arXiv subjects

Runhua Wu

Publications and source records attributed to Runhua Wu.

3 recordsLinked to original sources

arXiv2026

Maximizing Throughput in an M/G/1 Queue with Customer Abandonments

This paper studies the problem of identifying the optimal server assignment policy in single-server queues with customer abandonment. We consider a system with Poisson arrivals and exponentially distributed patience times. We show that when service times follow either an Erlang-$K$ or a hyperexponential distribution and the decision maker can observe the phase of a customer's service time, the Shortest Remaining Expected Processing Time (SREPT) policy maximizes the long-run average throughput, independent of the abandonment rate.

math.OC↗

arXiv2026

CAST: Causal Anchored Simplex Transport for Distribution-Valued Time Series

Many decision-facing stochastic systems are observed through aggregate distributions rather than scalar trajectories: queue occupancies, mobility shares, public-health mixtures, generation-source shares, ecological compositions, and air-quality severity profiles all live on the probability simplex and evolve over time. We study causal (online) forecasting for these distribution-valued time series and argue that the transition operator itself should be structured around the simplex. We introduce CAST (Causal Anchored Simplex Transport), a successor-local operator that (i) retrieves empirical successors from causal context, (ii) stabilizes them with a persistence anchor, and (iii) applies a bounded local stochastic transport on ordered supports; every stage preserves the simplex by construction. We identify a structural failure mode, latent transition-kernel aliasing, where similar observed distributions evolve differently under different contextual regimes, and prove that any forecaster depending only on an aliased summary incurs an irreducible weighted Jensen-Shannon excess-risk lower bound, while the CAST hypothesis class contains the regime-aware Bayes successor; for ordered supports an additional Pinsker separation holds whenever the transported successor lies outside the no-transport anchor hull. On eleven public and simulated benchmarks spanning ecology, energy, diet, mortality, employment, air quality, severe weather, mobility, and G/G/1, G_t/G/1 queue occupancy, CAST attains the best average rank on both one-step KL (1.27) and autoregressive rollout JSD (1.91), winning 8/11 sections on each metric against a broad statistical, compositional, recurrent, convolutional, and Transformer baseline set, and top-2 on all 11 sections for offline KL. Component ablations and a controlled synthetic aliasing experiment corroborate the theory.

stat.ML↗

arXiv2025

Service Rate Control in Queues with Abandonments

We consider a Markovian single server queue with impatient customers. There is a customer abandonment cost and a holding cost for customers in the system. We consider two versions of the problem. In the first version, customers pay a reward at the time of arrival whereas in the second version, reward is received at the time of service completion. Service rate attains values in a compact set and there is a cost associated with each service rate. Under these assumptions, our objective is to characterize the service rate policy that maximizes the infinite-horizon discounted reward and the long-run average reward. We show that for systems with an infinite buffer, the optimal service rate policy is monotone. However, the optimal policy is not necessarily monotone when the buffer size is finite. Furthermore, we prove that the set of possible optimal actions can be reduced to the lower boundary of the convex hull of the action space and develop an efficient policy iteration algorithm. Finally, we prove that the optimal service rate converges as the state tends to infinity. This property enables us to truncate the state space, facilitating the numerical computation of the optimal rate when the buffer space is infinite.

math.OC↗