Fully sequential selection procedures with control variates

In many practical problems, simulation models are used to support complex decision-making processes comparing hundreds or thousands of solutions. These problems typically have a key objective but the final decision may be dependent on other factors, which cannot be incorporated into the simulation model. In such cases, decision-makers may request a short list of ‘good’ solutions, which work well for the main objective and satisfy one or more chance constraints. While fully sequential ranking and selection procedures can be effective at solving these problems, surveys of experimentation practice suggest that they are under-utilized, potentially due to difficulties automating commercial software. We develop an approach with just two stages of replications. The approach, which has been designed to cope with the use of common random numbers, draws on ideas from indifference zones and makes use of bootstrapping to find a subset of high quality solutions. A Python implementation is freely available.

Common random numbers in multivariate simulations

We describe major improvements to the testing capabilities of SimOpt, a library of simulation-optimization problems and solvers. Foremost among these improvements is a transition to GitHub that makes SimOpt easier to use and maintain. We also design two new wrapper functions that facilitate empirical comparisons of solvers. The wrapper functions make extensive use of common random numbers (CRN) both within and across solvers for various purposes; e.g., identifying random initial solutions and running simulation replications. We examine some of the intricacies of using CRN to compare simulation-optimization solvers.

The double exponentially weighted moving average (dEWMA) feedback controller is a popular run-to-run (RTR) controller that is used to adjust semiconductor manufacturing processes that have a linear drift. Although this controller, with suitable fixed discount factors, can guarantee long-term stability, it usually requires a moderately large number of runs to bring the process output to its target value. To overcome this difficulty, an enhanced controller called the “adaptive variable EWMA controller,” is proposed. This controller brings the process output to a desired target much more quickly. An analytical expression of the process output of this controller and its long-term stability conditions are derived. Furthermore, examples are presented that compare the performance of the proposed controller with several competing controllers. It is demonstrated that the proposed controller has the best performance of the considered controllers in terms of the reduction of total mean square errors. [Supplementary materials are available for this article. Go to the publisher's online edition of IIE Transactions for the following supplemental resource: Appendix]

The Exponentially Weighted Moving Average (EWMA) feedback controller is a popular model-based run-to-run feedback controller that primarily uses data from previous process runs to adjust the settings of the next run. The long-term stability conditions and the transient performance of the EWMA controller have received considerable attention in the literature. Most of these studies have assumed that the process Input–Output (I-O) relationship is static and simply considered colored noise models for the process disturbance. However, process dynamics and disturbance dynamics may occur simultaneously. Under this circumstance, using EWMA-based controllers will usually lead to an unsatisfactory performance. To overcome this weakness, this article first proposes a quasi-minimum mean square error controller. The theoretical results of the long-term stability conditions are derived under a first-order transfer function model together with a general disturbance model. Furthermore, comprehensive simulation studies are conducted to compare the proposed controller with existing popular controllers. The results demonstrate that the proposed controller outperforms those popular controllers for most cases, even when the process I-O model is mis-specified or the process parameters are not precisely estimated. Supplementary materials are available for this article. Go to the publisher's online edition of IIE Transactions for the following supplemental resources: Tables 5 to 7 and Appendix.

Objective: To highlight the issues and discuss the research evidence regarding safety, mobility, and other consequences of different licensing ages. Methods: Information included is based on presentations and discussions at a 1-day workshop on licensing age issues and a review and synthesis of the international literature. Results: The literature indicates that higher licensing ages are associated with safety benefits. There is an associated mobility loss, more likely to be an issue in rural states. Legislative attempts to raise the minimum age for independent driving in the United States—for example, from 16 to 17—have been resisted, although in some states the age has been raised indirectly through graduated driver licensing (GDL) policies. Conclusions: Jurisdictions can achieve reductions in teenage crashes by raising the licensing age. This can be done directly or indirectly by strengthening GDL systems, in particular extending the minimum length of the learner period. Supplementary materials are available for this article. Go to the publisher's online edition of Traffic Injury Prevention for the following supplemental resource: List of workshop participants.

We describe a practical two-stage algorithm, BootComp, for multi-objective optimization via simulation. Our algorithm finds a subset of good designs that a decision-maker can compare to identify the one that works best when considering all aspects of the system, including those that cannot be modeled. BootComp is designed to be straightforward to implement by a practitioner with basic statistical knowledge in a simulation package that does not support sequential ranking and selection. These requirements restrict us to a two-stage procedure that works with any distributions of the outputs and allows for the use of common random numbers. Comparisons with sequential ranking and selection methods suggest that it performs well, and we also demonstrate its use analyzing a real simulation aiming to determine the optimal ward configuration for a UK hospital.

