Refine
Year of publication
Document Type
- Conference Proceeding (211) (remove)
Is part of the Bibliography
- yes (211)
Keywords
- Nachhaltigkeit (8)
- Landscape Planning (3)
- Nutritional Footprint (3)
- Baumpflege (2)
- Bodenverdichtung (2)
- Climate Resilience (2)
- Dendrologie (2)
- Green Fingers (2)
- Grünflächenmanagement (2)
- Hauswirtschaft (2)
Institute
- Fakultät AuL (211) (remove)
In this paper, we evaluate the application of Bayesian Optimization (BO) to discrete event simulation (DES) models. In a first step, we create a simple model, for which we know the optimal set of parameter values in advance. We implement the model in SimPy, a framework for DES written in Python. We then interpret the simulation model as a black box function subject to optimization. We show that it is possible to find the optimal set of parameter values using the open source library GPyOpt. To enhance our evaluation, we create a second and more complex model. To better handle the complexity of the model, and to add a visual component, we build the second model in Simio, a commercial off-the-shelf simulation modeling tool. To apply BO to a model in Simio, we use the Simio API to write an extension for optimization plug-ins. This extension encapsulates the logic of the BO algorithm, which we deployed as a web service in the cloud.
The fact that simulation models are black box functions with regard to their behavior and the influence of their input parameters makes them an apparent candidate for Bayesian Optimization (BO). Simulation models are multivariable and stochastic, and their behavior is to a large extent unpredictable. In particular, we do not know for sure which input parameters to adjust to maximize (or minimize) the model’s outcome. In addition, the complex models can take a substantial amount of time to run.
Bayesian Optimization is a sequential and self-learning algorithm to optimize black box functions similar to as we find them in simulation models: they contain a set of parameters for which we want to identify the optimal set, they are expensive to evaluate, and they exhibit stochastic noise. BO has proven to efficiently optimize black box functions from varius disciplines. Among those, and most notably, it is successfully applied in machine learning algorithms to optimize hyperparameters.
SimBO is a flexible framework for optimizing discrete event-driven simulations (DES) using sequential optimization algorithms. While specifically designed for Bayesian Optimization (BO) in the context of DES, SimBO can be applied to any black-box problem with other optimization algorithms. The framework consists of four encapsulated components - the black-box problem, the sequential optimization algorithm, a database for experiment configuration and results, and a web-based graphical user interface - that communicate via well-defined interfaces. Each component can be run in different environments, allowing for cooperation between different hardware- and software configurations. In our research context, SimBO’s architecture enabled BO algorithms to be run on a high-performance cluster with GPU support, while the simulation is executed on a local Windows machine using the Simio simulation software. The framework’s flexibility also makes it suitable for evolving from a research-focused tool to a production-ready, cloud-based optimization tool for modern algorithms.
Wirkung eines Vitamin D3-Metaboliten (25-OH-D3) auf den Geburtsverlauf von hochleistenden Sauen
(2022)
Fütterung von Sportpferden
(2021)
Fütterung von Jungpferden
(2021)
Fütterung von Zuchtstuten
(2022)
Fütterung von Jungpferden
(2022)
Fütterung von Jungpferden
(2023)
Fütterung von Zuchtstuten
(2021)
Fütterung von Zuchtstuten
(2023)
Iron deficiency is a global issue and can lead to a variety of clinical pictures. The biofor-tification of vegetables with iron could complement the existing portfolio of iron-rich products, thus improving iron supply in the long term. In order to determine whether the iron-biofortified vegetables could meet this demand and would address appropriate target groups, a quantitative online survey was conducted in Germany. Based on 1000 consumer responses, a cluster analysis was performed. The results showed a four-cluster solution. The first cluster was holistically engaged, the second was fitness-affine but health unconcerned, the third cluster consists frugal eaters with a focus on medical prevention, and the fourth cluster are hedonists. No cluster focused its consumption on iron-enriched products, but instead all developed an individual mix of the three product groups.
