Developing a Learning Hybrid NoSql Algorithm for Vehicle Routing Problem Araf Rotalama Problemi ifin Ogrenen Hibrit NoSql Algoritma Gelistirilmesi


Ozdemir R. H., SERTBAŞ A.

8th International Conference on Computer Science and Engineering, UBMK 2023, Burdur, Turkey, 13 - 15 September 2023, pp.420-425, (Full Text)

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/ubmk59864.2023.10286629
  • City: Burdur
  • Country: Turkey
  • Page Numbers: pp.420-425
  • Keywords: Big Data, Hybrid Learning Algorithm, NOSQL Database Algorithm, Vehicle Routing Problem
  • Istanbul University-Cerrahpasa Affiliated: Yes

Abstract

In this study, a learning Vehicle Routing Algorithm was developed to optimize the delivery of the products in the warehouses located in one or more central locations to the customers in various locations using a certain number of vehicles, and the return of the vehicles to the warehouse/home, and the results were compared with other algorithms used in the literature in terms of accuracy and time performance. A hybrid algorithm that works on a hierarchical database has been developed for the distribution of multiple locations according to the Time Intervals, Types, In-Route Constraints, Optimization Purposes, Road Structure, Fleet Structure and Layout, Inter-Route Constraints. Past plans help to create new plans. New plans are versions of old plans adapted to the new situation. This process is called "learning by planning". 25% increase in accuracy and 97% increase in time performance was observed.