Quantile-Based Optimization of Noisy Computer Experiments with Tunable Precision

This article addresses the issue of kriging-based optimization of stochastic simulators. Many of these simulators depend on factors that tune the level of precision of the response, the gain in accuracy being at a price of computational time. The contribution of this work is two-fold: first, we propose a quantile-based criterion for the sequential design of experiments, in the fashion of the classical expected improvement criterion, which allows an elegant treatment of heterogeneous response precisions. Second, we present a procedure for the allocation of the computational time given to each measurement, allowing a better distribution of the computational effort and increased efficiency. Finally, the optimization method is applied to an original application in nuclear criticality safety. This article has supplementary material available online. The proposed criterion is available in the R package DiceOptim.

IMILAST: A community effort to intercompare extratropical cyclone detection and tracking algorithms: assessing method-related uncertainties

The variability of results from different automated methods of detection and tracking of extratropical cyclones is assessed in order to identify uncertainties related to the choice of method. Fifteen international teams applied their own algorithms to the same dataset - the period 1989-2009 of interim European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERAInterim) data. This experiment is part of the community project Intercomparison of Mid Latitude Storm Diagnostics (IMILAST; see www.proclim.ch/imilast/index.html). The spread of results for cyclone frequency, intensity, life cycle, and track location is presented to illustrate the impact of using different methods. Globally, methods agree well for geographical distribution in large oceanic regions, interannual variability of cyclone numbers, geographical patterns of strong trends, and distribution shape for many life cycle characteristics. In contrast, the largest disparities exist for the total numbers of cyclones, the detection of weak cyclones, and distribution in some densely populated regions. Consistency between methods is better for strong cyclones than for shallow ones. Two case studies of relatively large, intense cyclones reveal that the identification of the most intense part of the life cycle of these events is robust between methods, but considerable differences exist during the development and the dissolution phases.

Counting scars on tree stems to assess rockfall hazards: A low effort approach, but how reliable?

Rockfall is a widespread and hazardous process in mountain environments, but data on past events are only rarely available. Growth-ring series from trees impacted by rockfall were successfully used in the past to overcome the lack of archival records. Dendrogeomorphic techniques have been demonstrated to allow very accurate dating and reconstruction of spatial and temporal rockfall activity, but the approach has been cited to be labor intensive and time consuming. In this study, we present a simplified method to quantify rockfall processes on forested slopes requiring less time and efforts. The approach is based on a counting of visible scars on the stem surface of Common beech (Fagus sylvatica L.). Data are presented from a site in the Inn valley (Austria), where rocks are frequently detached from an ~ 200-m-high, south-facing limestone cliff. We compare results obtained from (i) the “classical” analysis of growth disturbances in the tree-ring series of 33 Norway spruces (Picea abies (L.) Karst.) and (ii) data obtained with a scar count on the stem surface of 50 F. sylvatica trees. A total of 277 rockfall events since A.D. 1819 could be reconstructed from tree-ring records of P. abies, whereas 1140 scars were observed on the stem surface of F. sylvatica. Absolute numbers of rockfalls (and hence return intervals) vary significantly between the approaches, and the mean number of rockfalls observed on the stem surface of F. sylvatica exceeds that of P. abies by a factor of 2.7. On the other hand, both methods yield comparable data on the spatial distribution of relative rockfall activity. Differences may be explained by a great portion of masked scars in P. abies and the conservation of signs of impacts on the stem of F. sylvatica. Besides, data indicate that several scars on the bark of F. sylvatica may stem from the same impact and thus lead to an overestimation of rockfall activity.

Resource allocation with project management software

We present results of a benchmark test evaluating the resource allocation capabilities of the project management software packages Acos Plus.1 8.2, CA SuperProject 5.0a, CS Project Professional 3.0, MS Project 2000, and Scitor Project Scheduler 8.0.1. The tests are based on 1560 instances of precedence– and resource–constrained project scheduling problems. For different complexity scenarios, we analyze the deviation of the makespan obtained by the software packages from the best feasible makespan known. Among the tested software packages, Acos Plus.1 and Project Scheduler show the best resource allocation performance. Moreover, our numerical analysis reveals a considerable performance gap between the implemented methods and state–of–the–art project scheduling algorithms, especially for large–sized problems. Thus, there is still a significant potential for improving solutions to resource allocation problems in practice.

