Suleyman
Yildirim, Ph.D.
|
Transforming complex data into strategic intelligence — at the intersection of Machine Learning, Mathematical Optimization, and Decision Science.
SY
Bridging AI Research
with Enterprise Impact
Dr. Suleyman Yildirim is a hands-on Data Science leader with a Ph.D. in Industrial Engineering & Operations Research and a Computer Science minor — a rare combination that enables him to operate fluidly across advanced mathematics, machine learning engineering, and C-suite strategic communication.
With over 10 years of combined academic and industry experience, Dr. Yildirim currently leads high-impact AI and optimization initiatives at Blend360, where he manages a team of 4+ data scientists and delivers production-grade AI systems for Fortune 500 enterprise clients across hospitality, healthcare, and retail sectors.
Beyond the technical, Dr. Yildirim operates as a strategic thought partner to clients — translating complex model outputs into business decisions, leading cross-functional workshops, mentoring junior data scientists, and shaping the data science roadmap at the engagement level.
His academic roots remain active: his research combines machine learning with discrete event simulation, and bridges predictive modeling with prescriptive optimization — building systems that not only forecast outcomes but drive optimal decisions. This integration of ML and mathematical optimization is the thread that connects his academic work and industry practice.
Six Pillars of Applied AI Leadership
A unique blend of research-grade modeling depth and enterprise-scale deployment experience.
Applied Machine Learning
Production-grade ML from research to deployment. Expertise in gradient boosting, ensemble methods, deep learning, time-series forecasting, and uplift modeling for high-stakes business decisions.
Optimization & Decision Science
Deep fluency in mathematical programming — formulating and solving MIP, LP, and multi-objective optimization problems using Gurobi, PULP, and custom solvers for real-world resource allocation and decision-making.
AI Strategy & Consulting
Translating advanced AI capabilities into business strategy. Working directly with executives and client stakeholders to identify high-value AI opportunities, design solution architectures, and drive enterprise-wide adoption.
Price & Promotion Optimization + Uplift Modeling
Modeling price elasticity and designing promotion strategies using statistical regression and mixed-integer optimization. Combines causal inference frameworks, A/B experimentation pipelines, and T-Learner/S-Learner uplift models to measure true incremental impact — enabling data-driven pricing and offer allocation decisions at enterprise scale.
ML & Discrete Event Simulation
Combining machine learning with discrete event simulation to model complex clinical and operational systems. Research includes ML-enhanced simulation calibration, sepsis prediction, and early-stage clinical decision support using real-world EHR datasets.
Enterprise AI Systems
End-to-end ML pipeline design and deployment at scale — from feature engineering and model training to production inference, monitoring, and MLOps on cloud-native architectures across AWS and Azure.
Tools & Technologies
Programming
ML & Modeling
Optimization
Cloud & MLOps
Simulation
Professional Experience
A decade of progressively impactful roles at the frontier of applied AI and enterprise data science.
Manager, Data Science
Leading a team of 4+ data scientists in delivering end-to-end AI and optimization solutions for Fortune 500 hospitality and healthcare clients. Accountable for project strategy, model architecture, team performance, and C-suite-level stakeholder communication.
- Built an ML uplift-driven optimization engine (T-Learner + MIP in PySpark/SageMaker) to reallocate loyalty program offers across 2M+ members — maximizing incremental revenue while holding campaign cost flat.
- Deployed a real-time loyalty member recognition model integrating ML prediction with mathematical optimization, lifting elite-tier upsell conversion and raising NPS scores measurably.
- Formulated a Gurobi MILP for DC-to-DC transfer optimization across 20 healthcare distribution centers, cutting transfer miles by nearly 50% and significantly elevating service levels.
- Led the end-to-end refresh of a Marketing Mix Model (MMM), delivering channel-level ROI insights and a roadmap toward future-ready budget optimization capabilities.
Lead Data Scientist
Led technical delivery for Fortune 10 retail and healthcare clients, focusing on predictive analytics, demand forecasting, and inventory optimization that directly influenced supply chain strategy.
- Developed XGBoost classification model to predict inventory write-offs — boosting recall from 40% to 68% and enabling proactive inventory intervention.
- Implemented Prophet + LightGBM ensemble for daily demand forecasting, achieving 6% MAPE and enabling significant reduction in safety-stock investment.
- Developed LSTM and XGBoost models to forecast SKU-level Balance-on-Hand, improving prediction accuracy by 25% and delivering significant working capital savings.
Senior Data Scientist
- Modeled category-level price elasticity using linear regression and integrated outputs into a mixed-integer optimization model for pricing decision support.
- Led Provider Search Driver Analysis project to identify internal and external factors impacting patient satisfaction for a Fortune 10 healthcare client.
- Data Scientist on OmniRx — creating user-friendly order, tracking, and reordering capabilities across fulfillment, PBM, and specialty pharmacy business units.
- Finalist for Client Excellence Award and Team Excellence Award at Blend360 Annual Awards 2023.
Postdoctoral Researcher
- Led data science efforts on a U.S. Army-funded combat vehicle resilience project using Set-Based Design methodology, applying ML to defense systems engineering.
