This paper considers a capacity allocation problem in a two-channel service system. Customers can receive service from either a single-server queueing system, which serves the customers waiting in line one by one, or a clearing service system, which serves a fixed number of customers simultaneously according to its capacity. Customers who join the queueing system should wait till they receive service. In contrast, customers who join the clearing system face the risk of service denial when there are more customers than the clearing system’s capacity. The social planner aims to minimize the total expected cost of all customers by determining the capacities and the arrival rates for the two channels. There are two settings: an unobservable setting where only the expected waiting time information is available and an observable setting where real-time information about the exact workload of the queueing system is known. We also consider the same system under the same settings with strategic customers who choose one of the two channels strategically to minimize their costs. The planner still has the same objective but can now decide only on the capacity allocation. Comparing the performance of the resulting systems allows us to understand the value of coordination and information. Extensions of these systems that serve two customer types are also explored.

We consider an outpatient clinic with strategic patients who choose between making an appointment with an indirect wait cost (advance patients) and walking in with an inconvenience cost that includes the risk of being rejected and waiting in the clinic (walk-ins). Patients have different indirect waiting costs and show up with some probability. The clinic allocates slots to advance and walk-in patients to minimize the expected blockage of walk-in patients. We characterize the equilibrium behavior of patients and investigate the optimal capacity allocation, for unobservable (patients know the expected waiting time) and observable (patients know their exact waiting time) cases. For the unobservable case, one of the three options is optimal: allocating all slots to advance patients, allocating all slots to walk-ins, or allocating a certain number of slots to advance patients so that only urgent patients would choose the walk-in option. In contrast, for the observable case, no such structure exists. We investigate the value of information numerically. Finally, we develop a simulation platform to examine the effects of model assumptions. We find the optimal capacity allocation for the simulation model to benchmark the performance of the theoretical models and two simple policies. These analyses verify that our models work well in realistic simulations, offering a useful tool in practice. In contrast to the common practice of allocating some slots to walk-ins, our results suggest that the clinics should prefer a system that allocates all slots to advance patients in certain environments due to the strategic behavior of patients.

We consider the appointment scheduling for a physician in a healthcare facility. Patients, of two types differentiated by their revenues and day preferences, contact the facility through either a call center to be scheduled immediately or a website to be scheduled the following morning. The facility aims to maximize the long-run average revenue, while ensuring that a certain service level is satisfied for patients generating lower revenue. The facility has two decisions: offering a set of appointment days and choosing the patient type to prioritize while contacting the website patients. Model 1 is a periodic Markov Decision Process (MDP) model without the service-level constraint. We establish certain structural properties of Model 1, while providing sufficient conditions for the existence of a preferred patient type and for the nonoptimality of the commonly used offer-all policy. We also demonstrate the importance of patient preference in determining the preferred type. Model 2 is the constrained MDP model that accommodates the service-level constraint and has an optimal randomized policy with a special structure. This allows developing an efficient method to identify a well-performing policy. We illustrate the performance of this policy through numerical experiments, for systems with and without no-shows. Supplemental Material: The online appendix is available at https://doi.org/10.1287/stsy.2022.0029 .

IntroductionThe number of people diagnosed with dementia is increasing, creating significant economic burden globally. With the progression of the disease, patients need a caregiver whose wellbeing is important for continuous care. Providing respite as a service, through sharing the responsibility of caregiving or support for the caregiver, is a costly initiative. A peer-to-peer online support platform for dementia caregivers, motivated by the sharing economy, putting exchange of knowhow, resources, and services at its center, has the potential to balance cost concerns with a search for respite. The aim of this research is to assess caregivers' intention to engage in peer-to-peer exchange. MethodsA survey including sociodemographic, technology use, and caregiving variables, structured questionnaires (Zarit caregiver burden, WHO brief quality of life scale, ADCS-ADL and chronic stress scale) were administered, January 2018-May 2019, in the dementia outpatient clinic of a university hospital, to a convenience sample of n = 203 individuals identifying themselves as primary caregivers. A path analysis exploring the drivers of an intention to engage in peer-to-peer service exchange was conducted. ResultsIn the path model, caregivers experiencing higher caregiver burden showed higher intention to engage (0.079, p < 0.001). Disease stage had no effect while patient activities of daily living, chronic social role related stressors of the caregiver and general quality of life were significant for the effect on the caregiver burden. Existing household support decreased the caregiver burden, affecting the intention to engage. Caregivers who can share more know-how demonstrate a higher intention to engage (0.579, p = 0.021). Caregiver technology affinity (0.458, p = 0.004) and ability and openness to seek professional help for psychological diagnoses (1.595, p = 0.012) also increased intention to engage. ConclusionThe model shows caregiver burden to be a major driver, along with caregiver characteristics that reflect their technology affinity and openness to the idea of general reciprocity. Existing support for obtaining knowhow and exchanging empathy have a direct effect on the intention to engage. Given the scarcity of caregiver support in the formal care channels, the identified potential of enlarging informal support via a peer-to-peer exchange mechanism holds promise.

