A neural network for resource constrained project scheduling programming


Creative Commons License

Ozkan O. , Gulcicek U.

JOURNAL OF CIVIL ENGINEERING AND MANAGEMENT, cilt.21, ss.193-200, 2015 (SCI İndekslerine Giren Dergi)

  • Cilt numarası: 21 Konu: 2
  • Basım Tarihi: 2015
  • Doi Numarası: 10.3846/13923730.2013.802723
  • Dergi Adı: JOURNAL OF CIVIL ENGINEERING AND MANAGEMENT
  • Sayfa Sayısı: ss.193-200

Özet

The resource constrained project-scheduling problem (RCPSP) aims to minimize the duration of a project. RCPSP is prevalently used in programming the projects with high number of activities and resources such as construction projects. In this study, 240 projects such as residential, office, school, etc. are designed and programmed under limited resources. The resource amounts of these projects are determined using three priority rules, these are Latest Finish Time, Minimum Slack Time and Maximum Remaining Path Length which have the highest performance according to the literature, in the amounts of 2, 4, 6 and 8. The project times are estimated using artificial neural network (ANN). A correlation coefficient of 0.70 was obtained from the ANN estimation model.