflexive analytic team process. Results: Our study sample included 28 participants with a mean age of 73.0 ± 6.4 years of whom 16 (57%) were women. Participants’ mean self-confidence score of successful T2DM management was 8.0 ± 1.7 on a scale from 1 to 10 with 10 representing the greatest amount of confidence. Participants’ mean HbA1c was 7.3% ± 1.5%. Overall, participants discussed T2DM management within five themes: 1) sociocultural factors, 2) causes and consequences, 3) cognitive and affective assessment, 4) diet and exercise, and 5) medical management. Conclusions: It is important to be aware of the beliefs and attitudes of patients. Lay understandings can help identify factors underlying health and illness behaviors including motivations to maintain healthy behaviors or to change unhealthy behaviors. Such information can be helpful for health educators and health promotion program staff to ensure their efforts are effective and in alignment with patients’ realities.

Quant data science meets dexterous artistry

Data Science is a bridge discipline connecting fundamental science, applied disciplines, and the arts. The demand for novel data science methods is well established. However, there is much less agreement on the core aspects of representation, modeling, and analytics that involve huge and heterogeneous datasets. The scientific community needs to build consensus about data science education and training curricula, including the necessary entry matriculation prerequisites and the expected learning competency outcomes needed to tackle complex big data challenges. To meet the rapidly increasing demand for effective evidence-based practice and data analytic methods, research teams, funding agencies, academic institutions, politicians, and industry leaders should embrace innovation, promote high-risk projects, join forces to expand the technological capacity, and enhance the workforce skills.

Prompting Technology and Persons With Dementia: The Significance of Context and Communication.

Smart home auto-prompting has the potential to increase the functional independence of persons with dementia (PWDs) and decrease caregiver burden as instrumental activities of daily living (IADLs) are completed at home. To improve prompting technologies, we sought to inductively understand how PWDs responded to auto-prompting while performing IADL tasks. Fifteen PWDs completed eight IADLs in a smart home testbed and received a hierarchy of verbal auto-prompts (indirect, direct, multimodal) as needed for task completion. Two researchers viewed archived videos and recorded the observed behaviors of the PWDs and their reflections watching the PWDs. Using qualitative descriptive methods, an interdisciplinary analytic team reviewed transcripts and organized data into themes using content analysis. Context and Communication emerged as the major themes, suggesting that positive user experiences will require auto-prompting systems to account for a multitude of contextual factors (individual and environmental) such as level of cognitive impairment, previous exposure to task, and familiarity of environment. Communicating with another human rather than an automated prompting system may be important if individuals begin to exhibit signs of stress while completing activities. Additional work is needed to create auto-prompting systems that provide specific, personalized, and flexible prompts. Holistic conceptualization of "successful task completion" is needed and a positive end-user experience will be key to utility. Such systems will benefit from including positive reinforcement, training, and exploration of how, and whether, direct human involvement can be minimized during the provision of in-home care.

A generic data analytics system for manufacturing production

The increase in the amount of manufacturing information available means that big data can be collected and, with appropriate deep analysis, could be of great value to manufacturers. However, most small manufacturers cannot afford the overhead of a professional data analytics team. To address this problem, in this paper a generic data analytics system, Generic Manufacturing Data Analytics system (GMDA), is proposed. This system can perform most manufacturing data analytics tasks and users can easily carry out data analysis even if they have no prior knowledge or experience of data analytics. To establish such a system, we designed an abstract language, GMDL, to describe the manufacturing data analytics tasks. Aimed at factory data analytics, several algorithms were selected, tuned, optimized, and finally integrated into the system. Some noteworthy techniques were developed in GMDA such as proper algorithm selection strategy and an optimal parameter determination algorithm. Case studies show the practicability and reliability of the system.

Back-Testing of Expected Shortfall: Main Challenges and Methodologies

In a context of an ever-changing regulatory environment over the last years, Banks have witnessed the draft and publication of several regulatory guidelines and requirements in order to frame and structure their internal Risk Management. Among these guidelines, one has been specifically designed for the risk measurement of market activities. In January 2016, the Basel Committee on Banking Supervision (BCBS) published the Fundamental Review of the Trading Book (FRTB). Amid the multiple evolutions discussed in this paper, the BCBS presents the technical context in which the potential loss estimation has changed from a Value-at-Risk (VaR) computation to an Expected Shortfall (ES) evaluation. The many advantages of an ES measure are not to be demonstrated, however this measure is also known for its major drawback: its difficulty to be back-tested. Therefore, after recalling the context around the VaR and ES models, this white paper will review ES back-testing findings and insights along many methodologies; these have either been drawn from the latest publications or have been developed by the Global Research & Analytics (GRA) team of Chappuis Halder & Co. As a conclusion, it has been observed that the existing methods rely on strong assumptions and that they may lead to inconsistent results. The developed methodologies proposed in this paper also show that even though the ES97.5% metric is close to a VaR99,9% metric, it is not as easily back-tested as a VaR metric; this is mostly due to the non-elicitability of the ES measure.

The Effect of Incentives in Non-Routine Analytical Team Tasks

Despite the prevalence of non-routine analytical team tasks in modern economies, little is known about how incentives influence performance in these tasks. In a field experiment with more than 3000 participants, we document a positive effect of bonus incentives on the probability of completion of such a task. Bonus incentives increase performance due to the reward rather than the reference point (performance threshold) they provide. The framing of bonuses (as gains or losses) plays a minor role. Incentives improve performance also in an additional sample of presumably less motivated workers. However, incentives reduce these workers’ willingness to “explore” original solutions.

A Visual Analytics Platform and Advanced Visualization Tools for Interpreting and Analyzing Wind Energy Time-Series Data

Abstract The Bonneville Power Administration (BPA) connects nearly 5,100 MW of wind energy to its transmission system through dozens of wind energy projects located in the states of Oregon and Washington. These wind energy projects are producing large amounts of real-time data that need to be efficiently monitored and analyzed by BPA operators in order to combine and balance this variable source of energy with other energy resources to meet energy demands. To support wind energy monitoring, operations, and integration, visual analytics researchers at the Pacific Northwest National Laboratory have developed a visual analytics platform that is able to connect to existing wind operations data sources, fuse and integrate wind energy data, and efficiently serve integrated wind energy data to advanced web-based visualization tools. In addition, the visual analytics team has also developed a set of advanced visualization tools that are available through the visual analytics platform to directly support wind energy operations and analysis needs.