Patient classification system: An integrated method for measuring nursing intensity and optimizing resource allocation

A patient classification system was developed integrating a patient acuity instrument with a computerized nursing distribution method based on a linear programming model. The system was designed for real-time measurement of patient acuity (workload) and allocation of nursing personnel to optimize the utilization of resources.^ The acuity instrument was a prototype tool with eight categories of patients defined by patient severity and nursing intensity parameters. From this tool, the demand for nursing care was defined in patient points with one point equal to one hour of RN time. Validity and reliability of the instrument was determined as follows: (1) Content validity by a panel of expert nurses; (2) predictive validity through a paired t-test analysis of preshift and postshift categorization of patients; (3) initial reliability by a one month pilot of the instrument in a practice setting; and (4) interrater reliability by the Kappa statistic.^ The nursing distribution system was a linear programming model using a branch and bound technique for obtaining integer solutions. The objective function was to minimize the total number of nursing personnel used by optimally assigning the staff to meet the acuity needs of the units. A penalty weight was used as a coefficient of the objective function variables to define priorities for allocation of staff.^ The demand constraints were requirements to meet the total acuity points needed for each unit and to have a minimum number of RNs on each unit. Supply constraints were: (1) total availability of each type of staff and the value of that staff member (value was determined relative to that type of staff's ability to perform the job function of an RN (i.e., value for eight hours RN = 8 points, LVN = 6 points); (2) number of personnel available for floating between units.^ The capability of the model to assign staff quantitatively and qualitatively equal to the manual method was established by a thirty day comparison. Sensitivity testing demonstrated appropriate adjustment of the optimal solution to changes in penalty coefficients in the objective function and to acuity totals in the demand constraints.^ Further investigation of the model documented: correct adjustment of assignments in response to staff value changes; and cost minimization by an addition of a dollar coefficient to the objective function. ^

Effects of externally supplied protein on root morphology and biomass allocation in Arabidopsis

Growth, morphogenesis and function of roots are influenced by the concentration and form of nutrients present in soils, including low molecular mass inorganicN(IN, ammonium, nitrate) and organicN(ON, e. g. amino acids). Proteins, ON of high molecular mass, are prevalent in soils but their possible effects on roots have received little attention. Here, we investigated how externally supplied protein of a size typical of soluble soil proteins influences root development of axenically grown Arabidopsis. Addition of low to intermediate concentrations of protein (bovine serum albumen, BSA) to IN-replete growth medium increased root dry weight, root length and thickness, and root hair length. Supply of higher BSA concentrations inhibited root development. These effects were independent of total N concentrations in the growth medium. The possible involvement of phytohormones was investigated using Arabidopsis with defective auxin (tir1-1 and axr2-1) and ethylene (ein2-1) responses. That no phenotype was observed suggests a signalling pathway is operating independent of auxin and ethylene responses. This study expands the knowledge on N form-explicit responses to demonstrate that ON of high molecular mass elicits specific responses.

Distributed Rate Allocation in Inter-Session Network Coding

In this work, we propose a distributed rate allocation algorithm that minimizes the average decoding delay for multimedia clients in inter-session network coding systems. We consider a scenario where the users are organized in a mesh network and each user requests the content of one of the available sources. We propose a novel distributed algorithm where network users determine the coding operations and the packet rates to be requested from the parent nodes, such that the decoding delay is minimized for all clients. A rate allocation problem is solved by every user, which seeks the rates that minimize the average decoding delay for its children and for itself. Since this optimization problem is a priori non-convex, we introduce the concept of equivalent packet flows, which permits to estimate the expected number of packets that every user needs to collect for decoding. We then decompose our original rate allocation problem into a set of convex subproblems, which are eventually combined to obtain an effective approximate solution to the delay minimization problem. The results demonstrate that the proposed scheme eliminates the bottlenecks and reduces the decoding delay experienced by users with limited bandwidth resources. We validate the performance of our distributed rate allocation algorithm in different video streaming scenarios using the NS-3 network simulator. We show that our system is able to take benefit of inter-session network coding for simultaneous delivery of video sessions in networks with path diversity.

