4th African International Conference on Industrial Engineering and Operations Management

Barge Capacity Plan Analysis for Coal Transshipment Process with Decomposition Forecasting and Fuzzy Logic

Yeni Nuraeni & Dendi Ishak
Publisher: IEOM Society International
0 Paper Citations
2 Views
1 Downloads
Track: Operations Management
Abstract

PT XYZ is a company in the supply chain of coal distribution activities from Jetty Loa Tebu to Muara Berau. The coal distribution fleet operated is a barge guided by tugs. PT XYZ must be able to plan the needs of effective coal distribution capacity to avoid uneven distribution, anticipate fluctuations and increase demand for transhipment. This study aims to calculate the projected number of coal transhipment and distribution capacity needs by barges at PT XYZ in the next 5 years and determine the appropriate capacity planning strategy to be implemented in meeting the needs. The research method used is forecasting to project the number of coal transhipment from 10 years ago, the methods being compared are regression and decomposition which appropriate for long-term forecasting. The accuracy of the forecasting model is compared and shows that forecasting using decomposition has better results to apply because it has the smallest RMSE and MAPE values. Coal transhipment projection in 2027 is 1,008,638 MT and we build with Fuzzy Logic Mamdani Optimization model with some rules and parameter. Furthermore, based on forecasting and optimization result we need to add 8 barges to balance the transhipment projection in 2027, a cost analysis comparison is conducted for alternative lead, lag, and average capacity strategies. The calculation results show that the lead capacity strategy has the lowest cost value and is the best capacity planning strategy by investing in or adding sets of tugs and barges.

Published in: 4th African International Conference on Industrial Engineering and Operations Management, Lusaka, Zambia

Publisher: IEOM Society International
Date of Conference: April 4-6, 2023

ISSN/E-ISSN: 2169-8767