In the nested simulation literature, a common assumption is that the experimenter can choose the number of outer scenarios to sample. This paper considers the case when the experimenter is given a fixed set of outer scenarios from an external entity. We propose a nested simulation experiment design that pools inner replications from one scenario to estimate another scenario’s conditional mean via the likelihood ratio method. Given the outer scenarios, we decide how many inner replications to run at each outer scenario as well as how to pool the inner replications by solving a bilevel optimization problem that minimizes the total simulation effort. We provide asymptotic analyses on the convergence rates of the performance measure estimators computed from the optimized experiment design. Under some assumptions, the optimized design achieves [Formula: see text] mean squared error of the estimators given simulation budget [Formula: see text]. Numerical experiments demonstrate that our design outperforms a state-of-the-art design that pools replications via regression. History: Accepted by Bruno Tuffin, Area Editor for Simulation. Funding: This work was supported by the National Science Foundation [Grant CMMI-2045400] and the Natural Sciences and Engineering Research Council of Canada [Grant RGPIN-2018-03755]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0392 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0392 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Discrete-event simulation models generate random variates from input distributions and compute outputs according to the simulation logic. The input distributions are typically fitted to finite real-world data and thus are subject to estimation errors that can propagate to the simulation outputs: an issue commonly known as input uncertainty (IU). This paper investigates quantifying IU using the output confidence intervals (CIs) computed from bootstrap quantile estimators. The standard direct bootstrap method has overcoverage due to convolution of the simulation error and IU; however, the brute-force way of washing away the former is computationally demanding. We present two new bootstrap methods to enhance direct resampling in both statistical and computational efficiencies using shrinkage strategies to down-scale the variabilities encapsulated in the CIs. Our asymptotic analysis shows how both approaches produce tight CIs accounting for IU under limited input data and simulation effort along with the simulation sample-size requirements relative to the input data size. We demonstrate performances of the shrinkage strategies with several numerical experiments and investigate the conditions under which each method performs well. We also show advantages of nonparametric approaches over parametric bootstrap when the distribution family is misspecified and over metamodel approaches when the dimension of the distribution parameters is high. History: Accepted by Bruno Tuffin, Area Editor for Simulation. Funding: This work was supported by the National Science Foundation [CAREER CMMI-1834710, CAREER CMMI-2045400, DMS-1854659, and IIS-1849280]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0044 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0044 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Often in manufacturing systems, scenarios arise where the demand for maintenance exceeds the capacity of maintenance resources. This results in the problem of allocating the limited resources among machines competing for them. This maintenance scheduling problem can be formulated as a Markov decision process (MDP) with the goal of finding the optimal dynamic maintenance action given the current system state. However, as the system becomes more complex, solving an MDP suffers from the curse of dimensionality. To overcome this issue, we propose a two-stage approach that first optimizes a static condition-based maintenance (CBM) policy using a genetic algorithm (GA) and then improves the policy online via Monte Carlo tree search (MCTS). The static policy significantly reduces the state space of the online problem by allowing us to ignore machines that are not sufficiently degraded. Furthermore, we formulate MCTS to seek a maintenance schedule that maximizes the long-term production volume of the system to reconcile the conflict between maintenance and production objectives. We demonstrate that the resulting online policy is an improvement over the static CBM policy found by GA. Note to Practitioners—: This article proposes a method of scheduling maintenance in complex manufacturing systems in scenarios where there is frequent competition for maintenance resources. We use a condition-based maintenance policy that prescribes maintenance actions based on a machine's current health. However, when several machines are due for maintenance, a maintenance technician must choose between multiple competing jobs. While a common approach is to establish rules that dictate how maintenance jobs should be prioritized, such as the first-in, first-out rule, the goal of this work is to improve upon static policies in real time. We do this by strategically evaluating sequences of maintenance actions and playing out many "what-if" scenarios to see how the system will behave in the future. Implementation of the proposed method relies on the construction of a simulation model of the target system. This model is capable of retrieving the current state of the physical system, including the degradation state of machines, the availability of maintenance resources, and the distribution of parts throughout buffers in the system. We present several simulation experiments that demonstrate the improvement in system performance that our approach provides. Future work will aim to improve the efficiency of maintenance prioritization through online learning as well as more accurately identify manufacturing system configurations that will yield the greatest benefit of these methods.

Inference-based optimization via simulation, which substitutes Gaussian process (GP) learning for the structural properties exploited in mathematical programming, is a powerful paradigm that has been shown to be remarkably effective in problems of modest feasible-region size and decision-variable dimension. The limitation to “modest” problems is a result of the computational overhead and numerical challenges encountered in computing the GP conditional (posterior) distribution on each iteration. In this paper, we substantially expand the size of discrete-decision-variable optimization-via-simulation problems that can be attacked in this way by exploiting a particular GP—discrete Gaussian Markov random fields—and carefully tailored computational methods. The result is the rapid Gaussian Markov Improvement Algorithm (rGMIA), an algorithm that delivers both a global convergence guarantee and finite-sample optimality-gap inference for significantly larger problems. Between infrequent evaluations of the global conditional distribution, rGMIA applies the full power of GP learning to rapidly search smaller sets of promising feasible solutions that need not be spatially close. We carefully document the computational savings via complexity analysis and an extensive empirical study. Summary of Contribution: The broad topic of the paper is optimization via simulation, which means optimizing some performance measure of a system that may only be estimated by executing a stochastic, discrete-event simulation. Stochastic simulation is a core topic and method of operations research. The focus of this paper is on significantly speeding-up the computations underlying an existing method that is based on Gaussian process learning, where the underlying Gaussian process is a discrete Gaussian Markov Random Field. This speed-up is accomplished by employing smart computational linear algebra, state-of-the-art algorithms, and a careful divide-and-conquer evaluation strategy. Problems of significantly greater size than any other existing algorithm with similar guarantees can solve are solved as illustrations.