liveBook · Manning

Perishability

As explained in Applied Reinforcement Learning

Perishability is the property of losing product value when an item remains unsold beyond its usable or marketable period.

Perishability means an item loses value if it remains unsold beyond its usable or marketable period. Fresh bread may go stale by day’s end; flight tickets, hotel rooms, and seasonal fashion lose value after their selling window closes. section 3.3

With limited inventory and unpredictable, price-sensitive demand, setting prices involves a trade-off: high prices may leave stock unsold, while low prices may sell it too quickly and forgo revenue. Dynamic pricing adjusts prices over the selling window to pursue higher total revenue. section 3.3

Table 3.2 Elements of Markov decision process for perishable dynamic pricing
Markov decision process elements
Definitions
State
How many sales periods are left.
How many items are left to sell.
Action
The agent chooses a price from a predefined list, e.g., [8, 10, 12, 15, 18].
Reward
Calculated as the revenue from sales in that period.
Penalty if items remain unsold after the last period.
Transition Dynamics
Demand drops as the price increases (price sensitivity).
Demand slightly increases over time due to urgency (people rush to buy near the end).
Random noise is added to simulate uncertainty.

See how a perishable-product pricing problem represents remaining time and inventory, price choices, sales revenue, and penalties for unsold items. section 3.3

How Applied Reinforcement Learning covers it

How deep it goes
Explains the value-loss and pricing trade-off, while leaving domain-specific complexities aside.
Examples
Uses fresh bread, flight tickets, hotel rooms, and seasonal fashion to illustrate limited selling windows.
What you can do after
Understand why unsold inventory creates a pricing trade-off across a limited selling window.
How it connects
Connects perishability to dynamic pricing, price-sensitive demand, demand uncertainty, and inventory constraints.

This explainer was generated by Manning's knowledge graph from the text of Applied Reinforcement Learning. The section links point to the parts of the book it draws on.