Research on supply chain coordination, machine learning for forecasting and bandit pricing, and process improvement.
Supply Chain Coordination
I study how pricing and operational policies can coordinates supply chains and achieves optimal profits, with particular interests in business volume discounts, lot-sizing decisions, and perishable-goods supply chains.
Supply Chain Coordination Business Volume Discounts Perishable Supply ChainsMachine Learning for Forecasting & Bandit Pricing
I apply machine learning to operational decision-making, with current work spanning supervised learning for forecasting demand and capacity and online bandit learning for dynamic pricing and revenue management.
Machine Learning Forecasting Bandit PricingProcess Improvement Projects
I study process improvement projects with a focus on Six Sigma projects, and the organizational and project-level factors that affect outcomes in manufacturing and service settings.
Process Improvement Six Sigma Project SuccessThis study synthesizes participatory action research across 13 Six Sigma projects in manufacturing and service firms to identify factors associated with project success. The analysis highlights timeliness and prioritization as recurring themes and examines how project conditions shape their effects.
This study compares machine-learning and traditional time-series approaches for forecasting U.S. manufacturing capacity utilization. It evaluates the relative predictive value of internal operational variables and external macroeconomic risk indicators, with implications for capacity and inventory planning.
This study develops a probabilistic machine-learning approach for forecasting intermittent demand by combining Natural Gradient Boosting with a negative binomial distribution. The model improves demand forecasts with respect to Poisson GLM and Croston's benchmarks, explicitly accounts for uncertainty, and supports inventory decisions such as safety stock and reorder-point planning.
This study develops an analytically grounded pricing approach for coordinating a multi-product supply chain under business volume discounts. The proposed policy combines a uniform markup with a calibrated discount breakpoint to improve system profitability under linear price-sensitive, isoelastic, and exponential demand scenarios.
This study empirically tests existing Six Sigma project-success frameworks using survey data from U.S. manufacturers. It examines how project-management practices, timeliness, prioritization, project complexity, and uncertainty shape Six Sigma project success.
This study develops a bid-price-consistent primal-dual online learning algorithm for hotel pricing under unknown demand. The algorithm learns nightly prices from booking responses while accounting for room scarcity and operational pricing constraints, with the goal of maximizing revenue over time.
This project examines how time, temperature, product movement, and temperature-abuse scenarios affect the remaining shelf life and value of cherries across the cold supply chain. The goal is to quantify operational losses and support more profitable distribution decisions.
* Presentation by a co-author.