M otivated by service capacity-management problems in healthcare contexts, we consider a multiresource allocation problem with two classes of jobs (elective and emergency) in a dynamic and nonstationary environment. Emergency jobs need to be served immediately, whereas elective jobs can wait. Distributional information about demand and resource availability is continually updated, and we allow jobs to renege. We prove that our formulation is convex, and the optimal amount of capacity reserved for emergency jobs in each period decreases with the number of elective jobs waiting for service. However, the optimal policy is difficult to compute exactly. We develop the idea of a limit policy starting at a particular time, and use this policy to obtain upper and lower bounds on the decisions of an optimal policy in each period, and also to develop several computationally efficient policies. We show in computational experiments that our best policy performs within 1 8% of an optimal policy on average.

Authors are encouraged to submit new papers to INFORMS journals by means of a style file template, which includes the journal title. However, use of a template does not certify that the paper has been accepted for publication in the named journal. INFORMS journal templates are for the exclusive purpose of submitting to an INFORMS journal and should not be used to distribute the papers in print or online or to submit the papers to another publication.

We study the inventory replenishment of a product whose demand can be manipulated by restricting the supply. This research is motivated by a novel marketing tactic employed by manufacturers of fashion and luxury items. Such a tactic combines innovative marketing with deliberate understocking in an attempt to create shortages (i.e., waitlists) that add to the allure and sense of exclusivity of a product and stimulate its demand. We model the problem as a finite-horizon, periodic-review system where demand in each period is a decreasing function of the net ending inventory in the previous period. Although the optimal structure can be complex in general, under certain conditions we are able to characterize the optimal policy as a state-dependent, monotone, base-stock policy. We compare this policy with the optimal policy for the case in which demand is independent of the net inventory. We also show that understocking is optimal in various scenarios. We then propose a novel strategy, called the inventory-withholding strategy, to further explore the wait-list effect by making customers wait even when there is inventory on hand to satisfy them. Our numerical experiments study the impact of various model parameters in combination with the wait-list effect on the optimal policy and the corresponding expected profits.

We study a fundamental model of resource allocation in which a finite number of resources must be assigned in an online manner to a heterogeneous stream of customers. The customers arrive randomly over time according to known stochastic processes. Each customer requires a specific amount of capacity and has a specific preference for each of the resources with some resources being feasible for the customer and some not. The system must find a feasible assignment of each customer to a resource or must reject the customer. The aim is to maximize the total expected capacity utilization of the resources over the horizon. This model has application in services, freight transportation, and online advertising. We present online algorithms with bounded competitive ratios relative to an optimal off-line algorithm that knows all stochastic information. Our algorithms perform extremely well compared with common heuristics as demonstrated on a real data set from a large hospital system in New York City. This paper was accepted by Yinyu Ye, optimization.

Despite the fact that hospital care is often delivered in successive stages, current healthcare scheduling and capacity planning methods usually treat different hospital units in isolation. To address such a shortcoming, we introduce the first Markov decision process model for scheduling surgical patients on a daily basis, explicitly taking into account patient length‐of‐stay in hospital after surgeries and inpatient census. By way of a simple and yet innovative variable transformation, we reveal the hidden submodularity structure in our model. This transformation, in particular, allows us to show that the optimal number of patients to admit increases when the waitlist of surgical patients is longer, given the number of patients recovering downstream is fixed. We conduct extensive simulation experiments to study the applicability of our theoretical model in various settings. Our simulations based on real data demonstrate substantial values in making integrated scheduling decisions that simultaneously consider capacity usage at all locations in a hospital, especially when demand and system capacities are balanced or more elective patients present in the patient mix. The traditional scheduling policy, which is solely driven by operating room usage, however, can lead to significantly suboptimal use of downstream capacity and, as our numerical experiments show, may result in up to a three‐fold increase in total expenses. In contrast, a scheduling policy based on downstream capacity usage often performs relatively close to the integrated scheduling policy, and therefore may serve as a simple, effective scheduling heuristic for hospital managers—especially when the downstream capacity is costly and less flexible.

