1714. Service rate flexibility in diagnostics under congestion
Invited abstract in session MC-11: Scheduling and queuing in healthcare, stream OR in Healthcare (ORAHS).
Monday, 12:30-14:00Room: Clarendon SR 1.03
Authors (first author is the speaker)
| 1. | Kjell Raaijmakers
|
| Industrial Engeneering and Innovation Sciences, Eindhoven University of Technology |
Abstract
In this work, we study a diagnostics model in which an agent must determine the type of arriving customers through imperfect tests. However, testing takes time, during which additional customers arrive, which leads to congestion costs. To mitigate these costs, the agent can adjust the test rate against an increased cost, or release a customer by providing a diagnosis. Misclassifying a customer also incurs a cost, creating a trade-off between testing accuracy, waiting costs, and service rate costs.
We formulate this problem as a Markov Decision Process (MDP) and establish the existence of a search interval: whenever the agent's belief about a customer's type falls within the search interval, testing continues; otherwise, classification occurs. We will demonstrate that, as congestion increases, the test interval shrinks. Furthermore, we show that as congestion increases or as the agent's belief approaches the decision threshold, the optimal policy prescribes a higher service rate despite the associated increased rate costs.
Keywords
- Health Care
- Queuing Systems
- Service Operations
Status: accepted
Back to the list of papers