黑料福利网

earned her PhD from Eindhoven University of Technology on June 3, 2026.
In her research within the Operations Planning, Accounting & Control group (OPAC) of the department of Industrial Engineering and Innovation Sciences, she explores how dynamic selling strategies affect profit, waste and customer satisfaction in the context of perishable products.

Waste

Every day, businesses throw away enormous amounts of unsold products. From airline seats and hotel rooms to fresh groceries and fashion items, companies constantly face a difficult balancing act: stock too little, and customers leave disappointed; stock too much, and valuable products go to waste. This problem arises because businesses can never predict demand perfectly. Even with historical data and forecasting tools, they do not know exactly how many customers will want to buy a product on a given day. To avoid running out of stock, companies often keep extra inventory as a safety buffer, but if demand turns out to be lower than expected, those leftover products may become worthless and end up being discarded. This challenge is especially severe for perishable products, which lose all value once their selling period ends.

Customer choices

Sprenkels proposes a new way to model customer choice among comparable and substitutable products. The model shows that offering a larger assortment generally increases both expected revenue and the prices sellers can charge. However, these benefits eventually level off as assortments become very large. The research also shows that when customers are less willing to switch between products, revenues decline because sellers lose flexibility in steering demand toward available inventory.

Strategy matters

The thesis compares four dynamic selling strategies aimed at improving performance under uncertain demand: dynamic pricing, opaque selling, lottery selling, and a newly proposed strategy called haggling. Dynamic pricing adjusts prices over time based on inventory and demand, opaque selling conceals some product details until after purchase, lottery selling offers customers a random product from a set of alternatives, and haggling allows buyers and sellers to negotiate prices.

The results show that dynamic pricing delivers the highest profits in most scenarios. On average, it also improves resource efficiency and product availability, helping businesses reduce waste while continuing to meet customer demand. However, other strategies can outperform dynamic pricing in specific situations, which means companies must adapt their approach to their context that depend on customer behaviour or the objective of the company.

Faster decisions

Another key contribution is a method that simplifies the pricing decisions. Because determining the optimal selling prices becomes computationally extremely demanding as the number of products within the assortment grows, the thesis also introduces a new approximation method that dramatically reduces computation time. Instead of becoming exponentially more difficult with larger assortments, the proposed method scales efficiently while still producing highly competitive results. In many practical settings, it outperforms existing approximation techniques. This allows businesses to the dynamic pricing strategy without excessive computational effort.

Learning

Sprenkels also takes a deeper look at how these strategies can be introduced in situations where businesses do not yet understand how customers respond to this strategy. In these cases, sellers must learn from data over time how selling prices affect demand while still trying to maximize revenue. The research develops learning policies that adapt dynamically and shows how fast businesses can approach optimal performance under this absence of prior knowledge. This insight supports organizations that experiment with new sales formats in uncertain markets.

Laura Sprenkels defended her thesis on June 3, 2026.
Title of the thesis:
Supervisors: Ivo Adan, Z眉mb眉l Atan