Virginia Mason Franciscan Health-1

Healthcare and Pharma analytics

Healthcare

The goal of this project is to develop a data-driven decision support system to reduce operating costs for the accountable care organization within Virginia Mason Franciscan Health while simultaneously enhancing the quality of patient care through the implementation of advanced business analytics. The developed system will leverage a predictive approach and comprehensive data from various sources, including patient records, financial reports, and operational metrics, to provide insights and actionable recommendations for informed decision-making, leading to cost reduction and quality improvement in patient care. Ultimately, our project deliverable achieves sustainable cost savings and promotes efficient resource management, leading to improved patient outcomes and organizational success.

#Healthcare and Pharma analytics,Non-Profit

Virginia Mason Franciscan Health-2

    Healthcare and Financial Analytics

Healthcare

This project models risk scores for patients who are enrolled in Rainier Health Network ACO, which is comprised of Medicare beneficiaries. Medicare is the federal health insurance program for people who are 65 or older, under 65 who have disabilities or end-stage renal disease. Our team built a risk stratification model using the current and historical population data to predict the risk score of current beneficiaries for the next benefit year. The model will assess the health indicators like utilization, demographics, and medical history of the patients to predict their future risk score. The purpose is to enable provider groups and care managers to offer customized healthcare services. To benefit patients and healthcare providers by improving services and resource allocation.

#Healthcare,#Financial Analytics

HeyKiddo

Healthcare

HeyKiddo is very new in the Market and the goal is to help children with mental issues by providing Heykiddo apps and resources. Our team is building and developing the Heykiddo app which is in beta testing and developing its feedback mechanism to send customers relevant mental health resources on a daily basis. The app has a subscription model and our team continuously developing the application and its features. Our team also created a collaborative filtering machine learning model to learn the past activities of the user and send the relevant materials based on their past behaviors. Our team delivered 2 main solutions which include app development features in beta testing and a collaborative filtering ML model to launch their products in the future.

#Healthcare Analytics, #Machine learning Model development

MultiCare Health System

Healthcare

Multicare OCEDs are 10-bed emergency rooms staffed with board-certified emergency physicians providing emergency medical care. The key business problem is that there are no well-defined metrics to measure nurse performance and it is difficult to keep track of relevant data. Our aim is to identify solutions for measuring nurse performance through data analysis and implement an ongoing dashboard to track and monitor nursing productivity. The MSBA Team 4 has developed various key performance indicators by analyzing trends, scores, and seasonality to deliver an interactive dashboard and mobile application. The team built a ML model through Azure to identify significant data pills relationships including nursing shifts, facility, and severity level as well as a Power App for cognitive analysis.

#Healthcare Analytics, #NursePerformanceAnalysis

PropertyScout.io

Real Estate

PropertyScout.io is a startup in the rapidly growing real estate data domain. Since its establishment in 2019, the company has been operating under a B2B model, catering to diverse industries by providing access to property documents and details. However, the firm faces challenges related to a high churn rate. The project was divided across four stages, starting with business discovery, involving industry and client analysis, to gain understanding of the task. Next, data understanding was pursued by gaining access to the customer database, enabling initial data analysis and visualization. Data modeling followed, wherein Python and PowerBI were utilized to create a preliminary solution. Finally, the solution deployment phase ensured the project deliverables were finalized.

#Descriptive Analytics

Schnitzer Steel

Steel Manufacturing and Scrap Metal Recycling Company

Recognizing the need to reduce mill operational cost to be more profitable, Schnitzer Steel wants to minimize unscheduled machinery downtimes occur in their mill. Our project’s goal is to help Schnizter understand unscheduled downtime better through identifying the environments in which unscheduled downtime is long or frequent. To accomplish this, the team has developed a prototype decision tree model that examines performance metrics and predicts the length of unscheduled delays. An output method has been established to enable operations managers to closely monitor these predictions and minimize downtime caused by component failures. Thorough documentation has also been provided to support data governance practices at Schnitzer. Overall, the solution of this project will help Schnitzer understand unscheduled delays better, plan efficient actions, and become more data-oriented.