Problem definition: We study "comprehensive contracts," which are performance-based reimbursement schemes for managing patients with chronic conditions. If well designed, these contracts can transform primary care beyond the traditional fee-for-service model as they offer a more holistic approach to care by inducing providers to offer various services, including preventive care, chronic disease management, and care coordination. Methodology/results: We employ a stylized model within the Principal-Agent framework to study primary care practice (PCP) capacity allocation under a comprehensive contract offered by the payer. The contract is motivated by two current performance-based contracting systems for primary care in the US: Comprehensive Primary Care Plus and Primary Care First Model reimbursement systems of the Centers for Medicare and Medicaid Services. We first show that a pure fee-for-service contract fails to induce PCPs to allocate capacity for comprehensive care. We then analyze a comprehensive contract offering a capitation payment, a performance bonus, or a combination and show that such a contract is necessary but not sufficient in inducing a comprehensive care effort. Managerial implications: We identify PCP effectiveness in delivering comprehensive care as a crucial consideration in designing comprehensive contracts and show that offering comprehensive contracts to highly effective PCPs enhances access to healthcare and improves outcomes. In addition, we analytically characterize the optimal comprehensive contracts and show that they fail to achieve the first-best outcomes, even under the risk-neutrality of the PCP. Under the optimal contract, however, the PCP's revenue is increasing in their effectiveness level, which is a desirable spillover effect. Our numerical analysis suggests that a comprehensive contract with a strict budget limitation could fail to attract the majority of PCPs, and the payer should consider improving the revenue potential of PCPs for meaningful participation.

How to Assess the Benefits of Coordination in Managing Hospital Resources In providing patient care, hospitals rely on multiple types of resources, such as operating rooms, recovery beds, labs, and diagnostic equipment, that are often controlled and managed as separate entities and by different decision makers. In “Surgical Case-Mix and Discharge Decisions: Does Within-Hospital Coordination Matter?” Hessam Bavafa, Lerzan Örmeci, Sergei Savin, and Vanitha Virudachalam focus on the interaction between “front-end’’ resources, such as operating rooms, and “backroom’’ resources, such as recovery beds, and compare hospital profitability under the fully coordinated, optimal approach to hospital resource management and under alternative decentralized approaches often encountered in practice. The paper identifies settings in which the benefits of coordination are likely to be high as well as settings in which those benefits are at best moderate. In a given hospital, only hospital managers are in a position to estimate with any degree of certainty potential costs of coordinated management of hospital resources, and the paper’s analysis of the benefits of coordination empowers hospital managers to make informed decisions on the desirability of replacing the often decentralized “status quo” by centralized resource management.

Families that visit theme parks like Disneyland are debating on two aspects when they try to determine whether they prefer to join an activity of interest or would rather balk: (1) Is it better to join or balk as a group or allow the flexibility to get separated and jettison some members? and (2) Will it make any difference if they set a ranking among themselves beforehand as to who will be served first, second, etc.? We tackle the effect of flexibility and ranking knowledge and answer the above questions considering a single server Markovian queue with a generic batch size distribution. We consider two levels of flexibility: an inflexible setting, under which a family makes a common decision, and a flexible setting, under which each member makes her own decision. We pair each level with two sublevels with respect to the ranking knowledge: the case where the members set their ranking beforehand, and the case where they do not and assume they will be served according to a random order. We provide a full analytical characterization of the equilibrium and socially optimal strategies, and a comprehensive analysis of the intricate interplay among flexibility, ranking knowledge, and batch size variability, notions that do not exist in single‐ins arrival systems. We offer insights as to under which circumstances entity jettison is preferable. We investigate the corresponding implications of the above on system throughput and social welfare and determine which setting is preferable for the customers and which for the society, depending on the objective and the system dynamics. Further, we highlight key differences between single versus batch‐arrival models and provide high‐level guidelines for managers and policymakers as to how they can influence customer decisions so that they move toward the preferable setting (e.g., by revealing/concealing the ranking, encouraging flexibility, pricing, etc.).

