Open AccessApplication of Fuzzy Time Series (FTS) Algorithm in Production Planning of Indonesia’s Oil Refining Company
Author(s) -
Zakka Ugih Rizqi,
Tommy Aries Kurniawan,
Adinda Khairunisa
Publication year - 2021
Publication title -
indonesian scholar scientific summit taiwan proceeding
Language(s) - English
Resource type - Journals
ISSN - 2797-2437
DOI - 10.52162/3.2021108
Subject(s) - aggregate planning , production (economics) , production planning , heuristics , aggregate (composite) , refining (metallurgy) , operational planning , series (stratigraphy) , computer science , operations research , fuzzy logic , reliability (semiconductor) , industrial engineering , mathematical optimization , engineering , economics , artificial intelligence , mathematics , business , marketing , materials science , chemistry , macroeconomics , composite material , biology , operating system , paleontology , power (physics) , quantum mechanics , physics
Forecasting and aggregate planning are crucial phases in production planning especially for oil refining company that takes expensive production cost. Accurate forecasting greatly influences the success of production planning since it is the starting point of production planning. Whereas aggregate planning becomes important because it functions to bridge between the demand or production target with the existing resource requirements. Seeing the importance of accurate forecasting and aggregate planning, this research emphasizes the use of Fuzzy Time Series (FTS) Algorithm to forecast Premium sales in Indonesia’s oil refining company. The comparison is also done between FTS with the other classical techniques in time series forecasting to test the reliability of algorithm and FTS outperforms the others based on the lowest MAPE value as much as 0.87%. FTS result is then used as an input in the aggregate planning by using heuristics method and comparing 3 strategies which are Level Strategy, Chase Strategy, and Hybrid Strategy. The result shows that Hybrid Strategy is the most efficient one because it produces the lowest production cost for three months production period as much as Rp 3,272,000,000.