PEMODELAN RANTAI PASOK FLEKSIBEL PADA PEMURNIAN MINYAK SAWIT: STUDI KASUS DI PT X

The development of the CPO-producing industry provides great benefits for the CPO refining industry.CPO is used as the main raw material for refined derivative products such as oleopangan products (cooking oil,margarine, shortening) and oleochemicals (fatty acids, fatty alcohol, glycerine). A related problem is theinefficiency in the distribution network caused by the complexity of the existing system and related to the manytypes or product variants that must be integrated in shipping. The aim of this research was to design a flexiblesupply chain network for domestic palm oil refined products using a soft computing approach with a geneticalgorithm. The simulation results using GA showed that the optimal solution completion time of themathematical model only takes 57 seconds. The optimization of the model that has been formulated was tested incases where the results were measured by the variables of the number of products distributed, distribution flowin the supply chain network, minimum cost (Z), and network facilities. The design was also tested by verificationof the optimization model by correcting errors in the model until the model produces a feasible solution andvalidating the optimization model to experts through the Association of Indonesian Palm Oil Entrepreneurs(GAPKI) North Sumatra, the Medan Palm Oil Research Center (PPKS) and Public Appraisal ServiceConsultants that often handles problems in the oil palm sector. According to experts, the designed modelrepresented the real situation so that it can be said that the model has been validated. The design is flexiblebecause it is able to meet the existing demand with utilities above 100% for 9 periods at a cost of Rp630,892,386.Keywords: multi-product, CPO refinery supply chain, genetic algorithm, optimization, flexible supply chainnetwork desig