Abstract:

Vehicle Rotation Problems (ARPs) are often used to find the best routes to meet the company’s distribution and collection requirements. In real life, the meeting of new collection requests coming within the day by the vehicles in circulation will reduce transportation costs. In this context, a new mathematical model for the Dynamic Co-Time Collective-Distributed Vehicle Rotation Problem (DETD_ARP) has been developed in order to meet new collection requests from customers by vehicles in circulation.  Test problems in literature have been used to investigate the effectiveness of the developed mathematical model and the results have been evaluated. The results obtained by the mathematical model have been evaluated and as the problem size increases, the solution time has increased overly. In this study, intuitive algorithms have been used to solve the problem in a short time. A new algorithm called the variable neighbor landing (R - GO - DKI) has been developed.  R - GOOD - The application order of neighbouring structures by using the upper (hyper) intuitions in the DKI has been constantly changed according to the reduced permutation method. To evaluate the effectiveness of the developed algorithm, the results obtained by the mathematical model were used in the study. In view of the results, the intuitive algorithm developed has been shown to give results close to the results obtained by the mathematical model. Finally, big-dimensional problems were solved with the recommended R - GO - DKI algorithm and the results were shared.

Keywords: