Sensitivity Analysis Using Risk Measures

In a quantitative model with uncertain inputs, the uncertainty of the output can be summarized by a risk measure. We propose a sensitivity analysis method based on derivatives of the output risk measure, in the direction of model inputs. This produces a global sensitivity measure, explicitly linking sensitivity and uncertainty analyses. We focus on the case of distortion risk measures, defined as weighted averages of output percentiles, and prove a representation of the sensitivity measure that can be evaluated on a Monte Carlo sample, as a weighted average of gradients over the input space. When the analytical model is unknown or hard to work with, nonparametric techniques are used for gradient estimation. This process is demonstrated through the example of a nonlinear insurance loss model. Furthermore, the proposed framework is extended in order to measure sensitivity to constant model parameters, uncertain statistical parameters, and random factors driving dependence between model inputs.     

Decision model for product rollover in manufacturing operations

The purpose of this study is to investigate critical decisions when planning for product rollover at a manufacturing company, and how to organise these decisions. A literature review and a case study are used to develop a decision model. The findings indicate that product rollover can be organised in a five-phase decision model. The research quality is strengthened by a structured literature review, but it can be argued that more empirical research is needed for validation. For researchers, this paper contributes with the identification of critical decisions and a model for product rollover. For practitioners, the study highlights the need to acknowledge the importance of product rollover and its role as a competitive weapon. In earlier studies, product rollover has mainly been investigated from a market perspective, so that this study contributes by investigating the issue from a manufacturing perspective.     

Contrasting strategies for tree species to cope with heat and dry conditions at urban sites

Street trees planted at tree pits with impermeable pavements are exposed to a high level of both, atmospheric and edaphic stress. For many species tree physiology, optimal growth, and consequently long-term vitality may be hampered under these harsh conditions. Therefore, maintaining the vitality and promoting the ecosystem services provided by urban trees, for example aesthetic values and microclimatic benefits, it is important to establish trees with a high tolerance to stressful urban environments. Measurements of leaf gas exchange, leaf water potential, leaf surface temperature and chlorophyll-fluorescence were used for identification of specific response of Corylus corluna L., Ginkgo biloba L., Liriodendron tulipifera L., and Tilia cordata Mill. ‘Greenspire’. The summer months in 2013 with a period of heat and drought over eight weeks only interrupted by two rainy days were appropriate to access tree’s strategies to cope with low water availability. During the measurement period, the influence of high temperatures, high values of vapor pressure deficit (VPD) and low soil water availability on stomatal conductance, water-use efficiency (WUE) and leaf water potential (Ψmin, Ψmax ) was highest for Liriodendron followed by Ginkgo. These species showed high reductions in WUE in case of higher air temperatures and high values of VPD. Results indicate an avoidance strategy under increasing heat and VPD. In contrast, Corylus and Tilia followed a tolerance strategy indicated by decreasing values of Ψmin and lower reductions of WUE. In conclusion, wide species-specific variation in response to high temperatures and high VPD indicates that substantial potential exists in the selection for dry and hot urban places.     

$$D_s$$-optimality in copula models

Optimum experimental design theory has recently been extended for parameter estimation in copula models. The use of these models allows one to gain in flexibility by considering the model parameter set split into marginal and dependence parameters. However, this separation also leads to the natural issue of estimating only a subset of all model parameters. In this work, we treat this problem with the application of the \(D_s\)-optimality to copula models. First, we provide an extension of the corresponding equivalence theory. Then, we analyze a wide range of flexible copula models to highlight the usefulness of \(D_s\)-optimality in many possible scenarios. Finally, we discuss how the usage of the introduced design criterion also relates to the more general issue of copula selection and optimal design for model discrimination.     

The surprising effectiveness of earliest operation due-date sequencing

In make-to order production, schedule reliability is very important but still not sufficiently accomplished in industrial practice by the vast majority of companies. It has been known for long that processing the orders at a workstation in the order of their operation due-dates can compensate for lateness in the arrival at the workstation within certain boundaries. The paper analyses the effectiveness of earliest operation due date (EODD) sequencing by comparing it to an optimistic theoretical boundary. The surprising result is that EODD can nearly fully exploit the theoretical potential. It should therefore be used in practice whenever schedule reliability is important, with only few exceptions. Its effectiveness though is increasing with the workstation’s WIP level and thus is in conflict with the objective to reduce WIP levels and throughput times. A simple forecasting model allows to assess the extent to which lateness can be compensated by EODD sequencing and which schedule compliance can be achieved.     

Choosing summary statistics by least angle regression for approximate Bayesian computation

Bayesian statistical inference relies on the posterior distribution. Depending on the model, the posterior can be more or less difficult to derive. In recent years, there has been a lot of interest in complex settings where the likelihood is analytically intractable. In such situations, approximate Bayesian computation (ABC) provides an attractive way of carrying out Bayesian inference. For obtaining reliable posterior estimates however, it is important to keep the approximation errors small in ABC. The choice of an appropriate set of summary statistics plays a crucial role in this effort. Here, we report the development of a new algorithm that is based on least angle regression for choosing summary statistics. In two population genetic examples, the performance of the new algorithm is better than a previously proposed approach that uses partial least squares.     

Double ASN Minimax sampling plans by variables when the standard deviation is unknown

We deal with the double sampling plans by variables proposed by Bowker and Goode (Sampling Inspection by Variables, McGraw–Hill, New York, 1952) when the standard deviation is unknown. Using the procedure for the calculation of the OC given by Krumbholz and Rohr (Allg. Stat. Arch. 90:233–251, 2006), we present an optimization algorithm allowing to determine the ASN Minimax plan. This plan, among all double plans satisfying the classical two-point-condition on the OC, has the minimal ASN maximum.