Hybrid Discrete–Continuous Simulation for Dynamic Pricing with Strategic Customers and Evolving Willingness-to-Pay

Dynamic pricing in markets with forward-looking customers is shaped by strategic purchase delays, inventory constraints, and changing willingness-to-pay (WTP). Modeling these interactions is difficult because market activity occurs through discrete events, while latent customer valuations may evolve continuously after arrival. We develop an executable hybrid simulation framework that combines a discrete-event market engine with continuous-time stochastic WTP processes. Customer purchase timing follows a reduced-form threshold rule that balances current surplus against anticipated markdowns and waiting costs, including stockout risk. The event-driven implementation propagates WTP between scheduled market events and detects threshold crossings using a local numerical grid, interpolation, and an explicit near-tie convention. We evaluate the framework using controlled synthetic benchmarks, comparing it with fixed-step and static-WTP alternatives. Monte Carlo confidence intervals, paired policy comparisons, numerical-sensitivity tests, and synthetic-data recovery assess computational credibility and parameter support. The experiments reveal regime-dependent policy performance. Strong demand-responsive pricing can become fragile when strategic waiting suppresses early demand and induces further markdowns. Preannounced and robust pricing often stabilize purchase timing and revenue, whereas inventory scarcity discourages delay by increasing stockout risk. Recovery diagnostics also indicate that some latent parameters are only weakly separated by the available summaries. The framework provides a reproducible computational laboratory for studying how strategic timing, evolving valuations, inventory, and numerical execution jointly affect dynamic-pricing outcomes, while distinguishing controlled mechanism analysis from empirical market validation.