CLUSTERING APPROACH ON TWO-ECHELON LOCATION ROUTING PROBLEM WITH DIRECT SHIPMENT AND MAXIMUM TRAVEL TIME USING SIMULATED ANNEALING ALGORITHM

• Main Author: AMIR HAMZAH - NIM: 23414034 , DWINANTO
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/21836
• Institution: Institut Teknologi Bandung

Direct shipment generate the lowest cost in LNG distribution, yet not all destination point can be reach when the travel time to reach destination point is greater than the planning horizon. To be able deliver with timely manner required intermediary facility to connect LNG source location with destination location and devide the supply chain network into two echelon supply chain network. This problem is studied specifically in a research field called 2-echelon location routing problem (2E-LRP). <br /> <br /> <br /> In this research, developed 2E-LRP mathematical model by considering direct shipment and maximum travel time. Mathematical model proposed in this research is complex that analytic calculations can only be used on small data sets. In order to solve real problems with large data sets, simulated annealing algorithm was developed using clustering approach. The initial solution was generated by clustering based heuristic algorithm then improved by using simulated annealing algorithm with three operators, split cluster, combine cluster and origin swap or exchange. <br /> <br /> <br /> Based on the experimental results with 12 sets of hypothetical data generate an average solution of 99.89% approaching global optimum solution with computation time used only 13.34% of the total time average to obtain global optimum solution by using analytical method.