Saving the Australian War Effort in 1916? The Significance and Impact of Global Climatic Conditions and Pests on William Morris Hughes’ Negotiations with the British Government

Taking up the thesis of Dipesh Chakrabarty (2009) that human history (including cultural history) on the one hand and natural history on the other must be brought into conversation more than has been done so in the past, this presentation will focus more closely on the significance and the impact of global climatic conditions and pests on the negotiations that Australian Prime Minister William Morris Hughes carried on with the British government between March and November 1916. Whereas Australia had been able to sell most of its produce in 1914 and 1915 the situation looked more serious in 1916, not least due to the growing shortage in shipping. It was therefore imperative for the Australian government to find a way to solve this problem, not least because it wanted to keep up its own war effort at the pace it had been going so far. In this context intentions to make or press ahead with a contribution to a war perceived to be more total those of the past interacted with natural phenomena such as the declining harvest in many parts of the world in 1916 as a consequence of climatic conditions as well as pests in many parts of the world.

Evaluating gillnetting protocols to characterize lacustrine fish communities

Ecological research and monitoring of lacustrine ecosystems often requires a whole-lake assessment of fish communities. Gillnet sampling offers an efficient means of estimating abundance, biomass and fish community composition. However the choice of gillnet sampling protocol may influence lake characterization via physical properties of the nets and allocation of sampling effort between littoral, benthic and pelagic habitats. This paper compares two commonly used, whole-lake sampling protocols applied across 17 prealpine, subalpine and alpine European lakes ranging widely in size, depth and altitude to determine their relative strength for research and management applications. Effort-corrected estimates of abundance, biomass and species richness were correlated between the protocols and both distinguished the trout-dominated alpine communities from subalpine and prealpine lakes dominated by whitefish and perch. A considerable amount of variance remained unexplained between the two protocols however, which seemed to correspond with differences in the proportion of effort among benthic and pelagic habitats. We suggest that both the European standard (CEN) and vertical (VERT) netting protocols are suitable for assessing ecological status and monitoring changes in lake fish communities through time. However the details of each protocol should be kept in mind when comparing fish communities between lakes. Mesh sizes used in CEN nets produce a more even size frequency distribution, suggesting that this protocol is most appropriate for assessing size structure of fish assemblages. The high proportion of netting effort in benthic habitats shallower than 70 m depth under the CEN protocol means that, particularly in larger lakes, outcomes will be disproportionately influenced by the ecological condition of this habitat. The VERT protocol presumably provides a more accurate estimate of whole-lake CPUE and community composition because effort, in terms of net area, is more evenly distributed across the entire volume of the lake. This is particularly important in large and deep lakes where pelagic habitats occupy a high proportion of the lake volume.

Carbon-11 Reveals Opposing Roles of Auxin and Salicylic Acid in Regulating Leaf Physiology, Leaf Metabolism, and Resource Allocation Patterns that Impact Root Growth in Zea mays

Auxin (IAA) is an important regulator of plant development and root differentiation. Although recent studies indicate that salicylic acid (SA) may also be important in this context by interfering with IAA signaling, comparatively little is known about its impact on the plant’s physiology, metabolism, and growth characteristics. Using carbon-11, a short-lived radioisotope (t 1/2 = 20.4 min) administered as 11CO2 to maize plants (B73), we measured changes in these functions using SA and IAA treatments. IAA application decreased total root biomass, though it increased lateral root growth at the expense of primary root elongation. IAA-mediated inhibition of root growth was correlated with decreased 11CO2 fixation, photosystem II (PSII) efficiency, and total leaf carbon export of 11C-photoassimilates and their allocation belowground. Furthermore, IAA application increased leaf starch content. On the other hand, SA application increased total root biomass, 11CO2 fixation, PSII efficiency, and leaf carbon export of 11C-photoassimilates, but it decreased leaf starch content. IAA and SA induction patterns were also examined after root-herbivore attack by Diabrotica virgifera to place possible hormone crosstalk into a realistic environmental context. We found that 4 days after infestation, IAA was induced in the midzone and root tip, whereas SA was induced only in the upper proximal zone of damaged roots. We conclude that antagonistic crosstalk exists between IAA and SA which can affect the development of maize plants, particularly through alteration of the root system’s architecture, and we propose that the integration of both signals may shape the plant’s response to environmental stress.

What is the influence of randomisation sequence generation and allocation concealment on treatment effects of physical therapy trials? A meta-epidemiological study.

OBJECTIVE To determine if adequacy of randomisation and allocation concealment is associated with changes in effect sizes (ES) when comparing physical therapy (PT) trials with and without these methodological characteristics. DESIGN Meta-epidemiological study. PARTICIPANTS A random sample of randomised controlled trials (RCTs) included in meta-analyses in the PT discipline were identified. INTERVENTION Data extraction including assessments of random sequence generation and allocation concealment was conducted independently by two reviewers. To determine the association between sequence generation, and allocation concealment and ES, a two-level analysis was conducted using a meta-meta-analytic approach. PRIMARY AND SECONDARY OUTCOME MEASURES association between random sequence generation and allocation concealment and ES in PT trials. RESULTS 393 trials included in 43 meta-analyses, analysing 44 622 patients contributed to this study. Adequate random sequence generation and appropriate allocation concealment were accomplished in only 39.7% and 11.5% of PT trials, respectively. Although trials with inappropriate allocation concealment tended to have an overestimate treatment effect when compared with trials with adequate concealment of allocation, the difference was non-statistically significant (ES=0.12; 95% CI -0.06 to 0.30). When pooling our results with those of Nuesch et al, we obtained a pooled statistically significant value (ES=0.14; 95% CI 0.02 to 0.26). There was no difference in ES in trials with appropriate or inappropriate random sequence generation (ES=0.02; 95% CI -0.12 to 0.15). CONCLUSIONS Our results suggest that when evaluating risk of bias of primary RCTs in PT area, systematic reviewers and clinicians implementing research into practice should pay attention to these biases since they could exaggerate treatment effects. Systematic reviewers should perform sensitivity analysis including trials with low risk of bias in these domains as primary analysis and/or in combination with less restrictive analyses. Authors and editors should make sure that allocation concealment and random sequence generation are properly reported in trial reports.

Estimating whole-lake fish catch per unit effort

The European standard for gillnetsampling to characterize lake fish communities stratifies sampling effort (i.e., number of nets) within depth strata. Nets to sample benthic habitats are randomly distributed throughout the lake within each depth strata. Pelagic nets are also stratified by depth, but are set only at the deepest point of the lake. Multiple authors have suggested that this design under-represents pelagic habitats, resulting in estimates of whole-lake CPUE and community composition which are disproportionately influenced by ecological conditions of littoral and benthic habitats. To address this issue, researchers have proposed estimating whole-lake CPUE by weighting the catch rate in each depth-compartment by the proportion of the volume of the lake contributed by the compartment. Our study aimed to assess the effectiveness of volume-weighting by applying it to fish communities sampled according to the European standard (CEN), and by a second whole-lake gillnetting protocol (VERT), which prescribes additional fishing effort in pelagic habitats. We assume that convergence between the protocols indicates that volume-weighting provides a more accurate estimate of whole-lake catch rate and community composition. Our results indicate that volume-weighting improves agreement between the protocols for whole-lake total CPUE, estimated proportion of perch and roach and the overall fish community composition. Discrepancies between the protocols remaining after volume-weighting maybe because sampling under the CEN protocol overlooks horizontal variation in pelagic fish communities. Analyses based on multiple pelagic-set VERT nets identified gradients in the density and biomass of pelagic fish communities in almost half the lakes that corresponded with the depth of water at net-setting location and distance along the length of a lake. Additional CEN pelagic sampling effort allocated across water depths and distributed throughout the lake would therefore help to reconcile differences between the sampling protocols and, in combination with volume-weighting, converge on a more accurate estimate of whole-lake fish communities.