Intensive Care Units (ICUs) are scarce resources that usually operate under high occupancy rates. When faced with limited bed availability, arriving patients are sometimes rejected or admitted by discharging an existing patient early which may result in increased readmission rates or patient/hospital related costs. Yet, there is not a well-defined admission and discharge policy to decrease such adverse consequences. We formulate a comprehensive ICU model with multiple patient types, multiple health stages, and a so-called readmission orbit whose population consists of patients who will seek readmission after an ICU discharge. The ICU model also differentiates first-time and recurring patients. We propose two easy-to-implement heuristic control policies, one based on a Markov Decision Processes model and the other based on a deterministic fluid model. We compare the performances of these policies with the ones of two state-of-the-art policies via discrete-event simulation. According to the numerical results, our proposed policies significantly outperform the state-of-the-art policies by carefully balancing rejection and early-discharge decisions. Moreover, they complement each other since the fluid-based policy performs well in large ICUs that operate in an overloaded regime, while the one-bed policy performs well in small ICUs or ICUs operating in critically or under-loaded regimes.

This paper proposes a new formulation for the dynamic resource allocation problem, which converts the traditional MDP model with known parameters and no capacity constraints to a new model with uncertain parameters and a resource capacity constraint. Our motivating example comes from a medical resource allocation problem: patients with multiple chronic diseases can be provided either normal or special care, where the capacity of special care is limited due to financial or human resources. In such systems, it is difficult, if not impossible, to generate good estimates for the evolution of health for each patient. We formulate the problem as a two-stage stochastic integer program. However, it becomes easily intractable in larger instances of the problem for which we propose and test a parallel approximate dynamic programming algorithm. We show that commercial solvers are not capable of solving the problem instances with a large number of scenarios. Nevertheless, the proposed algorithm provides a solution in seconds even for very large problem instances. In our computational experiments, it finds the optimal solution for $42.86\%$ of the instances. On aggregate, it achieves $0.073\%$ mean gap value. Finally, we estimate the value of our contribution for different realizations of the parameters. Our findings show that there is a significant amount of additional utility contributed by our model.

Abstract Background Patient activation (PA), which is known to improve health outcomes, describes the knowledge, skills and confidence a person has in managing their own health care. We investigated the extent of PA and associated factors in adults with diabetes (DM) and/or hypertension (HT). This study is the first in Turkey evaluating PA, using the Patient Activation Measure (PAM) scale. The results of the study provide practitioners with information on the characteristics of patients who need support to increase their activation. Methods We conducted this cross-sectional study in 14 Family Health Centers in Istanbul. The participants were DM and/or HT patients. A questionnaire including the PAM, questions on patient characteristics, life style behaviors, healthcare utilization and health status was applied to 431 patients. Based on PAM score, patients were classified into two activation levels: level 1-2 (poor activation) and level 3-4 (good activation). χ2, t-test and logistic regression (LR) analysis were used. LR analysis was performed for all participants and for women and men separately. Results Of 431 patients (mean age: 63.6), 65% were women; 45% had a poor activation level (PAL). Based on LR analysis; low socioeconomic status (SES) (OR = 1.6; 95% CI:1.01-2.5), being illiterate (OR = 3.9, 95% CI: 1.5-10.7), being primary school graduate (OR = 2.1, 95% CI:1.1-4.2), lack of adult vaccination (OR = 1.9, 95% CI:1.1-3.1), higher BMI (OR = 1.1, 95% CI: 1.09-1.13) and worse self-reported health (OR = 1.2, 95% CI: 1.1-1.3) were factors associated with low PAL. The latter was associated with low PAL for both sexes; high BMI was an associated factor only among women, while low SES and lack of vaccination were factors only in men. Conclusions Almost half of the patients had low activation level in our sample. Associated factors may serve as the basis for the development of interventions needed to enhance activation for patients with DM/HT. Key messages Low patient activation is associated with low level of education, low SES, high BMI, lack of vaccination and worse self-reported health. The factors associated with patient activation vary by gender. A qualitative study with patients having different levels of activation would be useful to understand the underlying motivations.

We consider the appointment scheduling process of a physician in a healthcare facility. There are multiple patient types with different priorities in this facility. The facility observes the number of appointment requests from each patient type at the beginning of each day. The facility decides on how to allocate the arriving appointment requests to available slots over the booking horizon. Each type of pa- tient prefers a day in the booking horizon with a specific probability. We model this system with a constrained Markov Decision Process to maximize the infinite-horizon expected discounted revenue subject to the constraint that the infinite-horizon ex- pected discounted rejection cost is below a specific threshold. Patients have only one preference for the appointment day. Each patient is either given an appointment on the day he/she prefers or the appointment request of that patient is denied. We prove that the optimal policy is a randomized booking limit policy. To solve the model, we use Approximate Dynamic Programming (ADP) techniques. We conduct numerical experiments and compare the results obtained with ADP techniques with some benchmark policies.

