An Interactive Decision Support Dashboard for Monitoring and Forecasting UAC Care System Operations Using Streamlit and Power BI
This paper presents an interactive decision support dashboard for the UAC Care System, developed using Python, Streamlit, SQLite, and Power BI, which integrates real-time monitoring, predictive analytics, and scenario planning to enhance operational visibility and evidence-based decision-making for public-sector care managers.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine the UAC Care System as a massive, bustling train station where thousands of young travelers arrive every day. For a long time, the station managers were trying to run this complex operation using nothing but crumpled paper maps and handwritten logs. They were like conductors guessing when the next train would arrive, often missing the fact that the platform was already overflowing until it was too late.
This paper introduces a brand new "Mission Control" dashboard—a digital command center built to replace those messy notebooks. Think of it as a super-smart, interactive video game interface for real-life operations. Instead of guessing, managers can now see exactly how many people are on the platform, how fast they are moving through the station, and where the traffic jams are forming, all in real-time.
The Big Idea: A Unified Command Center
The authors built this system by mixing four powerful tools: SQLite (a tiny, super-fast digital filing cabinet), Python (the brain that does the math), Streamlit (the colorful, clickable screen you see), and Power BI (a high-tech telescope for deep dives).
They didn't just dump raw data onto a screen. Instead, they organized the data into neat "views"—like pre-sorted folders in a filing cabinet. This means the dashboard doesn't have to do the heavy lifting every time you click a button; it just reads the pre-sorted folders. This makes the system light, fast, and ready to answer questions instantly.
What the Dashboard Actually Does
The dashboard is split into different "tabs," each acting like a different tool in a mechanic's toolbox:
- The Executive Summary: This is the "at-a-glance" view. It shows big, bold numbers (KPIs) like the total number of kids in the system and a "risk score." It's like the speedometer and fuel gauge on a car dashboard.
- Operations Monitor: Here, you can watch the flow of people. It tracks how many kids arrive (intake), how many leave (discharges), and how many are stuck in the middle (backlog).
- Forecast & Anomalies: This is the crystal ball. Using past patterns, the system suggests that under current conditions, the total number of kids in the system could rise by about 30% over the next 30 days. It also acts like a smoke detector, flagging weird spikes or drops in the data that don't make sense.
- Scenario Lab: This is the "what-if" playground. Managers can slide a virtual slider to say, "What if we send 100 more kids out today?" or "What if fewer arrive?" The system instantly recalculates the numbers, showing how those changes would affect the backlog. It's like a flight simulator for decision-making.
- Data Quality: Before showing any numbers, the system checks its own homework. It counts the rows and checks the dates to make sure the data isn't broken or missing. If the data is shaky, it warns the user, because you can't trust a map if the roads are missing.
What the Numbers Tell Us
When the team tested this system on real data, they found some interesting patterns. In early 2023, the average number of kids in the system was about 7,414. During that time, the "net intake pressure" was negative, which is a fancy way of saying more kids were leaving the system than arriving. It was a moment of relief for the station.
However, the story changed in 2024. The data suggests that this year saw the highest buildup of a 14-day backlog. Even though many kids were being discharged, the number of new arrivals was growing even faster, creating a traffic jam. The system identified that the main "risk drivers" were the total number of kids in the system and how well the discharge process was working. When the system gets crowded and kids aren't leaving fast enough, the risk goes up.
What This System Is NOT
It's important to know what this dashboard doesn't do. The authors explicitly state that this is not a magic wand that solves the problem on its own. It doesn't replace the need for human managers or policy changes. Also, while it uses advanced math, the paper rules out the idea that this is a massive, expensive enterprise system requiring a supercomputer. Instead, it proves you can build a powerful tool using lightweight, open-source tech that fits on a regular computer.
How Sure Are We?
The results presented here are based on simulations and simulated scenarios run on historical data. The authors demonstrate that the dashboard works and that the tools can spot trends and risks. They suggest that this approach could be copied by other public organizations to save money and make better decisions. However, they don't claim to have solved the entire UAC care crisis; they've just built a much better pair of glasses to see the problem clearly.
The Bottom Line
This paper shows that by swapping out old spreadsheets for a modern, interactive dashboard, managers can stop reacting to emergencies and start planning for them. It turns a chaotic pile of data into a clear, actionable story, helping leaders see the traffic jams before they happen and steer the ship in the right direction. It's a lightweight, scalable, and cost-effective way to bring the future of data-driven decision-making into the public sector today.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.