The design and analysis of Monte Carlo experiments, with special reference to structural equation modelling, is discussed in this article. These topics merit consideration, since the validity of the conclusions drawn from a Monte Carlo study clearly hinges on these features. It is argued that comprehensive Monte Carlo experiments can be implemented on a PC if the experiments are adequately designed. This is especially important when investigating modern computer intensive methodologies like resampling and Markov Chain Monte Carlo methods. We are faced with three fundamental challenges in Monte Carlo experimentation. The first problem is statistical precision, which concerns the reliability of the obtained results. External validity, on the other hand, depends on the number of experimental conditions, and is crucial for the prospects of generalising the results beyond the specific experiment. Finally, we face the constraint on available computer resources. The conventional wisdom in designing and analysing Monte Carlo experiments embodies no explicit specification of meta-model for analysing the output of the experiment, the use of case studies or full factorial designs as experimental plans, no use of variance reduction techniques, a large number of replications, and "eyeballing" of the results. A critical examination of the conventional wisdom is presented in this article. We suggest that the following alternative procedures should be considered. First of all, we argue that it is profitable to specify explicit meta-models, relating the chosen performance statistics and experimental conditions. Regarding the experimental plan, we recommend the use of incomplete designs, which will often result in considerable savings. We also consider the use of common random numbers in the simulation phase, since this may enhance the precision in estimating meta-models. The use of fewer replications per trial, enabling us to investigate an increased number of experimental conditions, should also be considered in order to improve the external validity at the cost of the conventionally excessive precision.

The problem of constructing confidence interval for the largest of k normal means considered by Saxena and Tong (1969) and Tong (1970, 1973) is extended by considering the case when the populations have unequal and unknown variances. Several multi-stage estimation procedures (like, two-stage, three-stage and accelerated sequential procedures) are developed to deal with the estimation problem. Nice asymptotic properties of these estimation procedures are studied and their relative advantages and disadvantages are discussed.

The negative consequences of peer victimization on psychosocial adjustment are well documented. The consequences, however, may depend on who the bullies are. In this study, we examined the consequences of same- versus other-sex victimization. The sample consisted of 4,941 Finnish adolescents (ages 14–15; 47.7% boys). We used structural equation modeling to examine both concurrent and longitudinal associations of same- and other-sex victimization with depression, negative perception of peers, and social self-esteem. Both same- and other-sex victimization were related to psychosocial adjustment. Concurrently, the victimization experiences with same-sex peers in particular were associated with generalized cognitions about peers, whereas being bullied by other-sex peers was related to adolescents’ social self-esteem more strongly than victimization by same-sex peers. The longitudinal associations, in turn, showed that only being bullied by boys had carry-over effects on girls’ adjustment. Other-sex victimization can have serious consequences especially on girls’ psychosocial adjustment. [Supplementary material is available for this article. Go to the publisher's online edition of the Journal of Clinical Child & Adolescent Psychology for the following supplemental resource: Complementary information on model fit indices and the nested model chi-square difference tests.]

Revictimization and life stressors are common among dissociative disorder (DD) patients, yet no studies have examined the prevalence rates for these experiences or their relationships with treatment outcome. This study aimed to examine the rates of revictimization and victimization of others using therapist–DD patient pairs from the naturalistic Treatment of Patients with Dissociative Disorders (TOP DD) study while also considering the role of revictimization and life stressors among 49 patients who greatly improved or worsened during 30 months of treatment. Therapists reported that sexual and physical revictimization in the previous 6 months was high among the patients (3.5%–7.0% and 4.1%–7.1% in the overall TOP DD sample, respectively), and emotional revictimization was quite high (29%–36%). Revictimization showed a decreasing trend over the 30 months of the study. Therapists reported that more than a quarter of the patients who were revictimized were also occasionally emotionally or physically abusive to others. More patients showed sudden improvement versus sudden worsening in patient-reported symptoms at 1 or more time point(s). Patients who improved had significantly fewer revictimizations and stressors overall than patients who worsened, suggesting that revictimization and/or stressors may contribute to worsening in treatment. Further research is needed to learn more about the roles of revictimization, victimization of others, and stressors in DD treatment. [Supplementary material is available for this article. Go to the publisher's online edition of Journal of Trauma & Dissociation for the following supplemental resource: Baseline Demographic Information of TOP DD Improving and Worsening Subgroups]

