Logo do repositório

Percorrer por autor "Inden, Udo"

A mostrar 1 - 2 de 2
  • A multi-agent system tool for strategic planning in small-lot production environments
    Publication . Barbosa, José; Leitão, Paulo; Inden, Udo; Mascioni, Fosco
    The scale of output respectively sales matters in strategic and operations planning because it controls the degression of fixed costs, of learning curve effects (degression of average or perunit variable costs) or purchasing power that influences prime costs. But what is “small”? There is no clear quantitative definition of “small”, except that the minimum is 1. In the usecases of the ARUM project, airframers and producers of galley appliances, want to sell as much as possible of the same, just limited by the size of the markets for aircrafts and related appliances as well as the interplay of competitive forces [1]. The most sold aircrafts are the Airbus A320 family with an average output of about 180 units per year (6486 in total) [2] and the Boeing 737 family with 230 units per year (8350 in total) [3]. And Iacobuccy HF (IHF) in the long average produced about 1000 galley appliances per year. Compared to car industry that is very small. Compared to the size of the aviation market it is very reasonable. IHF is world marked leader for airworthy coffee makers, On the other side, in the purchasing market also the size of the bill of materials of the product matters. In the cases of the A320 and the B737 that are about 4 million parts per unit, a very big scale, while the galley appliances consist of some hundred parts – a small scale. The differences translate into variations of the leeway of pricing, of market shares and of the profitability. For the purpose of this paper and with regard to the use-case respectively the demonstration scenario it will be sufficient to have a look at the impact of small lots in the sales markets in aviation industry.
    2015Artigo científico Acesso aberto Ver mais
  • Parallelising multi-agent systems for high performance computing
    Publication . Leitão, Paulo; Inden, Udo; Rückemann, Claus-Peter
    Multi-Agent Systems (MAS) are seen as a promising technology to face the current requirements of large-scale distributed and complex systems, e.g., autonomous traffic systems or risk management. The application of MAS to such large scale systems, characterised by millions of distributed nodes, imposes special demanding requirements in terms of fast computation. The paper discusses the parallelisation of MAS solutions using larger-scale distributed High End Computing platforms as well as High Performance Computing as a suitable approach to handle the complexity associated to collaborative solutions for large-scale systems.
    2013Comunicação em conferência Acesso aberto Ver mais