Purpose -The performance measure approach (PMA) is widely adopted for reliability analysis and reliability-based design optimization because of its robustness and efficiency compared to reliability index approach. However, it has been reported that PMA involves repeat evaluations of probabilistic constraints therefore it is prohibitively expensive for many large-scale applications. In order to overcome these disadvantages, the purpose of this paper is to propose an efficient PMA-based reliability analysis technique using radial basis function (RBF). Design/methodology/approach -The RBF is adopted to approximate the implicit limit state functions in combination with latin hypercube sampling (LHS) strategy. The advanced mean value method is applied to obtain the most probable point (MPP) with the prescribed target reliability and corresponding probabilistic performance measure to improve analysis accuracy. A sequential framework is proposed to relocate the sampling center to the obtained MPP and reconstruct RBF until a criteria is satisfied. Findings -The method is shown to be better in the computation time to the PMA based on the actual model. The analysis results of probabilistic performance measure are accurately close to the reference solution. Five numerical examples are presented to demonstrate the effectiveness of the proposed method. Originality/value -The main contribution of this paper is to propose a new reliability analysis technique using reconstructed RBF approximate model. The originalities of this paper may lie in: investigating the PMA using metamodel techniques, using RBF instead of the other types of metamodels to deal with the low efficiency problem. IntroductionTo gain the increasingly global, competitive market, manufacturing companies strive to produce more economic and more reliable products. Therefore, reliability analysis and associated reliability-based design optimization techniques are received to obtain reliable products.During the past two decades, much effort has been made to develop efficient methods for reliability analysis problem. The commonly used reliability methods include Monte Carlo simulation (MCS), first-order reliability method (FORM), and second-order reliability method (SORM). For reliability-based design optimization (RBDO) problems, the approaches to evaluate probabilistic constraints can be categorized into reliability index approach (RIA) and performance measure approach (PMA). Youn and Choi (2004a) compared influences of different reliability analysis approaches employed in RIA and PMA on the behaviors of nonlinearity for RIA and PMA in the RBDO process. Choi and Youn (2003) proposed a hybrid mean value (HMV) method for effective evaluation of probabilistic constraints in the RBDO process in order to take advantages of PMA. Youn and Choi (2004b) proposed a response surface (RS) methodology for RBDO by integrating the proposed response surface method (RSM) and the HMV method in PMA, which is specifically developed for reliability analysis and optimization. Cheng et ...

General anesthesia or monitored anesthesia care sometimes is provided in nonoperating room (OR) locations during nights and weekends (e.g., for magnetic resonance imaging [MRI] or computerized tomography [CT]). Rational and consistent scheduling and sequencing decisions for these diagnostic imaging procedures, including coordination with OR cases, cannot be done without knowing how long each case can wait to be started without risking a worsening of the patient's condition. We reviewed the medical records of the 81 patients who underwent diagnostic imaging procedures (78 = MRI, 3 = CT scan) under general anesthesia or monitored anesthesia care either on weekends or between 6 pm and 6 am at the University of Iowa Hospitals between March 2012 and February 2014. For 77.8% of patients, the indications could have changed clinical management within 4 hours (N = 63/81). Among the 63 imaging studies with potential immediate impact, there was documentation of results having been communicated to the treating team within 4 hours of the completion of imaging for 39 of the patients. Among the 39 patients, 15 promptly received medications or underwent procedures based on the imaging results. Thus, 15 of the 81 patients had a change in care (18.5%, 95% lower confidence limit = 11.2%). Our results are important since we showed previously that it is not possible to make rational and consistent decisions in case sequencing without knowing how long each case (including diagnostic imaging procedures) can wait to be started without a change in the patient's risk. The scheduled surgical procedure itself provides sufficient information to assess safe waiting times to start add-on cases (e.g., appendectomy). In contrast, MRI provides no context as to how potential findings will influence treatment. Our results show that the assumption cannot reasonably be made when sequencing cases that all imaging studies can or cannot wait longer than pending surgical procedures. Our results show that, for evidence-based OR management decision-making, information to decide appropriate waiting should be obtained electronically or verbally for each imaging study.

No abstract