Perceptions of German consumers regarding methods for fortifying foods with iodine (Abstract)
(2022)
Besides yield reduction weeds have positive effects, like enhancing biodiversity or reducing soil erosion. The aim of the work is to enhance biodiversity without yield losses. In growing season 2021/22 a field experiment with winter triticale was conducted to experience the impact of mechanical weed control, sowing date and sowing rate. Especially hoeing had a positive impact on vegetative growth and grain yield while weed growth was inhibited but weed biodiversity stayed equal or increased. A lower seed rate or an earlier sowing date made in general no difference at flowering or harvesting stage.
Piglets experience a high level of stress during the onset of the fattening phase due to changes in social structures and feeding. Incorporation of flavors in the feed supports constant levels of feed intake thus avoiding drops in performance. This study aimed to evaluate if a combination of vanilla and fruity flavors could maintain high interest in variable feeds across different ages. 384 crossbred (Topigs 20 x Piétrain) piglets were separated at weaning into two groups (eight pens per group: four pens with females and four pens with uncastrated males; 24 animals per pen), according to gender and weight. The control group received commercial starter feeds without flavor, while the other group received diets with added flavor (500 g/t Cuxarom Fruit 210, a sweetish, berry multifruit note combined with creamy vanilla). After 49 days, 264 animals from both groups were rehoused and given an early-grower feed with or without added flavor for 7 days. Consequently, animals from the trial group again received flavor-supplemented feed and vice versa. During the early-grower phase, the feed intake of the trial group was 11.2% higher than that of the control group. Moreover, the trial group gained significantly more weight during the early-grower phase (+12.2%) in comparison to the control group. This indicates that the animals prefer the flavor they were familiarized with during the starter phase. This memory effect leads to an increased feed intake and weight gain in early-grower pigs and can counteract drops in performance, particularly when mixing animals or transitioning.
Piglets experience a high level of stress upon weaning due to changes in social structures and feeding. Incorporation of flavors in the maternal diet might be a promising tool to prenatally modulate flavor preferences of the offspring and prevent neophobia. This study aimed to evaluate if the supplementation of the maternal diet with Spicemaster GH 501, an intense, plant-based herbal flavor, could affect feed preferences of early-weaned piglets and thus improve feed intake, performance, and overall health conditions. On approx. day 95 of gestation, 59 pregnant sows (DanZucht) were separated into two groups. The control group received commercial gestation and lactation feeds without flavor, while the trial group received diets with added flavor (400 g/t Spicemaster GH 501). After weaning, 272 piglets from both groups were rehoused and given an early-grower feed, which was consistently supplemented in both groups with 400 g/t Spicemaster GH 501 throughout the whole trial. Strikingly, the trial group, which was prenatally exposed to the flavor, gained significantly more weight during the initial phase after weaning. Moreover, the trial group was less prone to develop ear tip necrosis and displayed an improved general health condition, which was reflected in a reduced number of antibiotic treatments during the duration of the trial. Collectively, the incorporation of flavoring compounds in the maternal diet provides a powerful mechanism to shape taste preferences in early-weaned piglets and conse-quently reduce stress levels and susceptibility to infections.
Aufgrund seiner starken Umweltauswirkungen gilt der Nahrungsmittelsektor durch Effekte in Produktion, Verarbeitung, Konsum und Entsorgung gemeinhin als ein wichtiges Handlungsfeld, soll eine gesellschaftliche Transformation in Richtung Nachhaltigkeit weiter vorangetrieben werden. Insofern Ernährungsgewohnheiten sowohl ökologische Auswirkungen induzieren als auch gesundheitliche Folgen für die Verbraucher haben, sind Konzepte gefragt, die ökologische mit gesundheitlichen Indikatoren kombinieren; diese sind jedoch bislang rar. Das vorgestellte Instrument des Nutritional Footprint greift diese Problemstellung auf und verbindet jeweils vier Kernindikatoren beider Dimensionen. Mit Hilfe des Konzeptes erhalten Verbraucher einen Überblick zu Umwelt- und Gesundheitswirkungen ihrer Ernährung. Unternehmen können wiederum interne Datensätze verwalten, Benchmarking betreiben und ihre externe Kommunikationsleistung erweitern.