#Manufacturing and Operations/Optimization

Sound Credit Union

Banking and Fintech

 Sound Credit Union have identified that while they’re gaining a certain amount of members each year there is also a very high rate of closed accounts that they would like to understand and reduce. Through our analysis and investigation of available data, our team has the goal of gaining a better understanding of the variables that affect member turnover at Sound as well as suggesting a retention strategy for proposing best offers. We provided a working predictive model to Sound for ongoing marketing functions and retention strategies along with a web app and inbuilt bot for deployment on member facing platforms. We discovered ways of differentiating Sound from competitors in a crowded financial marketplace by providing the right financial products at the right time to the right individuals.

#Marketing, #Strategy

South Sound Military Communities Partnerships (SSMCP)

Government

Our team was tasked with updating the 2020 Regional Economic Impact Analysis (REIA) of Joint Base Lewis-McChord (JBLM) for the South Sound Military Community Partnership (SSMCP) with the IMPLAN Analysis provided by the Thurston Economic Development Council (TEDC). The 2022 REIA analyzes the direct, indirect, and induced impacts that JBLM has on its local communities of Pierce and Thurston counties. As JBLM is the second largest employer in Washington State, and first in the South Sound, with over 54,000 military personnel and contractors, this report will have a significant impact in informing leaders in different government bodies to help them make decisions regarding the military base.

#Economic Analytics

Tacoma Power

Utilities Sector

Team 9 collaborated with Tacoma Power, a citizen-owned electric utility company that generates, transmits, and distributes electricity in a competitive marketplace. Our team was tasked with developing Tableau dashboards that provide valuable insights into future trends, spending, and resource prioritization for asset maintenance. Leveraging skills acquired from the MSBA program, including but not limited to machine learning, statistical regression and data mining, the team analyzed multiple data sets provided by Tacoma Power. The project resulted in the creation of a proxy dashboard and a current dashboard, which offer comprehensive details on assets and work order data. These powerful dashboards enable Tacoma Power to make well-informed decisions regarding cost, optimize resource allocation, and address data discrepancies effectively.

#Asset Management Analytics, #Financial Analytics,#Resource Allocation Analytics

UW Tacoma

Higher Education

University of Tacoma Institutional Research is partnering with our team of Master of Science in Business Analytics student consultants to develop a suite of data products to empower decision makers to utilize data to increase the efficiency and transparency of the decision-making process. Our team has developed a suite of dashboards which aim to illustrate a snapshot of institutional effectiveness before, leading up to, and after the COVID-19 pandemic as well as predictive model to better count students in resource allocations, thereby strengthening the bargaining power of department executives as well as increasing institutional knowledge about student behaviors. Contextualizing the data by including the COVID-19 impacts is critical, given that the Higher Education industry has been vastly impacted and now finds itself in the beginning of the recovery period.

#Operations, #Scheduling, #Measures of Institutional Effectiveness

Westwood Shipping Lines 

Shipping

Westwood Shipping Lines is an independent vessel operator specializing in trade between the Pacific Northwest and North Asia. Due to the dynamic nature of the shipping industry, maintaining schedule integrity can be challenging and time-consuming, so Westwood has partnered with the UWT MSBA student consultants to create an optimization tool to determine which routes will yield the highest profits based on cargo demands and reduce the current number of personnel hours spent on scheduling. Our team has created an optimization model that considers costs, revenues, cargo demands, and other parameters, then iteratively cycles through routing options and calculates which ports should be called and which should be bypassed to achieve maximum profit based on user inputs.

#Operations #Optimization, #Scheduling

Analytics Innovation Club 

Artificial Intelligence

Artificial Intelligence and Machine learning systems and algorithms are as perfect as the data they are built upon. Unfortunately, Data is often imperfect in the sense they incorporate the prejudices and bias of the decision maker or the system through how it’s collected or the prevalent traits in a society/business. When the data is biased the AI/ML algorithms become objectionable by unconsciously discriminating and putting certain privileged groups at advantage and certain unprivileged groups at disadvantage.

The project deals with testing and creating a use case presentation for the AI fairness 360 python package. The package consists of a comprehensive set of metrics and models to test for bias and a set of algorithms to mitigate the bias in the dataset and thus making the decision-making system fair.

#Bias Mitigation Algorithms, #AI Fairness