PERAMALAN PERMINTAAN SUKU CADANG OTOMOTIF KARET DENGAN INTEGRASI AGENT BASED MODELLING DAN DOUBLE EXPONENTIAL SMOOTHING

  • Dewi Auditiya Marizka Politeknik STMI Jakarta-Ministry of Industry, Indonesia
  • Yandra Arkeman Department of Agroindustrial Technology, IPB university
  • Sapta Raharja Department of Agroindustrial Technology, IPB university
  • Hartrisari Hardjomidjojo Department of Agroindustrial Technology, IPB university
  • Ono Suparno Department of Agroindustrial Technology, IPB university

Abstract

The objective of this study was to design an agent model for forecasting demand for rubber-based automotive parts using the Agent Based Modelling and Double Exponential Smoothing (ABMDES) approach.  Model design in rubber-based automotive spare parts forecasting using the integration of Agent Based Modelling (ABM) approach and Double Exponential Smoothing (DES) technique was done using agent design approach based on class diagram and definition function for each agent in mathematics models including DES-based forecasting. The ABM design has a structure consisting of an agent ID, an attribute to be calculated by the computer and function / process consisting of a function that has values ​​and voids (unstructured). Combining ABM and DES can guide us to see the forecast accuracy, monitoring the estimation of stock shortage and the excess of stock due to errors in DES forecast. Therefore, the Agent Based Modelling-Double Exponential Smoothing (ABMDES) approach is suitable for modelling the demand of rubber-based automotive parts in business simulation.

Keywords: double exponential smoothing, function, SMEs, shortage

Published
2020-12-29