Hospital care is one of the fundamental components in any healthcare delivery system. Within hospital care, surgical procedures account for the largest share of the revenue generated. Traditionally, the operating room (OR) capacity is viewed as a major constraint limiting a hospital's ability to increase the number of surgical procedures and the accompanying revenues. However, each procedure consumes not only the OR capacity but also, the hospital bed capacity. In “Managing Portfolio of Elective Surgical Procedures: A Multidimensional Inverse Newsvendor Problem,” Hessam Bavafa, Charles M. Leys, Lerzan Örmeci, and Sergei Savin investigate the effects of the interaction between the two resources (i.e., OR and recovery beds) on the optimal number of elective surgical procedures to be performed daily. They evaluate the performance of the “front-end” approach, which considers only the OR capacity, in different settings. Moreover, they show how the variability of the resource utilization by surgical procedures affects the optimal elective portfolio.

In standard stochastic dynamic programming, the transition probability distributions of the underlying Markov Chains are assumed to be known with certainty. We focus on the case where the transition probabilities or other input data are uncertain. Robust dynamic programming addresses this problem by defining a min‐max game between Nature and the controller. Considering examples from inventory and queueing control, we examine the structure of the optimal policy in such robust dynamic programs when event probabilities are uncertain. We identify the cases where certain monotonicity results still hold and the form of the optimal policy is determined by a threshold. We also investigate the marginal value of time and the case of uncertain rewards.© 2017 Wiley Periodicals, Inc. Naval Research Logistics 65: 699–716, 2018

Most service systems consist of multidepartmental structures with multiskill agents that can deal with several types of service requests. The design of flexibility in terms of agents' skill sets and assignments of requests is a critical issue for such systems. The objective of this study was to identify preferred flexibility structures when demand is random and capacity is finite. We compare structures recommended by the flexibility literature to structures we observe in practice within call centers. To enable a comparison of flexibility structures under optimal capacity, the capacity optimization problem for this setting is formulated as a two‐stage stochastic optimization problem. A simulation‐based optimization procedure for this problem using sample‐path gradient estimation is proposed and tested, and used in the subsequent comparison of the flexibility structures being studied. The analysis illustrates under what conditions on demand, cost, and human resource considerations, the structures found in practice are preferred.

We present a modeling framework for facilities that provide both screening (preventive) and diagnostic (repair) services. The facility operates in a random environment that represents the condition of the population that needs screening and diagnostic services, such as the disease prevalence level. We model the environment as a partially endogenous process: the population’s health can be improved by providing screening services, which reduces future demand for diagnostic services. We use event-based dynamic programming to build a framework for modeling different kinds of these facilities. This framework contains a number of service priority policies that are concerned with prioritizing screening versus diagnostic services. The main trade-off is between serving urgent diagnostic needs and providing screening services that may decrease future diagnostic needs. Under certain conditions, this trade-off reverses the famous cμ rule; i.e., the patients with lower waiting cost are given priority over the others. We define appropriate event operators and specify the properties preserved by these operators. These characterize the structure of optimal policies for all models that can be built within this framework. A numerical study on colonoscopy services illustrates how the framework can be used to gain insights on developing good screening policies.

The healthcare sector is facing major challenges worldwide in terms of higher demands for efficiency, quality and equity.Operational Research (OR) techniques offer valuable tools for improving the design and delivery of healthcare services (e.g., inpatient and outpatient care, admission processes, emergency services, home healthcare, etc.), as well as improving the decision-making processes and implementation of public health policies.The EURO Working Group on Operational Research Applied to Health Services (ORAHS) was formed in 1975 as a special interest group within the European branch (EURO) of the International Federation of Operational Research Societies (IFORS) [1].The group holds international conferences every year.The goal is to bring together the academicians and practitioners for disseminating ideas, knowledge and research on the application of OR methods to healthcare.

This paper explores how the capacity of colonoscopy services should be allocated for screening and diagnosis of colorectal cancer to improve health outcomes. Both of these services are important since screening prevents cancer by removing polyps, while diagnosis is required to start treatment for cancer. This paper first presents a basic compartmental model to illustrate the trade‐off between these two analytically. Further, a more realistic population dynamics model with resource constraints is introduced for colorectal cancer screening and analyzed numerically. The best resource allocation decisions are investigated with the objectives of minimizing mortality or incidence rates. We provide a sensitivity analysis with respect to policy and disease‐related parameters. We conclude that to minimize mortality, the capacity should be rationed to ensure that the wait for diagnosis is at reasonable levels. When the relevant performance measure is the incidence rate, screening is allocated more capacity compared to the case with mortality rate measure. We also show that benefits from increasing compliance to screening programs can only be realized if there is sufficient service capacity.