Ph.D. Research Assistant
- Developed deep learning-enhanced simulation frameworks and predictive clinical models using Python, MATLAB, and SimEvents.
- Built ML models using the MIMIC-3 clinical dataset for early-stage decision-making and patient outcome prediction.
- Taught graduate and undergraduate courses in optimization, simulation, decision analysis, and supply chain management.
- Awarded Thomas C. Rumble Ph.D. Fellowship, 1st Place Excellence in Ph.D. Research (2020), and Best Technical Paper at MSUG (2019).
Key AI & Optimization Projects
Enterprise-scale systems where research-grade modeling meets measurable business outcomes.
Uplift-Driven Loyalty Optimization Engine
Enterprise-scale ML + optimization system for a Fortune 100 hospitality company. Deployed a T-Learner uplift model in PySpark to estimate incremental response probability per loyalty member, then fed results into a Mixed Integer Programming (MIP) solver to reallocate offers across 2M+ members — maximizing incremental revenue while holding campaign cost flat.
Real-Time Member Recognition System
Integrated ML prediction model with a mathematical optimization framework to power real-time loyalty member recognition at hotel check-in. The system lifts elite-tier upsell conversion rates and measurably improves NPS by surfacing personalized, model-driven offers at the moment of interaction.
Retail Healthcare DC Transfer Optimization
Formulated a Gurobi MILP model to optimize distribution center-to-distribution center transfer decisions across 20 healthcare DCs. Achieved a near-50% reduction in transfer miles while significantly elevating service levels for a Fortune 10 retail distributor.
SKU-Level Inventory Forecasting Suite
Three-model ensemble (LSTM + XGBoost + Prophet) for hierarchical inventory forecasting across multiple SKU and store dimensions. Achieved 25% accuracy improvement in Balance-on-Hand forecasting and 6% MAPE on daily demand — driving significant reductions in working capital and safety-stock investment.
Inventory Write-Off Prediction Model
XGBoost classification model to identify at-risk inventory items before write-off events occur. Improved recall from 40% to 68%, enabling proactive supply chain intervention and preventing significant avoidable inventory write-offs annually for a Fortune 10 retailer.
Marketing Mix Model (MMM) Evolution
Led the end-to-end refresh and modernization of a Marketing Mix Model for a hospitality enterprise client — delivering channel-level ROI insights, data quality validation frameworks, and a strategic roadmap toward a future-ready MMM architecture for budget optimization across marketing channels.
Building AI That
Actually Works in the World
The highest-value AI is not the most mathematically elegant — it's the one that gets deployed, adopted, and trusted by the humans who use it. This belief shapes how Dr. Yildirim approaches every engagement: with equal weight on technical rigor and organizational change.
He leads by bridging two worlds: the precision of mathematical modeling and the pragmatism of enterprise execution. His team-building philosophy centers on developing data scientists who can think critically, communicate clearly, and own their impact end-to-end.
Technical Leadership
Managing and mentoring 4+ data scientists — from code reviews to career growth, performance evaluations, and promotion input.
Strategic Partnership
Translating model outputs into executive narratives. Operating comfortably at both the algorithm level and the boardroom level.
Responsible AI
Advocating for fairness, explainability, and human-in-the-loop design in production AI systems — especially in healthcare and personalization.
Research & Innovation
Interdisciplinary research at the frontier of AI, public health, and intelligent systems — bridging academic rigor with real-world application.
Research Focus Areas
AI for Healthcare & Health Equity
Applying machine learning and simulation modeling to healthcare systems, with a focus on equitable outcomes, clinical decision support, and population health analytics.
Mathematical Optimization & Network Design
Formulating and solving complex optimization problems — from network flow and transportation to large-scale MILP models — applied to supply chain networks, healthcare distribution, resource allocation, and operational decision-making using Gurobi, PULP, and GAMS.
ML-Enhanced Simulation & Change Point Detection
Integrating process knowledge and deep learning into discrete event simulation model calibration via automated change point detection — with applications in complex manufacturing and healthcare systems.
Stochastic Modeling & Distributed Simulation
Developing novel stochastic sampling-based frameworks for distributed and decomposed simulation of large-scale systems, enabling scalable uncertainty quantification.
Publications & Presentations
Process Knowledge Driven Change Point Detection for Automated Calibration of Discrete Event Simulation Models Using Machine Learning
IEEE Transactions on Systems, Man, and Cybernetics · In Review
Stochastic Sampling Based Distributed and Decomposed Simulation
Journal of Simulation · Under Preparation
Integrating Process Knowledge into Change Point Detection in Healthcare Systems
INFORMS Annual Meeting · 2019
Automated Discovery of Process Knowledge Using Deep Learning
WSU Graduate Research Symposium · 2019
CPD for Parameter Estimation in Simulation Models
Michigan Simulation Users Group Conference · 2019 · Best Technical Paper
Let's Build Something
Remarkable Together
Open to conversations about AI leadership roles, enterprise data science consulting, research collaborations, and speaking opportunities.