Menu
David Bayba
David Bayba

David Bayba

Principal Engineer
Intel Corporation
Bio

David (Dave) Bayba is a Principal Engineer at Intel Corporation, where he leads advanced analytics and artificial intelligence initiatives across Intel’s Global Supply Chain Operations. With three decades of experience in industrial engineering, operations research, and data science, Dave specializes in developing advanced modeling solutions for materials planning optimization, quality monitoring, and predictive analytics.

Dave is an active researcher, educator, and mentor in supply chain analytics. He has presented at prominent professional conferences and has held leadership roles in numerous Intel internal conferences. Dave also co-chairs Intel’s Manufacturing Strategic Research Segment which mentors Intel’s manufacturing, supply chain, and metrology research portfolio.  Dave holds two patents in automated monitoring systems.

Dave holds a Master of Engineering in Modeling and Simulation from Arizona State University and a Bachelor of Science in Industrial Engineering from the University of Arizona.


An Explainable ML Decision Support System for Spare Parts Inventory Management

This presentation describes Intel’s innovative machine learning framework for spare parts inventory optimization across their global manufacturing network. The solution addresses the classic challenge of balancing customer service with working capital by managing inventory for over 100,000 parts across dozens of locations. Three key innovations distinguish this approach: First, quantile regression forecasting directly estimates demand at target service levels, eliminating traditional safety stock calculations and normal distribution assumptions. Second, SHAP explainable AI provides transparency by highlighting influential factors like seasonality and maintenance schedules, building crucial user trust. Third, interactive simulation capabilities enable real-time what-if analysis for proactive planning. Built on a five-layer architecture integrating forecasting, optimization, and analytics, the system achieved impressive results: 31% reduction in forecast error, 52% runtime improvement, and over 80% user adoption rates, transforming inventory management from reactive to data-driven decision-making.

Essential / 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