Logo des Repositoriums

Auflistung nach Autor:in "Krcmar,Helmut"

1 - 2 von 2
  • Zeitschriftenartikel
    SiaaS: Simulation as a Service
    (Softwaretechnik-Trends Band 36, Heft 4, 2016) Willnecker, Felix; Vögele, Christian; Krcmar,Helmut
    One major advantage of performance models over tests using real systems is the ability to simulate design alternatives by simply modifying or exchanging parts of such models. However, the evaluation of numerous design alternatives can be time consuming depending on the number of alternatives and the complexity of the model. To overcome this challenge, this work presents a scalable simulation service for the Palladio Component Model (PCM) workbench based on a headless Eclipse instance, a Java EE application server, packaged in a docker container and run in kubernetes. The simulation service supports parallel simulation runs, multiple PCM instances in the same container and scales out automatically, when resources of one container instance exceed. Simulation jobs are triggered by a platform-independent REST interface and can be re-used by other applications. This allows to simulate a vast amount of model instances in parallel on cloud or on-premise installations.
  • Textdokument
    Using Machine Learning to Predict POI Occupancy to Reduce Overcrowding
    (INFORMATIK 2022, 2022) Bollenbach,Jessica; Neubig,Stefan; Hein,Andreas; Keller,Robert; Krcmar,Helmut
    Due to the rapid growth of the tourism industry, associated effects like overcrowding, overtourism, and increasing greenhouse gas emissions lead to unsustainable development. A prerequisite for avoiding those adverse effects is the prediction of occupancy. The present study elaborates on the applicability and performance of various prediction models by taking a case study of beach occupancy data in Scharbeutz, Germany. The case study compares different machine learning models once as supervised machine learning models and once as time series models with a persistence model. XGBoost and Random Forest as time series demonstrate the most accurate prediction, followed by the supervised XGBoost model. However, the short prediction span of time series models is a disadvantage for longer-term visitor management to avoid the explained unsustainable effects through steering measures, so depending on the use case, the XGBoost model is to be favoured.