This study introduces two semi-quantitative methods, Structured Subjective Assessment (SSA) and Control of Substances Hazardous to Health (COSHH) Essentials, in conjunction with two-dimensional Monte Carlo simulations for determining prior probabilities. Prior distribution using expert judgment was included for comparison. Practical applications of the proposed methods were demonstrated using personal exposure measurements of isoamyl acetate in an electronics manufacturing facility and of isopropanol in a printing shop. Applicability of these methods in real workplaces was discussed based on the advantages and disadvantages of each method. Although these methods could not be completely independent of expert judgments, this study demonstrated a methodological improvement in the estimation of the prior distribution for the Bayesian decision analysis tool. The proposed methods provide a logical basis for the decision process by considering determinants of worker exposure. [Supplementary materials are available for this article. Go to the publisher's online edition of Journal of Occupational and Environmental Hygiene for the following free supplementary resource: Coding of the MC Simulation (SSA Method)]

An original one-pot synthesis of bromomethyl ketones is achived using the Petasis reagent (dimethyltitanocene) as a key for enamine generation. Several amides were used to test the limits of the procedure by changing either the alkyl chain R or the amino portion of the starting materials. The enamines generated in situ were allowed to react with bromine at low temperature followed by hydrolysis to yield bromomethyl ketones in excellent yields (85 to 95%). Mechanistic details and optimum conditions for the reaction are briefly discussed. The present approach offers several advantages such as regioselectivity in enamine formation, good yields, mild reaction conditions, and ease of experimentation. [Supplementary materials are available for this article. Go to the publisher's online edition of Synthetic Communications® for the following free supplemental resource: Full experimental and spectral details.]

Hydrogen peroxide was employed as an oxidizing agent for a new method for trapping unstable gaseous arsines. Resin cartridges wetted with hydrogen peroxide trapped over 90% of the gaseous arsines (arsine, methylarsine, dimethylarsine, and trimethylarsine) from laboratory samples. The oxidized products (arsenic acid, methylarsonic acid, dimethylarsinic acid, and trimethylarsine oxide) are relatively stable in Milli-Q water, so they could be extracted from the cartridges and analyzed by liquid chromatography-inductively coupled plasma-mass spectrometry at nanogram per milliliter concentrations. We also used our new method to determine arsine concentrations in field samples from Nishinomaki mine and Tamagawa hot spring in Japan. Inorganic arsine was detected at the highest concentration for each sampling point, and trimethylarsine was observed in the Nishinomaki mine and at one sampling site of the Tamagawa hot spring. This trimethylarsine may have been produced by a variety of microorganisms. Supplementary materials are available for this article. Go to the publisher's online edition of Analytical Letters for the following free supplemental resource(s): tables.

This study assessed a professional pipefitter/welder performing shielded metal arc welding on carbon steel under field conditions. The resulting breathing zone (near field) and area (far field) welding fume concentration data were applied to the two-zone model for the purpose of determining field-derived personal exposure emission (generation) rates during actual welding work. The study is unique in that one welder was evaluated under high production conditions for 2 days at two different welding locations: a boiler room and a breezeway. Samples were collected and analyzed for total particulate following NIOSH Method 0500 and for select metals following NIOSH Method 7300. Breezeway average personal breathing zone sample total particulate concentrations ranged from 2.89 mg/m3 to 4.38 mg/m(3), Fe concentrations ranged from 0.53 to 0.63 mg/m3, and Mn concentrations ranged from 0.10 to 0.12 mg/m3. The boiler room average personal breathing zone sample total particulate concentrations ranged from 4.73 mg/m3 to 5.90 mg/m3, Fe concentrations ranged from 0.48 to 0.85 mg/m3, and Mn concentrations ranged from 0.06 to 0.16 mg/m3. Average arc times ranged from 20 to 25% of the total sampling time. Both tracer gas and anemometer techniques were used to estimate ventilation of the boiler room. The steady-state form of the two-zone model was applied to long-term and short-term sample total particulate, Fe, and Mn concentrations obtained during welding in the boiler room and breezeway. The average generation rate in the boiler room was 39.2 mg/min for TP, 6.4 mg/min for Fe, and 1.3 mg/min for Mn. The average generation rate in the breezeway was 40.0 mg/min for TP, 6.6 mg/min for Fe, and 1.2 mg/min for Mn. The field-based generation rates were considerably lower than laboratory-derived published emission rates of between 280 and 650 mg/min for TP. This study emphasizes the need for field-derived welding fume generation rates and showed the personal breathing zone and area sample concentrations can be described by the two-zone model in a way that may help the industrial hygienist estimate exposures. [Supplementary materials are available for this article. Go to the publisher's online edition of the Journal of Occupational and Environmental Hygiene for the following free supplemental resource: Tables detailing the personal breathing zone and average area sample results for breezeway welding and boiler room welding, two-zone modeling results, and boiler room welding personal breathing zone and area sample results with mixing fans on.]