Menu
Deepti Bahel
Deepti Bahel

Deepti Bahel

Senior Data Engineer and AI analytics practitioner
VigourSoft Global Solutions
Bio

Deepti Bahel is a Senior Data Engineer and AI analytics practitioner with over 12 years of experience building data platforms, decision intelligence systems, and AI-driven analytics solutions across healthcare, finance, retail, and technology sectors. She has worked with organizations including Google, Intuit, Accenture, and Wayfair, specializing in data engineering, business intelligence, advanced analytics, and AI-enabled decision support.

Her recent work focuses on healthcare AI innovation, including clinical analytics platforms, predictive modeling for patient outcomes, and decision support systems such as her Blood Donation Forecasting DSS and hospital analytics applications. Deepti is also active in the data and AI community as a conference speaker, mentor, and advocate for data-for-good initiatives. She combines strong technical expertise in data engineering, AI, and analytics strategy with a passion for building human-centered solutions that translate data into meaningful real-world impact.


Saving Lives with Forecasts: A Blood Donation Decision Support System (DSS) for Data-for-Good

Blood shortages are not simply a logistics challenge—they are a decision-making problem under uncertainty. This session presents a practical, end-to-end Blood Donation Decision Support System (DSS) built to help blood centers and public health teams improve donor retention, forecast donation volume, and allocate outreach resources more effectively.

Using a working analytics application as a case study, the talk walks through a real-world workflow that moves from mission to action. Attendees will see how donation history and behavioral signals are transformed into reliable datasets; how exploratory analysis surfaces key patterns in donor recency, frequency, and campaign response; and how predictive models estimate repeat-donation likelihood using both interpretable baselines and ensemble methods. Time-series forecasting is used to anticipate donation trends and seasonality, while “what-if” scenarios allow teams to evaluate outreach strategies before deployment.

Rather than focusing on algorithms in isolation, the session emphasizes pragmatic design choices that matter in high-stakes, social-impact settings: thoughtful feature engineering from behavioral data, evaluation beyond raw accuracy (including calibration, operational thresholds, and cost-awareness), and optimization techniques that balance expected impact against real budget constraints. The system also demonstrates human-centered deployment patterns, combining transparent visualizations and AI-generated summaries to support—not replace—decision-makers.

Participants will leave with a reusable blueprint for building decision-ready analytics products for Data-for-Good, including common pitfalls to avoid and a clear framework for moving from dashboards to responsible, real-world action

Professional
Back To Top

The Institute for Operations Research and the Management Sciences

5521 Research Park Drive, Suite 200
Catonsville, MD 21228 USA

phone 1 443-757-3500

phone 2 800-4INFORMS (800-446-3676)

fax 443-757-3515

email

Get the Latest Updates

Copyright INFORMS. All Rights Reserved