Task scheduling and resource allocation with multi-variable heuristics

Tamás Rick, István Groma, Ádám Gránicz, Tibor Bercsey

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Nowadays product development processes are typically specified as bound work processes (as a workflow), and their treatment is performed by PLM systems. One of the benefits of this approach, as implemented by various PLM applications, is that it integrates PDM/EDM, and in some cases PPS, SCM tasks as well. On the other hand, the workflow-based process description is only applicable to modeling simple recurring development activities that can only be treated as a whole, and it is impossible to specify and treat iterations that often accompany the development process in a dynamic way. Furthermore, the assignment of resources to the tasks of the given process is not automated, and it is not possible to adapt the process in case of a change in the requirements due to the rigid, predefined structure. Therefore, the use of work processes is only effective if a pre-defined standard, repetitive processes was used. However development processes are unique and valid in a given context that changes with time. Therefore it is advised to use a DSM, through which one can get a more precise picture of the product and the relationship of its structural elements during the development planning process. This approach supports the modeling of iterations that are quintessential in practice. With the help of genetic algorithm-based optimization one can find the one development schedule with the least development time and cost by changing the schedule of the product elements. As a result a project plan can be obtained - and for each element the needed resources can be assigned. But in most cases, the enterprise resource profile does not fully cover the requirements; therefore it becomes necessary to create a modified project plan keeping the optimized schedule, one that is suitable for the enterprise resource profile. For solving the problem of resource assignment a heuristic method has been created where the project plan is fitted to the enterprise resource setting by applying various strategies, focusing on minimizing the time requirements of the development project.

Original languageEnglish
Title of host publicationProceedings of ICED 2007, the 16th International Conference on Engineering Design
Publication statusPublished - Dec 1 2007
Event16th International Conference on Engineering Design, ICED 2007 - Paris, France
Duration: Jul 28 2007Jul 31 2007

Publication series

NameProceedings of ICED 2007, the 16th International Conference on Engineering Design
VolumeDS 42

Other

Other16th International Conference on Engineering Design, ICED 2007
CountryFrance
CityParis
Period7/28/077/31/07

    Fingerprint

Keywords

  • Dsm
  • Genetic algorithms
  • Product development
  • Resource allocation
  • Task scheduling

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

Rick, T., Groma, I., Gránicz, Á., & Bercsey, T. (2007). Task scheduling and resource allocation with multi-variable heuristics. In Proceedings of ICED 2007, the 16th International Conference on Engineering Design (Proceedings of ICED 2007, the 16th International Conference on Engineering Design; Vol. DS 42).