Home pageCapabilitiesData Science & Advanced Analytics

Data Science & Advanced Analytics

The Data Science & Advanced Analytics practice provides important insight into clients operations, helps improves safety and productivity while reducing fraud, waste, and abuse.

TBD

Data pipeline creation via data cleansing and standardization
Data-driven business decisions and future planning
Business Intelligence
Statistical Analysis
Predictive Analytics via Artificial Intelligence (AI) and Machine Learning (ML)
Dashboards for performance monitoring and trend analysis
Custom coding and analytical models

Data Engineering

Data engineering requires creating a data pipeline by collecting, translating, and validating data for analysis.
It can be achieved by using tools in Microsoft Azure or Amazon Web Services or by building custom tools within the Python environment.
The ASC team has extensive capability in building data pipeline for real-world problems in transportation, health care, finance, and other domains.

Core System Transformation

ASC can assist with core system transformation, including:

Creating a data engineering platform and migration to cloud
Data standardization
Performing predictive analytics for budget forecast based on priorities
Creating dynamic dashboards to monitor daily operations

Microsoft Azure

Can create schema and tables in MS SQL database and access the data using a Python framework.
Can use Azure Databricks, which is an easy, fast, and collaborative Apache spark-based analytics platform. It accelerates innovation by bringing data science, data engineering, and business together, making the process of data analytics more productive, more secure, more scalable, and optimized for Azure.
Can perform data analytics; Azure SQL can be integrated with Python, and quick data visualization can be performed in Python, R, or other programming environment.

Azure SQL Python Integration

Azure SQL can be integrated with Python, and quick data visualization can be performed.

Amazon Web Services

Just like Azure, AWS provides easy to use tools for data management, governance, and cloud migration.

Dashboards for Tracking Real-Time Traffic Congestion and Hotspots using Traffic Sensor Data

Recurring Traffic Congestion is a common problem in urban areas. Using Maryland Department of Transportation live sensor data, we created dashboards to track hotspots along the Maryland portion of the Capital Beltway and I-95 between Washington, D.C., and Baltimore Beltway. The dashboard captures vehicles travelling at speeds less than 40 mph at a given time. The map on the right in the screenshot above shows the locations at which vehicle speeds are less than 40 mph on an average weekday afternoon peak hour. The plot on the top left shows the speed distribution histogram, and the plot with red bars shows the location of vehicles at or below 40 mph. Such a dashboard is very useful for displaying instantaneous vehicle positions and tracking hotspots.

Time-Series Analysis to Track Stock Market Performance of Financial Companies

We performed a time-series analysis to track stock market performance of some financial companies. The screenshots above show a five-year performance comparison of five financial companies. Such an analysis helps track spikes and dips over time to understand the market performance.

Analyzing Drunk Driving Related Fatalities

Using data from the National Highway Traffic Safety Administration (NHTSA) and US Census, we analyzed 2019 traffic fatalities and fraction of such fatalities due to drunk driving. We then analyzed the distribution of such fatalities over population density of various U.S. states. The result can be seen in the screenshot above. This analysis is useful to allocate resources appropriately to curb traffic fatalities due to drunk driving.